Return to flip book view

ResearchResearchPortfolioPortfolio20242024Transformative technologies for personalized, vigilant health monitoring.Transformative technologies for personalized, vigilant health monitoring.

Wearable EnergyHarvesting and StorageLow PowerSensor SystemsLow Power Circuitsand Systems on ChipSmart E-textiles andFlexible Materials Emerging CorrelatedSensing Applications

Electrospun Molecular Ferroelectric-Based Composite Nanofibers for Mechanical Energy HarvestingEnergy Harvesting and Self-Powered Smart Insoles for Gait and Balance DetectionFlexible Battery Yarns for Smart TextilesFlexible Electrochemical Capacitor for Wearable TechnologiesPrinting Metallic Pastes and Metallic ‘Gels’ for Wearable Energy HarvestingSpace Charge-Induced Flexoelectric Transducers to Replace Lead-Based Piezoelectric TransducersWearable Neuro-Biochemistry Sensors Powered by Enzymatic Fuel Cells45678910A Wireless Sensing Ecosystem to Support Biomedical InterfacesLabel and Bio-Active Free Electrochemical Detection of Testosterone Hormone Using MIP-BasedSensing PlatformMicroneedle-Integrated Sensing System for Minimally-Invasive Extraction & Analysis of Dermal Interstitial FluidMiniaturized, Wearable, and Flexible Wound Monitoring SystemUltra-Low-Power Metal Oxide Electronic Nose Array for Breath, Skin, and Environmental MonitoringUltra-Thin Polymer-Based Flexible Platforms for Wound and Skin Temperature SensingWearable Ultrasound Systems for Muscle Activity Sensing in Assistive Robotics12131415161718Bio-electro-photonic Microsystem Interfaces for Small Animal Health MonitoringBody Area Network of Inertial Sensors for Clinical Gait RehabilitationCompressive Sensing-Based Ultra low power PhotoPlethysmoGraphy (PPG) Module for Wearable SystemsMiniaturized Extended Gate Field Effect Transistor Readout System for Wearable Analyte SensingUltra-Low Power Integrated Systems-on-Chip (SoCs) for Wearable Devices and Internet of Things SolutionsUltrasonic Energy/Data Transfer for Implantable SystemsWirelessly-Powered Bluetooth Backscatter-Based Electrocardiography System20212223242526Advanced Flexible and Textile Antennas for On-Body Wearable ApplicationsComfortable Outfits from Utilitarian Textiles for Unobtrusive Recording of Events (COUTURE)Eco-friendly Screen Printing of Silver Nanowires for Flexible and Stretchable ElectronicsFlexible Organic Poly(Octamethylene Maleate (anhydride) Citrate) (POMaC) CircuitsIntegrated Tactile Interfaces and Physiological Sensing Arrays Using Sustainable 3D-Printed Hydrogels forWearable SystemsTextile Electromyography Sensors for Wireless Stroke Therapeutic SystemTextile Integrated Sensors for Inner Prosthetic Socket Environment Monitoring for Neurorehabilitation28293031323334A Wearable System for Continuous Monitoring and Assessment of Alzheimer's Disease and Related Dementia Artificial Intelligence Driven, Resilient and Adaptive Monitoring of Sleep (AI-DReAMS)Embedded Wearable System-Based Cough Detection Enhanced by Out-of-Distribution RecognitionLanguage Processing Using Cloud and Embedded Artificial Intelligence for Monitoring of Cognitive DeclineMultifunctional Plant Wearable Sensors for Agricultural Monitoring and One HealthMultimodal Flexible Electrochemical and Biophotonic Biosensing System for Metabolic MonitoringSensor-Integrated Microfluidic Chip to Monitor Nutrient Uptake in Plants for Smart AgricultureWearable Sensors-Based Health and Exposure Tracker for Asthma and Diabetes3637383940414243

AgingAsthmaNeurorehabilitationTechnologyApplicationsDiabetesSleepMonitoringMetabolicStatusCough/Speech DetectionCardiovascularAlzheimer’s/DementiaWound Monitoring/HealingStress Monitoring/Mental HealthGait/Fall DetectionForewordWelcome to the ASSIST 2024 Research Portfolio! Established in 2012, the NSF NanosystemsEngineering Research Center for Advanced Self-Powered Systems of Integrated Sensors andTechnologies (ASSIST) has envisioned always-on wearable devices tailored for continuousmonitoring in chronic disease management and gaining data-driven insights into humanhealth. Interdisciplinary teams from NC State University, the University of Virginia, Penn StateUniversity, Florida International University, the University of Michigan, and the University of Utahhave collaborated to build disruptive technologies in energy harvesting, low-power systems-on-chip, and integration on flexible platforms such as textiles. Through collaborations with cliniciansand industry partners, we've validated our wearable systems and translated our technologies toaddress critical health concerns like asthma, cardiac disease, diabetes, and wound monitoring.Since our graduation from NSF last year, we have been broadening the Center's mission byexpanding to new technical areas such as implantable devices, novel ultrasound-basedsensing, and innovative applications of textile fibers. Our health applications are alsoexpanding and include areas such as Alzheimer’s disease and related dementias, stressmonitoring, and animal-assisted interventions. Expressing our heartfelt gratitude to the NationalScience Foundation for their 10 years of invaluable support, we are now in a phase to translateour technologies into the field and to expand our industry program for a self-sufficientinnovation ecosystem. This portfolio summarizes our repertoire of innovations and researchendeavors to create revolutionary technologies for disruptive healthcare through continuousmonitoring. We invite you to reach out and explore collaborative opportunities together.ASSIST Co-DirectorsVeena Misra and Alper Bozkurt

Wearable EnergyWearable EnergyHarvesting & StorageHarvesting & Storage3

Electrospun Molecular Ferroelectric-Based CompositeNanofibers for Mechanical Energy HarvestingObjective:Developing flexible devices to scavenge ambient energy iscrucial to fulfilling the requirements of swiftly escalatingglobal energy consumption. Triboelectric nanogenerators,known for their ability to effectively transform kinetic energyinto electrical power, are viewed as a promising solution foradvancing self-powered wearables and electronics in thenext generation. In the extensive range of materials underexploration, ferroelectric materials emerge as the idealclass to achieve these objectives. However, up until now, ithas been limited to ceramic-based substitutes whichinvolve energy-intensive fabrication procedures, comprisetoxic element composition and are highly brittle hamperingtheir widespread utilization for on-body electronics. Thiswork aims to leverage the benefits of a newly emergingclass of molecular ferroelectrics which offer superiorfeatures of biocompatibility, flexibility and lightweightproperties all of which are indispensable for thedevelopment of self-powered wearable electronics.Approach:The approach involves incorporating an organicferroelectric into a polymer matrix through theelectrospinning technique. This modification not onlyfacilitates an increased surface area, enhancing chargegeneration but also allows for aesthetics that arechallenging to achieve in thin film counterparts.Additionally, the electrospinning tool offers noteworthyadvantages for an in-situ poling process, eliminating therequirement for an extra step for poling. Employing a fibrouselectrospun mat, we constructed an all-organic activelayer-based triboelectric nanogenerator capable ofefficiently harnessing energy from human mechanicalmovements with superior output performance.Key Accomplishments:The synergistic effects stemming from the alignment ofpolymeric chains and the presence of ferroelectric particlescoupled with a high surface area of the as-designed TENGgenerated the output voltage of 204 V and yielded themaximum power density of 416 mW/m2 when operated incontact separation mode. Moreover, when used to harvestenergy from human body motions, it produced anoverarching output voltage of 200 V, when activity likejumping was involved.Impact:This research introduces a reliable, economically viable,and readily scalable approach for boosting theperformance of the triboelectric nanogenerator. The outputpower of the device can be tailored by fine-tuning thediameter and employing an active layer with moretriboelectric features. Considering the structural tunabilitythat molecular ferroelectric possesses and viablemodulation approaches that can be employed, this studyunfolds a new research direction for realizing a sustainableenergy solution for wearable electronics.Principal Investigators:Veena Misra, PhD, Bongmook Lee, PhD, Wei Gao, PhDPostdoc/Students:Swati Deswal, PhD, Shima Arab, Nanfei HeFunding Source:ASSIST Center(a) FESEM image of as preparedelectrospun fiber mat.(b) Output voltage(c) power characteristics of thefabricated device at an appliedforce of 50 N and at a frequency of1.3 Hz.(d) Biomechanical energyharvesting as part of jumpingprocess.4

Key Accomplishments:Foot pressure sensors were prototyped on polished metalfoils passivated with 100m LaNiO3 and a 50 nm HfO2 layer.A piezoelectric layer with a 52/48 Zr/Ti ratio with 2% Nbdoping, 12% excess Pb, and a molarity of 0.6M. Theprocessed foil was withdrawn from the solution bath at arate of 30 mm/min, pyrolyzed, and crystallized. The pressure-sensing foil was characterized in tests in which a cylinderwas rolled over the array to mimic the action of the ball ofthe foot. We also fabricated a squishy prototype heel thatwas mounted to a material test machine (see Fig. 2) tomimic the heel strike. The piezoelectric thin film producessignificant voltages (~ 0.2 volts) without amplification. Thesevoltage signals can be used to predict a contact patcharea and a center of pressure. For flexoelectric-based harvesters, a process flow wasdeveloped for space charge-induced flexoelectric (SCIF)devices shown in Fig. 3. The major challenges areassociated with curvature of the sample on the releaseprocess, which leads to the pyramids delaminating from thepolymer matrix. Multiple approaches were adopted toameliorate this difficulty, resulting in the first devices thatcould be characterized electromechanically. In addition,we are working to validate basic constitutive equations,equivalent circuit models, and finite element-basedcomputational models. Standard finite elementformulations do not contain strain gradient as afundamental quantity nor coupling between strain gradientand polarization. An initial code was developed in Matlabto validate the SCIF pyramid structures. Once validated, thiscode will be moved to a more computationally efficientplatform, and carrier diffusion equations will beincorporated to more accurately model SCIF transducers.Finally, the validated models can be used in designoptimization to design SCIF transducers for the energyharvester.Impact:Falls result in > 3,000,000 trips to the emergency room, 28,000deaths, and $50B in costs, annually. This work aims toenable a technology that will allow long-term, low-burdenmonitoring of people at potential risk for falls.Objective:This project seeks to develop a self-powered pressure-sensing array in a shoe insole that can be used in studiestargeting conditions in which gait, balance, and activity arekey indicators of disease severity and/or risk. Approach:There are three technical thrusts in the project:development of a shoe insole harvester that is comfortableand does not affect the gait of the user, development ofnovel space-charge polarizable flexoelectric transducers,and development of a flexible large-area piezoelectricpressure sensing array. Recently, the energy harvester wasredesigned to improve comfort, power generation, androbustness. Several shoe insoles were characterized bymeasuring the force-displacement relationship (i.e. stiffness)and the energy dissipation. The harvester was thendesigned to mimic the stiffness and energy dissipation ofthese standard walking shoes. This re-design includes a verysmall ball-screw and a multi-lobe cam which was found tobe more robust. The new design can incorporate differentfrequency up conversion factors. With a 3-lobe cam theharvester generates 9.9 mW of power at 1 Hz actuation(similar to walking at about 2.5 mph) which nearly meetsour target of 10 mW. The estimated generated power fromthis design was 19 mW. It is believed that the discrepancy ismostly due to compliance in harvester components otherthan the piezoelectric beams (see Fig. 1).Fig. 3: Microfabricated array of space charge flexoelectric pillarsEnergy Harvesting and Self-Powered SmartInsoles for Gait and Balance DetectionPrincipal Investigators:Shad Roundy, PhD, Susan Trolier-McKinstry, PhDStudents:Travis Peters, Arash KazemiFunding Sources:ASSIST Center, Office of Naval Research5Fig. 1: Shoe-based energy harvester and example voltageoutput. Harvester generates 9.9mW when walking at 1Hz.Fig. 2 : Pressure sensing array on metal foil being testedwith a heel prototype (left). Example voltagewaveforms from four PZT sensors on array (right).

Schematic demonstrates the fabrication process of LIB yarns.Flexible Battery Yarns for Smart TextilesObjective:The advancement of smart textiles urgently calls for textile-compatible power sources for sensing, computing, anddata transmission, etc. Energy storage devices that can beefficiently and reliably integrated with other electroniccomponents, while maintaining the breathability andflexibility of the resulting fabrics, are favorable. Incomparison to their film-shaped counterparts, yarn-shapedenergy-storage units are capable of being incorporatedinto textile fabrics via automated weaving or knitting, indiverse sizes, locations, and shapes, and thus are readilydeployable to different parts of clothes. However, a robustenergy-storage yarn configuration that can withstandrigorous textile fabrication processes and various end uses isyet to be established, as is their long-term durability andbiosafety as an invisible energy source on clothing.Approach:To fabricate yarn-shaped lithium-ion batteries (LIB yarns),conventional textile techniques including yarn sizing, yarntwisting/plying, yarn encapsulation, etc., are employed toengineer active materials, current collectors, separator, andelectrolyte – all necessary components – into a single yarn(Fig. 1). Unique yarn twisting protocol is designed to induce better interfacial contact between each component inyarn electrodes, improving the performance consistencyand linear power density of LIBs. Separator ribbons areinnovatively introduced in-between yarn electrodes, withthe hope to achieve reliable power units in woven or knitfabrics, bearing repetitive mechanical deformations alongalmost all directions.Key Accomplishments:An ongoing funded project by SRI International and IARPA,to optimize the LIB yarn configurations to validate itsfunctionality and improve its durability and flexibility.Impact:Built upon our previous experience in the fabrication ofyarn-shaped supercapacitors, which are highly scalablethrough traditional textile manufacturing techniques. Thesuccessful fabrication of LIB yarns will meet the strongdemand we have received from relevant commercialentities regarding invisible power supplies for smart textiles.Principal Investigators:Amanda Mills, PhD, Veena Misra, PhD, Wei Gao, PhDFunding Sources:Prime: IARPA; Subcontracted under SRIThis research is based upon work supported in part by the Office of the Director of National Intelligence (ODNI), Intelligence AdvancedResearch Projects Activity (IARPA), via N66001-23-C-4515. The views and conclusions contained herein are those of the authors and should notbe interpreted as necessarily representing the official policies, either expressed or implied, of ODNI, IARPA, or the U.S. Government. The U.S.Government is authorized to reproduce and distribute reprints for governmental purposes notwithstanding any copyright annotation therein.6

Flexible Electrochemical Capacitor forWearable TechnologiesPrincipal Investigators:Clive Randall, PhD, Ramakrishnan Rajagopalan, PhDStudent:Linsea ParadisFunding Source:ASSIST CenterObjective:The project aims at developing a high energy densityelectrochemical capacitor with low self-dischargecharacteristics and long cycle life as a promising energystorage solution for the design of self-powered wearabletechnologies with continuous health monitoring. Thetechnology improves the volumetric energy density ofexisting state-of-the-art capacitor technologies by ~3x. Theresearch focuses on designing capacitors with flexible formfactors that are easily scalable and have good reliability.The long cycle life (>10,000 cycles), high energy density,and fast charging/discharging capabilities make it asuitable alternative for the lithium polymer batteries that areused in the current wearable device platforms.Approach:The technology is based on the development of high-capacity electrode materials that enable fabrication ofultra-thick electrodes with high fast charging anddischarging rates. The assembled capacitor also shows alow self-discharge rate with 90% capacity retention over 2months making it attractive to be used in conjunction withlow-power energy harvesting and sensor technologies. Thecapacitors can be assembled in various form factors suchas coin cell prototypes, pouch cells, or flexible cable orwire. The technology can be used either as a standaloneenergy storage solution or in conjunction with batteries forcontinuous health monitoring using wearable deviceplatforms.Key Accomplishments:Currently, we have fabricated lithium-ion capacitors (LIC)that have a cell capacitance of 4–5 F packaged in a 2016coin-cell prototype and a volumetric energy density of~13Wh/L. The cells were capable of being chargedbetween 2.2V and 3.8V. In comparison with a commercial3.6V rechargeable lithium-ion battery (LIR) of similar formfactor, the capacitor shows higher capacity retention atfast charging and discharging rates. Long-term stability testsunder constant current conditions showed that thecapacitor lasted three times longer relative to the battery(~550 hours) when charged and discharged at 8 mA. Theseresults show that LIC can provide an energy-efficientsolution for fast charging or high pulsed current loadingconditions. We have published a paper in thatdescribes the technology.Impact:High energy density lithium-ion capacitors packaged insmall form factors such as coin cell prototypes can offersignificant advantages over rechargeable and primarybatteries of similar form factor in terms of current ratings,cycle life, and higher energy efficiencies. Successfuldevelopment of the technology into various form factorsthat include flexible pouch cells, flexible wire, or cable canoffer the potential to fabricate cells/modules with differentcapacities and extend its application as a standaloneenergy storage device or its use in conjunction withbatteries. Carbon7Performance comparison of ASSISTcapacitor versus rechargeable lithiumion battery of similar form factor

Printing Metallic Pastes and Metallic ‘Gels’for Wearable Energy HarvestingObjective:To demonstrate a novel class of multiphase liquid metal(LM) ‘inks’ whose properties can be designed throughsystematic incorporation of solid additives and fluid micro-capsules with nanometer-thin oxide shells. The introductionof these easily tunable inks could enable large-scaleadoption of the materials in stretchable electronics, thermalmanagement, and medical applications. Currently, there isno simple way to print parts at room temperature withmetallic conductivity.Approach:By incorporating solid and fluid fillers, this project seeks torender LMs easier to pattern by additive manufacturing andto broaden their range of physical and chemical properties.To date, researchers have shown it is easy to distribute liquidmetal droplets in other materials (such as polymers), butnon-trivial to do the opposite. We have created metallicinks that consist of Cu particles connected by liquid metalparticles that create a ‘metallic gel’ with ideal rheologicalproperties for printing. The figure shows an example of aprinted part that resembles a spider.Key Accomplishments:To date, we have demonstrated the principle and haveshown it is possible to print highly conductive metallic partsat room temperature. Interestingly, the parts can bedesigned to change shape with time in a controlledmanner, which researchers call 4D printing (4th dimension istime).Impact:This project will create metallic materials with electricalconductivity. Areas of opportunity include expanding theink composition to a broader range of materials and thus,properties (including stretchable conductors). Our personalinterest is tuning the rheological properties to enable facileprinting of metallic materials at room temperature (Figure1). Yet, there are many other properties that can be tunedby forming other types of inks, including foams and pastes.Principal Investigator:Michael Dickey, PhDFunding Source:NSF in collaboration with ASUF3D printed ‘spider bot’ that has metallic conductivity but is printed at room temperature. The ‘metallic gel’ that makes this possible consists ofCu particles connected by liquid Ga. After printing, the part has excellent electrical and mechanical properties. We believe it is the onlymetallic ink that can be printed at room temperature to form solid parts.8

Objective:This project seeks to develop a new type ofelectromechanical transducer that can be used as areplacement for leaded piezoelectric materials, chiefly leadzirconate titanate (PZT). This new transducer is made fromdoped silicon, an environmentally benign material. Ifsuccessful, the technology could replace PZT in manyenergy harvesting, sensing, and actuation applications. Approach:It has recently been observed that semiconductingmaterials, including silicon, can exhibit significantlyenhanced flexoelectricity. This effect is referred to as SpaceCharge Induced Flexoelectric (SCIF) effect. There are twokey elements to achieve high performing SCIF transducers:1) generation of very large strain gradients, and 2) highflexoelectric coupling defined by the effective flexoelectriccoefficient. The approach being undertaken is to developmicrofabrication processes for pyramidal arrays in siliconthat will create very high strain gradients under mechanicalforce, and computational models for SCIF materials that willenable us to investigate the effect and optimize materialsand structures.Key Accomplishments:The SCIF effect in silicon was demonstrated using a boron-doped silicon beam subjected to bending. The 500 μmthick beam had 30 nm layers of hafnium oxide on eachsurface. The beam exhibited an effective flexoelectric(a) 500 μm doped silicon flexoelectric beam subject to bending.30 nm thick insulating HfO2 layer is deposited on both siliconsurfaces. Experimental measurement of effective flexoelectriccoefficient for the doped silicon beam.Space Charge-Induced Flexoelectric Transducers toReplace Lead-Based Piezoelectric Transducers coefficient of 4.9 μC/m which is orders of magnitude higherthan the underlying flexoelectric coefficient for silicon,verifying the space charge enhancement. Although stillbelow the highest reported values, the SCIF silicon beamdemonstrates the promise of this approach. A process flowhas been developed to create SCIF transducers from 3Dpyramidal arrays in silicon. A computational model offlexoelectricity has been developed in the open platformcomputational system, FEniCSx. This computational modelhas been validated against published results used tosimulate a flexoelectric pyramid (without space chargeenhancement). This model has then been extended toinclude the drift-diffusion equations for mobile chargecarriers to simulate the effective flexoelectric coefficient fora SCIF structure. Initial results indicate an enhancementfrom the presence of space charges similar to the firstexperimental demonstration with a silicon beam. This systemwill next be used to explore the effect of material (i.e.,doping level) and structural parameters on the SCIF effect.Impact:It is estimated that the annual worldwide production of PZTis between 1250 and 4000 tons. The lead contained in thePZT represents an environmental risk all along the valuechain (i.e. mining to end device disposal) and is increasinglysubject to health, safety, and environmental legislation.However, despite decades of research around the world,there is still not a lead-free drop-in replacement for leadedpiezoelectrics such as PZT. If successful, this project opens abroad array of applications for which Si can replace PZTsensors, actuators, and energy harvesters.Principal Investigators:Susan Trolier-McKinstry, PhD, Shad Roundy, PhDStudents:Travis Peters, Arash Kazemi, Ryan HawksFunding Sources:ASSIST Center, Office of Naval Research, NSF(a) Illustration of 3D array of pyramids to create aflexoelectric transducer. (b) Microfabricated arrayof space charge flexoelectric pillars. (a) Simulation of bulk flexoelectricity: the electric potential φand the through-thickness mechanical displacement u3 in thepyramid. (b) Simulation of effective flexoelectric coefficient ofthe SCIF silicon beam show in Fig. 2a. Slope is the effectiveflexoelectric coefficient including space charge effects and isthe underlying flexoelectric coefficient of silicon.9

Wearable Neuro-Biochemistry Sensors Poweredby Enzymatic Fuel CellsObjective:Our work is motivated by the ever-growing need forsustainable energy sources. Current wearable systemsgenerally rely on lithium-ion batteries, which are poor for theenvironment and are unsustainable in the long term. Ourintegrated solution combines the "Dual-BioCapacitor"system, a novel digital circuit, and a neuropeptide-Y sensor.Aiming to bridge this gap, presenting a wearable that offersa sustainable and efficient energy harvesting andmonitoring method.Approach:The proposed technology is an enzyme fuel cell employingdirect electron transfer engineered enzymes to harvestenergy from human sweat and convert electrochemicalreactions into electrical energy. This process is achieved byusing a charge pump that boosts the voltage supply fromthe enzyme fuel cells, filling a capacitor which can then beused for downstream processes. The current system isdeveloped to power a neuropeptide-Y sensor using a fieldeffect transistor through the interaction with an immobilizedaptamer. Other potential applications include wearabledevices for health monitoring, fitness tracking, and stressdetection. The status of our technology showcasessignificant progress in its design, efficiency, and potential forreal-world applications.Key Accomplishments:This work has successfully been used to construct andpower capacitors ranging from 10–1000 µF for various usecases. This occurred through 3 iterations of the Dual-BioCapacitor, improving on previous designs. We addressedissues associated with electrode polarization, dramaticallyimproving the operational stability of the system. Anotherkey accomplishment was the development of a noveldigital circuit, designed to pair with the Dual-BioCapacitorfacilitating downstream sensing. Current development isfocused on system integration between the Dual-BioCapacitor, and field effect transistor neuropeptide-Ysensor.Impact:This technology represents progress toward thedevelopment of standalone, self-powered wearables,addressing the constraints of sustainable power output. Asboth an independent module and an integratedcomponent in broader systems, it helps cater to the risingdemand for wearable technology. Given the global e-waste concerns, the Dual-BioCapacitor offers a moresustainable solution without compromising on performanceor usability.Principal Investigators:Koji Sode, PhD (UNC/NCSU), Michael Daniele, PhD(NCSU/UNC), Spyridon Pavlidis, PhD, Alper Bozkurt, PhDStudents:David Probst, Kartheek Batchu, Kaila Petersen,Hayley Richardson, Brendan ThompsonFunding Source:Nano-Bio Materials Consortium (NBMC)Dual-Biocapacitor platform capable of harvesting electrical energy from sweat to powerpotentiometric sensors, such as one for neuropeptide Y (NpY) as a marker for stress.10

Low PowerLow PowerSensing SystemsSensing Systems11

A Wireless Sensing Ecosystem to SupportBiomedical InterfacesObjective:This effort aims to create a comprehensive ecosystem ofwireless technologies tailored for applications in biomedicaland neuroscience research. These technologies combinetraditional flexible electronics and emerging soft,stretchable materials, enabling the development ofmechanically compliant wireless biomedical devices thatsupport bidirectional interactions, including stimulation andsensing, with living organisms.Approach:We adopt a multifaceted approach to this project. Wefocus on the creation of electronic hardware that forms thefoundation for wireless communication and sensor/actuatorinterfacing. We use wireless communication such asBluetooth Low Energy (BLE) and Near-Field Communication(NFC) technologies to establish bidirectional datatransmission to nearby data collection stations, includingcomputers and handheld devices. We then design andimplement advanced data analytic algorithms to enablereal-time data processing from a network of sensors, bothmultimodal and multinodal. The envisioned deviceconfigurations accommodate applications requiringimplantable, wearable, and hybrid implantable/wearableform factors.Key Accomplishments:Our efforts generated several wireless device prototypes: 1)Wireless electrochemical fast chronoamperometry devicethat is used, along with an injectable aptamer-functionalized sensor, for the detection of circulatorybiomarkers in-vivo, see Figure 1(a). This device platform canbe straightforwardly extended to other electrochemicalinterrogation protocols through firmware updates. 2)Reconfigurable NFC battery-free optogenetic stimulatorsdesigned for closed-loop behavioral brain synchronyresearch in rodents. 3) Wireless reconfigurableelectrophysiology platform which supports multimodal andmultinodal electrophysiology recordings (EEG, ECG, EMG,ENG) in both wearable and implantable versions. We havealso developed an iOS application that serves as theinterface for real-time data collection and devicereconfiguration.Impact:The technology development outlined in this projectrepresents the foundation for creating application-specificdevices and systems. These innovations will not only enablethe practical implementation of low-power, implantablewireless biomarker sensors, but they will also lay the essentialgroundwork for the design of materials- and mechanics-optimized systems, specifically tailored for seamlessinterfacing with soft tissue in long-term implantationscenarios. Moreover, the establishment of adaptable,multimodal, and multinodal wireless networks forphysiological sensors will supply a robust IOT platform,empowering both cutting-edge biomedical research andtransformative clinical applications, see Figure 1(b).Figure 1. (a) Example of the design and development of awireless BLE generic electrochemical sensor interface (Fastchronoamperometry in this specific project) and systemintegration for circulatory biomarker biosensing in-vivo. (b)Conceptualization of a technology ecosystem where multimodaldevice nodes interfaced with the body monitor physiologicalsignals that are fed into a data analytics hub for advancedinferences, prediction, and emergency actuation. Principal Investigator:Abraham Vázquez-Guardado, PhDStudents:Mohammad Riahi, Akshay Bhardwaj,Tanya Dubey, Ashwini MuralidharanFunding Source:ASSIST Center12

Label and Bio-Active Free Electrochemical Detection ofTestosterone Hormone Using MIP-Based Sensing PlatformObjective:Hormones play an important role in regulating andmaintaining physiological function. Imbalance or inability toproduce regular amounts of hormones causes severe andoften chronic illnesses such as diabetes, cardiovasculardisease, eating disorders, osteoporosis, and more.Testosterone, an anabolic steroid, is an androgen that actsas the primary sex hormone for men. The gonads and testesin men are responsible for the production of over 95% ofendogenous testosterone, producing approximately 6–7mg per day, leading to an average concentrationbetween 300 and 1000 ng/dL in the blood and aconcentration of around 263 to 544 pmol/L (7.6 ng/dL to15.7 ng/dL in saliva). Recently, biosensors have beenreceiving attention for their ability to detect testosteronenot only in a laboratory setting but with the potential forclinical and field practice as well. Therefore, with theimportance of testosterone, we developed an efficientsensing system: molecularly imprinted polymers (MIPs)technology for the sensing of hormones like testosterone.Approach:We present a MIP-based electrochemical sensor for sensingsweat testosterone levels. MIPs are synthetic polymers thathave cavities imprinted onto a matrix, through knownelectrochemical methods, that allow for the binding ofspecific molecules. The binding of a specific moleculeaffects the current passing through, leading to an electricsignal that correlates to a certain amount or concentrationof target molecule in the sample. The developed MIP-based testosterone sensor, when compared to traditionalmethods, is far simpler, cost-effective, non-invasive, andmass-producible while still being accurate, sensitive, andspecific.Key Accomplishments:The results of the research demonstrate a consistent positivecorrelation between the concentration of testosterone andoverall current behavior within the concentration range of 1ng/dL to 25 ng/dL. The developed sensor successfullydemonstrates bio-active and label-free electrochemicalsensing of testosterone using PoPD-MIP/SPCE system at low(1 ng/dL) and high levels (25 ng/dL), which is within thephysiological range for male saliva, 7.2 ng/dL to 15.7 ng/dL.The PoPD-MIP/SPCE approach for testosterone testing iseasy to execute for large-scale production and POC testingif integrated with miniaturized electronics.Impact:Screening for hormone changes using wearable andhandheld technology has the potential to revolutionizehealthcare and diagnostics. This work on this testosteronesensor can detect fluctuations in testosterone using a sensoron the skin and can also be explored for salivary hormonedetection. Further, utilizing an advanced M−P of PalmSenswith Bluetooth capabilities can be operated using asmartphone (introduction of Internet-of-medical-things(IoMT)). By combining the existing network of healthprofessionals and generating a quick and flexible flow ofmedical data, users could not only monitor testosteronelevels in real time but also receive consultation and ASSISTCenterance. Extended studies with an emphasis onintegrating the MIP-based sensing platform into stretchableelectronic technologies and software development willprovide an all-encompassing medical experience.Illustration of testosterone correlation with male body functionand its electrochemical estimation using sensing electrodemodified with MIPs of PoPD fabricated via electro polymerizationof oPD. Such a proposed sensing approach supportedminiaturized potentiostat (M−P) could be established for POCsensing of testosterone at a low and physiological range.Principal Investigators:Vivek Kamat, PhD, Shekhar Bhansali, PhDStudents:Justin Sanchez-Almirola, Alexander Gage,Raul Lopez, David YapellFunding Sources:NSF, ASSIST CenterPresentation of PoPD-MIP and electro-chemical sensing of testosterone usingPoPD-MIP/SPE electrode.Overlay of current (mA) output againstvoltage (V) for characterization tests ofthe functionalized sensor whenpresented with testosterone samples.13

Microneedle-Integrated Sensing System for MinimallyInvasive Extraction and Analysis of Dermal Interstitial FluidObjective:With the increasing interest in improving diagnostic tests,making them portable and more accessible is becoming asignificant research and commercial effort. Whileconventional blood-based diagnostic tests increase the riskof transmitting blood-borne pathogens and infection, theyare also painful, which reduces compliance across thepatient population. Sampling interstitial fluid (ISF) using ananocomposite microneedle (MN) patch enablesnoninvasive, painless alternative to common finger-stickblood draw. ISF contains comparable analytes to blood,plasma, urine, saliva, and feces and can be similarlyanalyzed.Approach:The proposed technology uses a biocompatible hydrophiliccomposite of methacrylated hyaluronic acid (MeHA) andTEMPO-oxidized cellulose nanofibers to fabricate MNs thatwhen inserted into the skin, swell with ISF that can easily berecovered from the patch for analysis. The microneedlepatches will be integrated into a wearable health andenvironmental tracker (HET) platform using an osmoticpump and screen printed electrochemical sensors fordiagnostics and monitoring. Custom fabrication of mastermolds for the MNs via 3D printing has allowed flexibility inthe geometry of the patches in order to optimize insertionand extraction of ISF.Key Accomplishments:The microneedles and osmotic hydrogel pump extractionsystem has allowed continuous extraction and transfer offluids from a skin model to a paper system. The effortsrecently have been in creating and optimizing bioreceptorsand transducers into a lateral flow system for diagnosticsand into an electrochemical sensor for monitoring. Whenthe systems are optimized, the next step will be to test thetechnology in human trials for future translation.Impact:A significant goal of the project is to develop wearabledevices for detection and monitoring of biochemistry whichminimally affects the users and can provide real timephysiological and metabolic status for an extended periodof time. This could provide personalized managementstrategies for several diseases including diabetes, and offera solution for medication compliance tracking.Principal Investigators:Michael Daniele, PhD, Alper Bozkurt, PhD, OrlinVelev, PhD, Michael Dickey, PhD, Trevor Tilly, PhDFunding Sources:DermiSense, NBMC, SEMI, AFRLStudents:Angelica F Aroche, Hannah Nissan, Kaila Peterson,Brendan Thompson, Chris Sharkey, Leslie UySEM image of microneedle patchHET platform with osmotic pumpand microneedle patch3D printed master mold for microneedle fabrication14

Miniaturized, Wearable, and FlexibleWound Monitoring SystemObjective:Diabetic foot ulcers are chronic wounds that affect millions,and increase the risk of amputation and mortality,highlighting the critical need for their early detection.Recent demonstrations of wearable sensors enable real-time wound assessment, but they rely on bulky electronics,making them difficult to interface with wounds.Approach:We have developed a miniaturized, wireless, battery-freewound monitor that measures lactate in real-time andseamlessly integrates with bandages for conformalattachment to the wound bed. The device detects woundlactate levels using a self-powered biofuel cell-based sensorand transmits the data wirelessly using near-fieldcommunication (NFC) protocols.Key Accomplishments:Our work exploits these unique prognostic properties ofwound lactate to demonstrate the miniaturized woundmonitor's capacity to predict wound closure rate early.Studies in healthy and diabetic mice reveal distinct lactateprofiles for normal and impaired healing wounds. Amathematical model based on the sensor data predictswound closure rate within the first three days post-injury with~76% accuracy which increases to ~83% when pH isincluded. Impact:Our work marks an important advancement in the field ofwound care as information provided by the describedwound monitor will be useful for timely course correction ifthe present treatment is not expected to lead to rapidwound closure.Principal Investigator:Amay J. Bandodkar, PhDFunding Source:ASSIST CenterPostdocs:Nate Garland, PhD, Abraham Vazquez-Guardado, PhD15

Ultra-Low-Power Metal Oxide Electronic Nose Array forBreath, Skin, and Environmental MonitoringKey Accomplishments:Our team successfully engineered a monolithic 4x4 array ofALD MOX sensors, capable of being heated to varyingtemperatures and producing diverse metal oxide surfacesthrough on-chip annealing. This innovative array wasemployed to measure trace concentrations of NO2, CO,Acetone, Ethanol, and ozone, all at very lowconcentrations. The portable system is compact, featuringgas sensors, and a Wi-Fi- and Bluetooth-enabled MCU, allintegrated onto a small 3 cm x 3 cm PCB board.Impact:The portable system we have developed has the capabilityto track toxins and VOCs, making it a versatile tool forvarious applications. One significant application is its use asa breath-analyzer, enabling passive detection of diseasesthrough the analysis of specific compounds present inexhaled breath. Additionally, the system proves invaluablein environmental monitoring, allowing real-time detection ofharmful gases and pollutants in the air.Objective:The project aims to develop an innovative Ultra Low-PowerMetal Oxide Electronic Nose Array, leveraging cutting-edgesensor technologies, for real-time and continuousmonitoring of diverse compounds, odors in human breath,skin emissions and environmental safety. This dedicatedelectronic nose system will be highly sensitive, selective, andenergy-efficient, enabling accurate analysis of volatileorganic compounds (VOCs) at low concentrations. Byestablishing a dynamic association between VOCs andphysiological/psychological health, the objective is tocreate a wearable, non-invasive solution for comprehensivehealth monitoring.Approach:Our approach centers on advancing single-layer metaloxide (MOX)/heterojunction metal oxide sensors andseamlessly integrating them with cutting-edge technology.This integration is achieved through a pioneering monolithicprocess rooted in CMOS processing, MEMS, and atomiclayer deposition (ALD) techniques. Notably, our systemexcels in monitoring VOCs and toxic gases over extendeddurations. Moreover, our e-nose array is designed for ultra-low powerconsumption, operating at less than 1 mW. Suchexceptional energy efficiency ensures prolonged batterylife and facilitates the development of compact, wearabledevices suitable for continuous, unobtrusive monitoring. Inaddition to its sensitivity, our system incorporatessophisticated machine-learning algorithms. Throughadvanced data analytics, the e-nose can distinguishbetween a wide range of gases and correlate them with airquality indicators and metabolic states.Principal Investigators:Veena Misra, PhD, Bongmook Lee, PhDPostdoc/Student:Mahaboobbatcha Aleem, PhD, Yilu ZhouFunding Sources:ASSIST Center, NSFDesign and Architecture of Sensor Array, Infrared ThermalImaging, and Wire-Bonded Sensor Array Integration for AdvancedGas Sensing TechnologiesPortable Gas Sensor System with Continuous Monitoring Portal and Principal Component Analysis for Various Gases and Mixtures16

Ultra-Thin Polymer-Based Flexible Platformsfor Wound and Skin Temperature SensingObjective:There is a critical need for at-home health monitoringcapabilities to provide real-time assessment withoutinterruptions to daily activities or treatment burden. Underthe current standard of care, mortality rates for diabeticlower extremity complications have become comparableto cancer. In order to enable long-term monitoring,electronic devices need to be imperceptible when appliedand robust to accommodate motion. Current devices aretoo bulky, rigid, and unreliable to be applied for long-termmonitoring in sensitive locations, such as chronic wounds.Our work aims to develop a next-generation platform forthe integration of novel sensors and electronics towardimperceptible long-term health monitoring.Approach:Our team has developed an ultra-thin polymer-basedsensing platform, which supports integration of Surface-Mount Devices (SMDs), additive fabrication of functionalsensors, and vertical integration of organic electronics.Material selection and fabrication methods are aimed atpatient comfort and high-volume manufacturability forhuman subject trials.Key Accomplishments:Our team has achieved a biocompatible (ISO 10993),hermetically-sealed, and mechanically stable (2 million+cycles) platform for health monitoring devices. Our platformwas demonstrated and evaluated for wireless temperaturesensing in human subject trials. The devices were evaluatedagainst polyimide flex devices and demonstrated fewermotion artifacts, higher wireless connection stability, andhigher user comfort. Electrochemical and physical wound-sensing devices have been fabricated and evaluated in ananimal study for monitoring of wound healing with differentdressing types.Impact:An ongoing collaboration is aimed at improving healthoutcomes in patients with diabetic foot ulcers through at-home monitoring with wound-interfacing devices. In otherwork, we aim to apply our platform to provide real-timefeedback during revascularization. Through imperceptiblelong-term monitoring, we aim to make healthcare moreaccessible and more personalized to improve healthoutcomes.Principal Investigators:Vladimir Pozdin, PhD, Alper Bozkurt, PhD, Michael Daniele, PhDFunding Sources:ASSIST Center, IndustryStudent:Mauro VictorioConformable temperature sensor enablesimperceptible readings during active andpassive activities. BLE enables real-time datavisualization with an iOS device.Sensing array is directly integrated into various wound dressings.17Autonomous woundmonitoring. Conformabledesign enables placementof the sensing arraydirectly on the wound withcontrol electronics nearthe wound site. Woundhealing was assessed inanimal models.

Wearable Ultrasound Systems for MuscleActivity Sensing in Assistive RoboticsObjective:In modern medicine, the control of assistive robotic (AR)devices hinges on the sensing of muscle activity. Yet,current techniques, such as EMG signals and traditionalultrasound, suffer from limitations and bulkiness unsuitablefor AR control. With the absence of reports on wearabletransducers for muscle activity in AR, this project's objectiveis to develop a wearable ultrasound transducer tailored forthe purpose.Approach:Employing 1-3 composite PZT-5H/Epoxy, we propose thefabrication of a 64-element linear ultrasound array with a7.5 MHz center frequency and optimized pitch size forreduced side lobes. The addition of matching and backinglayers enhances array performance, enabling ultrasoundimaging of the target muscle's movement. This linearultrasound array can also be customized into a flexible,wearable form to suit the specific needs of various musclelocations.Key Accomplishments:Our team has successfully fabricated a 64-elementultrasound transducer array, measuring aperture of 0.13 mm× 5.5 mm for each element. Initial experimentsdemonstrated the transducer's capability to monitor muscleactivity using A-mode signals from 64 elements, and we areactively working towards achieving real-time B-modemuscle imaging.Impact:Our accomplishments thus far include the successfulfabrication of this 64-element ultrasound transducer array,with demonstrated functionality in detecting musclemovement. These achievements mark significant progress inthe development of muscle activity sensing technology. Thepending realization of real-time B-mode imaging holdspromise for transformative advancements in assistiverobotics and modern medicine, addressing the criticalneed for precise and real-time muscle activity monitoring.Principal Investigators:Nitin Sharma, PhD, Xiaoning Jiang, PhDPostdoc/Students:Xiangming Xue, PhD, Sunho Moon,Vidisha Ganesh, Krysten LambethFunding Sources:NIH, NSF(a) 64-element linear US array (b) Optical images of active area of the array, showing interconnection (c)Linear US array integrated with 3D printed holder for connection with Verasonics system (d) A representativeA-mode ultrasound (US) dataset acquired from muscle activity monitoring via four designated channels.18

Low Power Circuits andLow Power Circuits andSystems-on-ChipSystems-on-Chip19

Bio-electro-photonic Microsystem Interfaces forSmall Animal Health MonitoringIn addition, wearable systems are being used within otheranimal training applications to analyze more complexmetrics like gait and pulling force. Key Accomplishments:The capsule system has been evaluated in clinicalenvironments for tracking physiological signals in rats andchickens. The collar and harness systems have beendeployed in the field with guide dog puppies to improvethe puppy training program outcomes. Machine learningtechniques such as convolutional Long Short-Term Memory(LSTM) networks have been able to attain 40% accuracy ofaction states from guide dog evaluation tests using thecollar alone. We have started to use latent variable statesthat correspond to certain data patterns to define a lexiconof data generated from common canine behavioral states.Impact:The microsystem under development enables physiologicalmonitoring of small animals in their natural environment.These systems have a broad range of applications andimpacts from agricultural production to veterinary researchand beyond. They will provide new methods and insightsthrough their minimally intrusive design and novelcapabilities. Truly accurate physiological monitoringnecessitates such minimally intrusive design and bolsters thepotential impact of this kind of system and range of itsapplications. The lexicon helps with predicting commoncanine behaviors in the field.Objective:This project aims to wirelessly monitor biophotonic andbioelectrical physiological signals in small animals usingremotely powered wearable and injectable systems. Thismicrosystem is in response to the critical need for a novelminimally invasive class of devices for the continuousrecording of key physiological parameters of animals innatural environments without disturbing their naturalbehavior.Approach:This project involves the development of two parallelphysiological and behavioral sensing platforms in twodifferent form factors: an injectable subcutaneous capsuleand a wearable harness and collar system for animals. Thecapsule platform provides subcutaneous photoplethys-mography, electrocardiography, accelerometry, andthermometry. This is used to calculate heart rate, respirationrate, oxygen saturation, pulse transit time, and core bodytemperature. The harness system is for more simplewearable applications with electrocardiography,photoplethysmography, inertial and environmentalmeasurements, and audio recording integrated into astandard dog harness and collar. This system has been usedin training dogs to follow or interact with unmanned-airvehicles with the goal of deploying this in working dog rolessuch as search and rescue operations and agricultural pestdetection. Recent efforts involve combining the system withhuman wearable devices to collect and analyzesynchronized biometric data in human-canine interactions. Principal Investigators:Alper Bozkurt, PhD, David Roberts, PhD,James Dieffenderfer, PhD, Yaoyao Jia, PhDPostdocs/Students:Parvez Ahmmed, PhD, James Reynolds, PhD, Devon Martin,Colt Nichols, Maxwell Noonan, Natalie Smit, Yifan WuFunding Sources:NSF, IBM20

Body Area Network of Inertial Sensors forClinical Gait RehabilitationObjective:Monitoring of gait and lower limb function is of great interestto clinicians overseeing patients recovering frommusculoskeletal injuries and surgeries, and those at risk oflower limb dysfunction due to advanced age or overallhealth status. Existing inertial sensors capable of obtainingthis information are prohibitively expensive, makingwidespread monitoring of these patients impractical. Ournovel system of inertial sensors is intended to provide a low-cost option for monitoring of patient gait using a wirelessbody-area network (WBAN) of self-contained sensorperipherals, capable of recording movement and sendingthese data to a central hub for analysis using machinelearning techniques.Approach:Our system consists of up to eight wearable devicesinterfaced wirelessly with an off-body control unit. Eachperipheral device contains an inertial measurement unit(accelerometer, gyroscope, and magnetometer) andsupporting circuitry needed to enable continuous on-bodyoperation for up to 4 hours. Data from each sensor channelis sent to the control unit in 44-byte packets at a rate of~50Hz/device, sufficient to capture detailed gaitinformation. Additional planned peripheral sensors aim tocapture other data streams, such as heel drop pressure, toimprove monitoring of patient status.Key Accomplishments:At present, we have validated the use of these peripheralsfor collecting gait data as part of a body-area network.Human trials involving deliberate over- or underloading ofparticipant vertical ground reaction forces havedemonstrated our system’s ability to capture subtlechanges in gait with high resolution. Additionally, an add-oncircuit for heel pressure sensing has been successfullydemonstrated and will be incorporated into upcomingtrials. Ongoing work will focus on integrating the WBAN withmachine learning to provide real-time feedback, as well ascontinued miniaturization of the individual sensors.Impact:Our WBAN system will enable clinicians to monitor patientsfor minute changes in gait during rehabilitation, as well asproviding prophylactic surveillance for harmful changes ingait within vulnerable patient populations (e.g., post-surgery). Importantly, the low cost of this system willfacilitate detailed monitoring in real-world settings, ensuringa complete picture of the user’s lower limb function isobtained during the performance of daily tasks, rather thanstereotyped movements in a lab setting.Principal Investigators:Michael Daniele, PhD, Jason Franz, PhD,Brian Pietrosimone, PhDFunding Sources:NSF, UNC Eshelman School of PharmacyStudents:Jack Twiddy, Max Yates, Kaila Peterson,Ethan Cove, Ricky PimentelPrototype modular WBAN peripheral (right) and miniaturized version 2 device (left)Representative data from on-body trials, showingdistinct differences in measured gait data betweenoverloading and underloading conditionsWBAN peripheral interfaced with prototype heel drop pressure sensor21

Compressive Sensing-Based Ultra-Low PowerPhotoPlethysmoGraphy (PPG) Module for Wearable SystemsObjective:This project focuses on developing ultra-low power andnovel biophotonic sensing techniques for wearablephysiological monitoring that can be incorporated withother sensors for correlated sensing experiments. The state-of-the-art biophotonic circuit designs reduce powerconsumption using techniques like logarithmic amplifier,heartbeat-locked loop, light-to-digital converter, etc.However, most of these circuits were evaluated on thefinger using a benchtop system. This project demonstratedthat compressive sensing is one of the lowest powerconsuming techniques on the system level and evaluatedits performance on the wrist.Approach:Photoplethysmography (PPG) systems, although convenientto deploy in wearable devices, consume a lot of energydue to the light source used for capturing photonic datasamples. Compressive sensing, which implements a randomsampling scheme to optimize signal acquisition, along witha random-sample based heart rate extraction methodnamed Lomb-Scargle Periodogram, minimizes the energyusage of the light source in the PPG system. This projectenabled the incorporation of the compressive sensingtechnique into an ultra-low power application specificintegrated circuit (ASIC) in collaboration with IMEC. Weworked on the miniaturization and integration of this novelcompressive sensing based PPG ASIC into ASSIST’s Healthand Exposure Tracker (HET) testbed in the form factors of awearable wristband and evaluated its accuracy andusability to track heart rate on the wrist.Key Accomplishments:The system-level miniaturization aimed for a wearable formfactor was achieved along with no compromise in theperformance of the ASIC. The ASIC consumes 172 μWpower to extract heart rate from the PPG signal where thewhole system consumes 1.66 mW power for continuousstreaming of heart rate data over the commercial off-the-shelf Bluetooth Low Energy radio of the HET-engineeredsystem. After the initial proof-of-concept validation, wecompleted in-vivo preclinical studies with 15 subjects toformally evaluate the accuracy of heart rate valuesextracted from the PPG signals of different compressionlevels implemented in the ASIC in comparison to acommercial gold-standard system. We were able todemonstrate the wrist-worn system as an efficient platformfor future integration of the compressive sensing-based PPGsubsystem with other sensing modalities available in the HETTestbed environment.Impact:Photonic measurements, such as PPG and pulse oximetry,are the most common methods in wearable systems totrack cardiovascular health. On the other hand, these arealso some of the most power-consuming modalities due tothe necessity of generating a large number of photons.Although most of this is lost due to absorption andscattering in the tissue, important hemodynamicparameters are assessed in return. An ultra-low power ASICfor PPG is required to overcome translational barriersrelated to the use of these systems as a part of self-poweredor extended battery life-based operation.Principal Investigators:Alper Bozkurt, PhD, James Dieffenderfer, PhDPostdoc:Parvez Ahmmed, PhDFunding Sources:ASSIST Center, IMECPPG ASIC micrograph(top left), Wire-bondedASIC (top middle),assembled boards (topright), and wearablewristband (bottom)22

Miniaturized Extended Gate Field Effect TransistorReadout System for Wearable Analyte SensingKey Accomplishments:Our EGFET circuit has demonstrated expected behaviorover a wide range of simulated input conditions. Toevaluate performance in solution, an electrodefunctionalized to detect Levodopa (L-DOPA) - used in themanagement of Parkinson’s disease - was evaluated as atest case. Using various bias settings and a gain of10,000V/A, saline blanks (no L-DOPA) were successfullydifferentiated from a 60μM L-DOPA solution (also in saline).Additional characterization experiments are beingperformed to optimize biasing conditions to ensuresensitivity over the therapeutically relevant range. Further,additional revisions to the device are being prepared todramatically reduce the system size, ensuring maximumusefulness in a wearable context.Impact:Our system will allow for low-power and small form factordetection of a range of clinically relevant analytes, limitedonly by the availability of suitable recognition elements. Theflexibility inherent to our design maximizes the applicabilityof this device to a huge number of possible use cases.Objective:Measurement of analytes in human biofluids represents amajor avenue for longitudinal monitoring of an individual’shealth, generating data throughout the course of the dayand varied user activities. Detection of changes inelectrochemical potential that occur when an analytebinds to an electrode-bound recognition element allows forrapid, label-free monitoring of specific biomarkers.However, direct measurement of open circuit potential(OCP) necessitates specialized high-impedance inputelectronics, which can be costly, complicated, and power-hungry.Approach:Our system implements a small form factor, wearableelectronic system designed to allow for sampling andamplification of analyte binding events using an extendedgate field effect transistor (EGFET). In this scheme, the gateof a field effect transistor allows for amplification of thevarying OCP, while providing an extremely high inputimpedance, using only a single transistor. This current is thenconverted into a readout voltage signal using atransimpedance amplifier, which is then digitized. Becausethe EGFET sensor can be used with an electrodefunctionalized for any desired biorecognition event, systemparameters can be varied according to specificexperimental needs (-2.5V to 2.5V for both transistor drainand electrochemical cell bias potential, and any valuegreater than 1V/A for amplifier gain).Principal Investigators:Michael Daniele, PhD, Koji Sode, PhDStudents:Jack Twiddy, Ethan Cove, David ProbstFunding Source:ASSIST Center23Detection of L-DOPA in vitro to validate efficacy, showing adecrease in transduced current corresponding to thepresence of L-DOPA, which varies with varying transistorand cell bias potentials.Wearable EGFET board for systemdemonstration and characterization

Ultra-Low Power Integrated Systems-on-Chip (SoCs) forWearable Devices and Internet of Things SolutionsObjective:This project seeks to develop a core electronics platform forintegrating the technologies from other ASSIST thrusts into aunified self-powered sensing system with a total powersustainable by energy harvesting, targeting a budget of lessthan 50 uW, and having flexible multimodal capabilities forthe second-generation Self Powered Adaptive Platform.The core research in the effort investigates and developscustom chips to enable the system level integration andperformance at these low power levels.Approach:The multi-chip platform, centered around a system-on-chip(SoC), includes circuits for data collection, data storage,data processing, node control, power management, powerharvesting, power delivery, and wireless communication.This project plays a vital role in the strategic plan of thecenter, since SAP 2.0 cannot operate from body-harvestedenergy without the low power electronics from this project.This project is the hub around which many of the otherprojects across different thrusts are arranged, and it isdriving the core capability for self-powered operation.Key Accomplishments:Demonstrated SoC (with RISC-V MCU, 8kB memory, bootROM, on-chip clocks) integrated with flexible analogfront-end chip. Demonstrated a new flexible analog front end (AFE)with photoplethysmography (PPG) power as low as 9.35μW.Built printed circuit boards for integrated wearabledesign using custom ASSIST chips and components.Demonstrated full system integration of ASSISTcomponents into a working PPG system at the targetpower levelsPython system model for duty-cycled, hierarchical self-powered systems.Impact:This work is enabling the ASSIST SAP 2.0 as well as futuregenerations of the SAP systems to operate entirely fromharvested energy, due to the low power operation andflexible functionality of the custom chips.Principal Investigator:Ben Calhoun, PhDStudents:Xinjian Liu, Sumanth Kamineni, Shourya Gupta, Natalie Ownby,Katy Flynn, Peter Le, Suprio Bhattacharya, Omar Faruqe,DaeHyun Lee, Akiyoshi Tanaka, Anjali Agrawal, Nugaira MimFunding Source:ASSIST CenterSAP 2.0 multi-chip systemblock diagramDemonstration of SoC with analog front end chip and PPG power as low as 9.35 μW (left) and close-up view of SoC and AFE chips (right).24

Ultrasonic Energy/Data Transfer for Implantable SystemsObjective:A diverse set of stimulators, sensors, and data transceiverscan be integrated into micro-scale medically implantabledevices. Ultrasonic power transfer provides substantiallyhigher power density and reaches much deeper in tissuecompared to alternative sources using inductive couplingor radio-frequency (RF). A key advantage of ultrasonicenergy transfer over the competing RF technology is thatthe maximum allowed power level in tissue for diagnosticultrasound is 7.2 mW/mm2, which is about 70x highercompared to RF energy limits. Furthermore, attenuation ofultrasonic signals in tissue is far less than RF, and thewavelength of the ultrasonic energy in tissue is on the orderof millimeters. These advantages translate to a small devicesize and excellent range in biological systems. In this project,we aim to develop a miniature ultrasonically powereddevice integrated into an endo-vascular aneurysm repair(EVAR) stent-graft that could provide on-demanddiagnostic information about the presence of endoleak (acondition leading to pressure buildup in the aneurysmsack), based on measurements of the aneurysm sackdimensions, and of the stent-graft inside the vessel lumen.Approach:Our approach to implement the described implantabledevice relies on using a capacitive micromachinedultrasonic transducer (CMUT) with integrated electroniccircuits to function as an ultrasonic power receiver, adistance measurement sensor, and a transmitter for wirelessdata transfer to an external unit.Key Accomplishments:To date, our team has demonstrated the feasibility of thepresented approach which includes key concepts for animplantable intravascular ultrasound device tomonitor/diagnose endoleak in endovascular aneurysmrepair stent-grafts. In early benchtop studies using externallybiased CMUTs and off-the-shelf discrete components, wedemonstrated: 1) Greater than 1 mW power recovery froma 3-mm2 device with incident ultrasound intensity of 5mW/mm2, which is less than the spatial-peak temporal-average ultrasound intensity (ISPTA) limit of 7.2 mW/mm2 setfor diagnostic devices. Ultrasonic biphasic communicationconcepts with potential for high data rate and pulse-echoranging from sensor to EVAR structures have also beenshown. Most recently we have demonstrated a customintegrated circuit (IC) that interfaces with a pre-chargedCMUT device for ultrasonic energy harvesting. Weimplemented an adaptive high voltage charge pump(HVCP) in the proposed IC, which features low power,overvoltage stress (OVS) robustness, and a wide outputrange. The ultrasonic energy harvesting IC is fabricated inthe 180-nm HV BCD process and occupies a 2 × 2.5 mm2silicon area [1]. We have also demonstrated a novel devicestructure to implement pre-charged CMUTs [2].Impact:The results accomplished to date show the potential ofCMUT-based powering, sensing, and wirelesscommunication for implantables in a broad range ofapplications ranging from cardiovascular health to neuralsensing and stimulation. Principal Investigators:Ömer Oralkan, PhD, Yaoyao Jia, PhD,F. Yalcin Yamaner, PhDStudents:Muhammetgeldi Annayev, Linran ZhouFunding Sources:NSF, ASSIST Center (C2C Program)Pre-charged CMUTs fabricated on a4-inch borosilicate glass wafer.25Prototype ultrasound powerreceiver including a custom ICand a pre-charged CMUT.

Objective:Electrocardiograms (ECGs) are commonly done using up to12 long leads attaching a patient to a large machine. Thesetethered measurements are problematic as the long leadsare a source of failure, do not allow the patient freedom ofmotion, and commonly get tangled with themselves andother sensor wires causing incorrect readings anddisconnections. Most problematic is the time it takes inemergency situations to remove the ECG tethers to move apatient or to re-connect them when time is critical, andseconds can make a difference. Current wirelessapproaches that remove the need for tethers suffer fromlarge power requirements and depend on large and heavybatteries, creating bulky and uncomfortable devices.Approach:Our proposed solution is a wireless ECG system that utilizesbackscatter communication – communication viareflections – in the form of Bluetooth packets. By leveragingbackscatter communication, the power consumed by theECG sensor is driven extremely low (~10 uW), allowing thesensor to be self-powered by human sweat. This approachdoesn’t require a battery and its footprint can be madeextremely small, on the order of a postage stamp. Wirelessly-Powered Bluetooth Backscatter-Based Electrocardiography SystemTypically, backscatter systems suffer from overhead andinfrastructure problems as they require RF sources to coverthe desired area of communication, but this approachsolves this conventional issue with an innovative and smallRF source that plugs into the phone/tablet/PC that isdisplaying the ECG information. This approach has thepotential to create multi-sensor body-area networks withlow-cost and infinite shelf-life sensors.Key Accomplishments: We have fabricated and tested prototypes of the body-worn ECG sensor, the RF source board, and a customphone app that displays the recorded ECG. We have alsotested this prototype with a sweat-powered “battery” withno drop in performance. We are currently optimizing thedesign of the ECG sensor board on a flexible substrate andintegrating it with ECG electrodes and a sweat-powered“battery.”Impact:The test results of the wireless Bluetooth backscatter ECGsensor system show its potential to revolutionize ECGmeasurement methods in a hospital setting and as aplatform for next generation body-area sensing networksand the future of wireless health monitoring.Principal Investigators:Jordan Besnoff, PhD, David Ricketts, PhD, Amay Bandodkar, PhDFunding Sources:ASSIST Center and NC State Chancellor's Innovation FundSelf-powered Bluetooth backscatter ECG sensor system diagram showingbody-worn ECG sensor block diagram and plug-in RF sourceBluetooth backscatter ECG sensor board prototype on the leftwith major areas highlighted, and plug-in RF source connectedto a phone showing ECG recording app on the right.26

Smart E-Textiles andSmart E-Textiles andFlexible MaterialsFlexible Materials27

Advanced Flexible and Textile Antennas forOn-Body Wearable ApplicationsObjective:Low system complexity paired with high system efficiency,to enable low power operations in extremely small formfactors, are key design goals for on-body or wearablecommunication systems. RF switches, which areconventionally used to facilitate dual transmit/receiveoperations, increase the system complexity and result insub-optimal efficiency. In our work, we aimed to bothdecrease system complexity and significantly increasesystem performance by developing a dual-port duplexingantenna such that either of the two ports can be directlyconnected to the Tx or Rx modules. The resulting antenna isa great candidate for minimally intrusive on-bodycommunication devices.Approach:In general, dual-port duplexing antennas suffer from lowisolation ratio between the ports which decreasesperformance. To combat this, we introduced loading stripsbetween the ports and antenna to improve port isolation.An additional benefit of the loading strips is that theycontribute to the second electromagnetic resonance ofthe device, resulting in a wider bandwidth than theunmodified antenna. Parametric simulations of the stripgeometries reveal significant tuning capability of thedevice which may allow for optimization in a wide range of applications. These simple additions can be widely utilizedfor wearable antenna applications because of easyfabrication and the correspondingly small footprint.Key Accomplishments:Our main accomplishment is the high isolation ratiobetween the two ports and the wide bandwidth. As can beseen in Fig. 2, we can increase the bandwidth of theantenna by varying the length of the strip. The -10 dBbandwidth of the proposed antenna is 120 MHz, which isthree times larger than the antenna without the loadingstrips. This bandwidth enhancement is due to the additionalresonance produced by the strips. Moreover, an excellentisolation ratio of about -25 dB is achieved, which is 15 dBlower than the antenna without the strips.Impact:Our work simultaneously realized wide bandwidth operationand high isolation ratio between the transmit and receiveports by introducing loading strips. Our prototype iscompact and easy to fabricate, demonstrating that thistechnique can be widely used in the future of wearableantenna design. Aside from wearable applications, the low-complexity nature of our design can be readilyimplemented for PCB, textile, and on-chip antennaapplications.Principal Investigator:Douglas Werner, PhDPostdoc/Students:Chunxu Mao, PhD, Yuhao Wu, Dongha YangFunding Sources:Penn State College of Engineering and an endowment fromthe John L. and Genevieve H. McCain Chair ProfessorshipFig. 5. (a)Textile antenna mounted on different parts of a human body subjects.Our work simultaneously realized wide bandwidth operation and high isolationratio between the transmit and receive ports by introducing loading strips.(b)Measured S-parameters of the duplex textile antenna.Fig. 3. Simulated S21 of the proposedantenna and a conventionalantenna without stripsFig. 4. Fabricated antennaprototype (a) Rigid PCB version(b) Textile version based onscreen-printing technology.Fig. 1. Configuration of theproposed dual-port duplextextile antenna.Fig. 2. Simulated S-parametersvary with different values ofstrip length.28

Comfortable Outfits from Utilitarian Textiles forUnobtrusive Recording of Events (COUTURE)Objective:The objective of this research is to develop an integrated,textile system that can record a conversation between twoor more people. In particular, all components used in thesystem: the sensor, computation and data storage unit,power source, and interconnects, must be integratedseamlessly within the textile substrate. Approach:The NC State team is leading efforts to develop a lithium-ionbattery yarn, inkjet printed soft capacitive switch, printedhaptic feedback mechanism, and integrated connectionsand interconnects. In addition, the garment pattern anddesign are being generated at NC State concurrently withdevice development. Findings from the garment designprocess inform the device integration and vice versa. Thus,we have a novel approach where technical and artisticdesign are being considered at each stage of researchand development.Key Accomplishments:Our team has made significant progress on the creation ofa woven electronic grid that serves to bus power and dataover a defined area. This grid remains functional afterdeformation and multiple laundering cycles. Additionally,we have integrated active filament (which houses rigidcomponents) and microphone filament into the base,woven electronic grid textile which is the first step tocreating the complete system. We have begun to evaluatethe longevity and durability of soldered connectionsbetween discrete components.Impact:By developing a smart system that is entirely textile-basedand washable, we will create a new type of electronicdevice. In particular, the fiber format of common electronicdevices that we are using in this effort will lead to additionalresearch for both materials and soft electronics. What isparticularly unique about this garment is that all electronicsare washable, including the battery, without having to bedetached from the fabric.Principal Investigators:Amanda Mills, PhD, Veena Misra, PhD,Wei Gao, PhD, Cassandra Kwon, PhDPostdocs/Students:James Reynolds, PhD, Faisal Abedin, PhD, Nanfei He, PhD,Emily Schmidt, Erin Parker, Faisal Ahmed,Daniel Weispfenning, Amanda O’BrienFunding Sources:Prime: IARPA;Subcontractedunder SRIThis research is based upon work supported in part by the Office of the Director of National Intelligence (ODNI), Intelligence AdvancedResearch Projects Activity (IARPA), via N66001-23-C-4515. The views and conclusions contained herein are those of the authors and should notbe interpreted as necessarily representing the official policies, either expressed or implied, of ODNI, IARPA, or the U.S. Government. The U.S.Government is authorized to reproduce and distribute reprints for governmental purposes notwithstanding any copyright annotation therein.Initial ‘listening’ garment design and proposed component layout.29

Eco-Friendly Screen Printing of Silver Nanowiresfor Flexible and Stretchable ElectronicsObjective:Screen printing is a promising route towards high throughputprinted electronics. Currently, conductive inks involvecomplex formulations with non-biodegradable binders,organic solvents, and toxic surfactants in the ink'scomposition, making them unsuitable as an eco-friendlyprinting technology. Besides, high-conductive particleloading and multiple post-printing steps are typicallyrequired to achieve the desired electrical conductivitywhich impairs scalability of the screen-printing process.Approach:In this work, we developed an eco-friendly silver nanowire(AgNW)-based ink that is comprised of poly(ethylene)oxide as a biodegradable binder and deionized water as agreen solvent with no toxic surfactants in the inkformulation. With AgNWs as conductive fillers and thermalannealing as a single step post-printing treatment, highlyconductive features with high resolution and solventresistance were fabricated.Key Accomplishments:The developed AgNW-based ink produced patterns with aconductivity of as high as 6.70×106 S m-1 even at a lowconductive particle loading of 7 wt%. A single-step post-printing treatment with a relatively low thermal annealingtemperature of 150°C made the process compatible withplastic substrates like polyethylene terephthalate (PET) andalso suitable on a large scale. In addition, the capability toprint complex patterns on a diverse range of substratessuch as polydimethylsiloxane (PDMS), polyimide, plastics,glass, and even rough textile surface was achieved, whichis difficult to achieve with other printing methods. Exploitingthese features, wearable devices based on the printedAgNWs are demonstrated by fabricating flexible heaters ontextiles and wearable hydration sensors on PDMS.Impact:This work provides a promising pathway for sustainablyprinted electronics. The screen-printing process incombination with the AgNW-based ink and post-printingtreatment creates large-scale manufacturing possibilities ofprinting highly conductive complex patterns on diversesubstrates.Principal Investigator:Yong Zhu, PhDFunding Source:NSFPostdoc/Student:Shuang Wu, PhD, Darpan Shukla(a) Demonstration of screen-printing process, (b) thermal annealing as post-printing treatment,(c) conductivity of the printed AgNWs and (d) demonstration of printing capability on various substrates. 30

A POMaC circuit composed ofresistor and LED operating in PBS.Scale bar is 9 mm.A. Stress strain curves of POMaC polymer achievable with changing processing conditions. B. Resistanceunder strain for encapsulated POMaC circuit over 1000 cycles. Image shows the circuit tested.Principal Investigator:Michael Daniele, PhDPostdoc/Students:He Sun, PhD, Brendan Turner, Kirstie QueenerFunding Sources:ASSIST Center, NSFFlexible Organic Poly(Octamethylene Maleate(anhydride) Citrate) (POMaC) CircuitsObjective:This work seeks to develop an elastic, bioresorbablesubstrate capable of supporting stretchable electronicsalong with cell payloads with the goal of providing bothmonitoring and therapeutic capabilities. There is a need fordevelopment of implantable medical devices (IMDs)capable of integration with host tissue. Biointegration ofdevices with tissue has been limited by non-stretchable,non-degradable materials utilizing a non-interactivematerial design paradigm. These non-interactive materialsrequire removal surgery at end of implant life and typicallyresult in additional local inflammation due to non-matchingphysical and biological properties.Approach:We have developed a citric acid-based elastomer as asubstrate for flexible and stretchable electronics. Thepolymer is made up of human metabolites and can bedegraded in vitro into its substituent monomers. The materialhas a history of usage as cell scaffolds in regenerativemedicine and can be tuned to match physical propertiesof various native tissues. Devices fabricated using thistechnology could be used for transient implants, single-usedegradable biosensors, tissue regeneration, andcombinations thereof. As part of this work, we plan tofabricate a degradable, wireless power module that couldbe used to power a variety of systems.Key Accomplishments:The mechanical properties of POMaC films can be tunedwith moduli ranging from 0.4-2.3 MPa (matching a widerange of human tissue). A fabrication methodology forPOMaC circuits has been developed and used to producecircuits capable of operating in simulated biofluid. POMaCcircuits have been characterized under strain showing R/R0~2.3 after 1000 strain cycles to 20% strain. Proof-of-conceptphotopatterning has been shown and illustrates thepotential for biomolecular patterning of the material toenhance cell applications.Impact:Successful development of the POMaC circuit system willlead to implantable devices capable of providing both thetherapeutic and diagnostic functions offered by currentIMDs along with specific biotherapeutic payloads (cells orbiomolecules). For example, a device capable of targetedmyocardial regeneration while providing transient pacingalong with diagnostics to monitor healing would be anexciting application. Further, soft, degradable andstretchable electronics would be desirable in fields likewearables, soft robotics, and green technology.313D-printedPOMaC/carbonpaste mixture ontoPOMaC substrate

Integrated Tactile Interfaces and Physiological Sensing ArraysUsing Sustainable 3D-Printed Hydrogels for Wearable SystemsObjective:This project brings a transformative novel approach inmaking high-performance sustainable hydrogel sensingpatches including piezoresistive, piezoionic, andbiomolecular sensing elements. These interface skin andreadout signals are obtained using a miniaturized low-power wireless module. The sensing arrays are enclosed inbiopolymer matrices that match and conform well tohuman skin. The patches are fabricated by 3D printingbased on our pioneering methods in additivemanufacturing with hydrogels. These are inexpensive,transparent, ultrasoft, skin-friendly and completelybiodegradable.Approach:Emerging technologies for touch-based skin-mountedsensors require the development of wearable sensingpatches that are naturally sourced and based onsustainable sources, thus reducing the problem of e-waste.This project is based on a transformative approach and setof techniques for making broadband piezoresistive andpiezoionic patches. The new patch-like sensors use anumber of breakthrough inventions and scientific advancesmade by the dynamic collaborative team. Theseinnovations include the synthesis of new fibrillar networksfrom biopolymers, forming hierarchical piezoresistivehydrogel networks that generate changes in electricalresistance through the reversible breakage and adhesion ofnanofibrils within a 3D network. Principal Investigators:Lilian Hsiao, PhD, Orlin D. Velev, PhD,Amay Bandodkar, PhD, Alper Bozkurt, PhDFunding Sources:NSF, ASSIST CenterStudent:Pedro Henrique Wink ReisHydrogel sensor array making and structure. (A) 3D printing of the ionically conductive mesh for sensing. (B) Prototype ofthe soft sensing patch. (C) Data from a test device proving that it has spatial sensitivity, as well as to applied forcemagnitude and rate. (D) Soft biodegradable substrate.Key Accomplishments:The use of hydrogel medium poses several challenges suchas mechanical stability, fracture, creep and drying. Ourteam has thus far successfully increased the overallsensitivity and resiliency of biocompatible hydrogels andpublished the results in Nature Communications. We solvedmany of these problems by introducing a new class ofprintable touch sensors made completely out of hydrogelswith a hierarchical network microstructure. Onebreakthrough of this project is the formulation of newclasses of hydrogels with double network – colloidal andmolecular, which are synergistically reinforced while madeof the same material. These “homocomposite” hydrogelsserve as an excellent base material for 3D-printablepiezoconductive sensors operating on changes in ionicconductance. These sensors could be easily 3D printed andthe hierarchical microstructure of the ionic hydrogelsprovides superior tactile sensitivity as compared toconventional hydrogel. Our team developed a prototypeand demonstrated wireless interfacing/readout of thesensor patches and seek to use machine learning methodsfor multi-touch data processing.Impact:The outcome of this project is the demonstration of newtechnology-transformative sensing arrays and user-interfacing patches that are completely naturally sourced,made by using common 3D printing techniques andinterfaced to wireless data transmission modules. Thesepatches have three unique highly desirable features: theseare hypoallergenic and biodegradable, cost-effective, sothat they could be used in daily or periodically replaceableconsumables, and they could be made completelytransparent and attractive to the human users.Nature Communications.32

Textile Electromyography Sensors forWireless Stroke Therapeutic SystemObjective:The objective of this research is to improve the comfort anddata quality of garment-based electromyography (EMG)monitoring. This includes exploring the inevitable tradeoffsassociated with garment design and e-textile performance.For example, one area of research is in fabricating textilevias for connections between printed sensors andelectronic wiring.Approach:The arrangement of electrodes and their connectionmethod to a data aggregator are the key variables toconsider when trying to improve the signal to noise ratio(SNR) for an EMG garment. Our team uses screen-printed,Ag/AgCl dry electrodes that can be added to garmentsduring or after production which provides the opportunityfor modular and customized design approaches. Oursimulation efforts, utilizing CLO3D, inform the garmentdesign in order to customize garment fit to achieve an idealcontact pressure between the sensor and body, whichsignificantly improves data quality. Key Accomplishments:Our team has designed and fabricated two iterations of anarmband to monitor EMG signals. The first armband usesscreen-printed, Ag/AgCl dry electrodes and the secondarmband uses inkjet-printed Ag dry electrodes. We haveacquired high-fidelity EMG signals using each armband.Ongoing work is centered on improving the connectionmethod between printed interconnect and PCB. We willalso use CLO3D to design the armband so that it is bothstylish and functional.Impact:Biometric data quality of an e-textile smart garment isheavily reliant on the garment design and sensorintegration strategy and can still vary from user to user. Thisresearch explores and tunes the key factors in achieving ahigh performing smart garment for EMG monitoring, with aspecific end application of at-home rehabilitation andtelehealth. The outcomes of which can be applied to otheron-body sensing methods, rehab techniques, andtelehealth applications. Screen printed armband (left) and inkjet printed armband (right) where both armbands utilize crimped connections to externalwiring. The armbands also have adjustable sizing via velcro so that the appropriate contact pressure can be achieved. Principal Investigator:Amanda Mills, PhDFunding Source:NIDILRR: SBIR, Subcontract under Impulse WellnessStudents:Beomjun Ju, Prateeti Ugale33

Textile Integrated Sensors for Inner Prosthetic SocketEnvironment Monitoring for NeurorehabilitationObjective:This project aims to develop a novel Flexible InneR-socketSensing Technology (FIRST) that is seamless, unobtrusive,and elegantly integrated into the lower-limb prosthesissocket. FIRST is based on an electronic fabric structurewhere the fibers of the fabric act as sensory elements andcan simultaneously track tactile forces, moisture/wetness,electromyography, and body temperature at multiplesensing points around the residual limb. The majorchallenge is to develop a fundamental understanding ofthe coupling and interaction between multi-componentfiber cross-sectional architecture, fabric structure, and itselectro-mechanical response to achieve a multimodalsensor that can be unobtrusively integrated into 'textile-based' sensory devices in general. The interpretation of thedata is to identify locations of skin problems to enablepatient self-management and allow for more objectiveclinical evaluation to avoid the occurrence of potential skinbreakdown and the resulting complications.Approach:Our collaborative research team works on melt-extrudedmulti-component fiber and seamline-based sensordevelopment where we carefully engineer the fiber cross-section, fabric structure, and its electrical response. Thistargets a sensitive and specific multimodal response usingmicrofabricated and, ultimately, textile-based polymericfibers with ordered segments of conducting and insulatingareas in the fiber cross-sectional structure. We aim tounobtrusively integrate these into many electronic small- orlarge-area textile-based sensory devices and systems of thefuture especially for health monitoring. Key Accomplishments:We manufactured arrays of multi-component fiber andseam-line-based sensor patches and wearable, garment-integrated sensor arrays of expanded size connected to awireless high-speed data recording and transmission systemvia textile interconnects. We tested the sensors on an in-vitroartificial limb testing setup and two in vivo experimentsinvolving an able-bodied subject donning a bent-kneeadapter and a bilateral transtibial amputee participant. Inall these cases, the sensor array successfully detectedpressure changes within the inner socket during weight-shifting and walking experiments.Impact: Amputation is one of the major causes of disability. Socketsare the important prosthesis components and physicalinterface to integrate the prosthetic limbs mechanicallywith the amputee's residual limb to replace lost function.Objective monitoring of the inner socket environment (i.e.,pressure, temperature, and humidity) and residual muscleactivity during daily prosthesis use requires flexible,unobtrusive, and multi-modal sensors that can beintegrated into the socket structure without causing subjectdiscomfort. The lack of such an inner-socket sensortechnology has been a long-standing problem forevaluating the prosthesis socket, preventing thecomplications elicited by poor socket design and fit, andadvancing the socket technologies. Therefore, advancedsocket technologies are urgently needed and will bedeveloped under this project to significantly reduce thenumber of clinic visits, lower the healthcare costs foramputees, and ultimately improve their quality of life.Principal Investigators:Alper Bozkurt, PhD, Tushar Ghosh, PhD, Helen Huang, PhDFunding Source:NSFStudent:Brendan ThompsonSchematic depictions of the fiber and seam-line sensor arrays, images ofthe integrated textile sensors, and examples of integration and testing withhuman subjects, demonstrating successful detection of pressure changes.34

35Emerging CorrelatedEmerging CorrelatedSensing ApplicationsSensing Applications

A Wearable System for Continuous Monitoring andAssessment of Alzheimer's Disease and Related Dementia run time between charges when 1500 mAh batteries areused. The design process focused on lowering the samplingrate of the voice recording to preserve privacy and onlowering the power consumption to improve complianceby eliminating the need for frequent charging of thedevices. We have demonstrated the basic functionality ofthis system in controlled clinical environments and in theprocess of arranging more data collection. The next stagewill explore predictive artificial intelligence methods to usethe data collected and the indices and metrics generatedby the system for the assessment of speech impairment,gait decline, and cognitive stress for ADRD diagnosis andmanagement.Impact:Recent developments in wearables allow the recording ofsome of the related parameters such as voice or bodymovement. The higher power consumption of these devicesrequires a need for daily charging of the batteries whichmay cause compliance issues, especially for a userpopulation potentially being challenged by decreasedmemory function. Users of these devices may also find therecording of their speech as concerning to their privacyand confidentiality. Most of the available devices provideonly a limited number of sensors and are deployedgenerally only in the wrist region. The approach developedunder this project for early detection of ADRD providesnovel, cost-effective, user-friendly, wear-and-ignorewearable devices that can passively, longitudinally, andcontinuously generate data for simultaneous assessment ofspeech impairment, gait decline, and cognitive stressduring daily life activities. Objective:Early detection of cognitive decline is essential to study mildcognitive impairment and Alzheimer’s disease in order todevelop targeted interventions and prevent or stop theprogression of dementia. This requires continuous andlongitudinal assessment and tracking of the relatedphysiological and behavioral changes during daily life.Approach:Our research team’s earlier work focused on the effect ofaging on speech, gait, and cognitive analysis using datacollected during clinical tests. This project is our joint effort topresent a one-of-a-kind application-specific wearablesystem to move these assessments outside labenvironments. This custom-designed system recordsphysiological and behavioral signals related to speech,gait, and cognitive stress continuously and longitudinallyusing multiple sensors and two body locations (wrist andwaist). These include sound, inertial measurements, heartrate variability, pulse transit time, and electrodermalactivity. Our initial efforts have focused on the hardwaredesign processes, where the future work will look at thedata collected with these devices in real-life environments,and perform an artificial intelligence-based machine-learning analysis on the data for early detection ofAlzheimer’s Disease and Related Dementia (ADRD).Key Accomplishments:In the form factors of a wristband and waist patch, wedeveloped a multimodal, multi-sensor system that measuresinertial signals, sound, heart rate, electrodermal activity,and pulse transit time. Total power consumption of 2.6 mWwithout any duty cycling allows for more than 3 weeks of Principal Investigators:Alper Bozkurt, PhD, Veena Misra, PhD, James Dieffenderfer, PhD,Shevaun D. Neupert, PhD, Edgar Lobaton, PhD, Sujit K. Ghosh, PhD,Jason R. Franz, PhD, Katarina L. Haley, PhD, Adam Jacks, PhDStudents:Alec Brewer, Maxwell Noonan,Madeline Smith, Emily EichenlaubFunding Source:ASSIST Center(A) The application specific custom designedwearable physiological and behavioral monitoringsystem for early detection of ADRD consisting of a (B-C) waist-patch and (D) wristband. (Bottom) Systemblock diagram for both systems containing threesubsystems with the integrated circuits used in each(indicated in brown font). ECG sensor is only on thewaistband and the EDA on the wristband with all theother parts all common for both.Representative data collectedfrom the sensors during a walkingand talking experiment todemonstrate the functionality ofthe system.36

Artificial Intelligence Driven, Resilient andAdaptive Monitoring of Sleep (AI-DReAMS)Key Accomplishments:This recent project stems from an earlier clinical studyfunded by the National Institute of Health to explore the useof NIRS and machine learning to bring a new perspective tosleep studies. The team previously demonstrated flexibledevices to perform NIRS and electroencephalography inthe form factors of a flexible bandage. We thenconstructed a reconfigurable version of the hardwaresystem, validated novel conformable electrodes, anddeveloped the associated software platform for seamlessdata acquisition. The current efforts are focusing onconducting in-vivo studies to collect data from at-homeand in-clinic sleep studies and support the development ofthe proposed artificial intelligence techniques for studyingsleep more efficiently.Impact:A majority of the population in the US and around the worldsuffers from chronic sleep disorders that are not diagnosedor treated. There is a vital need for new wearabletechnologies to increase the capacity of sleep researchersto make further advances in investigating sleep,understanding sleep pathologies, and improving the abilityof clinicians to reliably detect and treat sleep disorders. Theresearch results from this award have the potential topositively influence the continuous monitoringinstrumentation required for other chronic conditions suchas heart diseases. In addition to allowing a novel, AI-driven,and reconfigurable tool design for sleep research, this effortwill also shed light on novel multimodal biomarkers assessednoninvasively in wearable form factors for the detection ofsleep stages and disorders.Objective:This project investigates the use of a data-drivenreconfigurable sleep monitoring system to transform sleepresearch in the clinic and at-home. A sensor fusion strategyto ultra-miniaturize the sleep assessment instruments andartificial intelligence (AI) techniques to explore novel sleep-related biomarkers have the transformative potential toinvigorate sleep research for more efficient and accuratediagnosis and treatment of sleep disorders. There is a needfor combining lower-cost hardware with better ergonomiccomfort, and more efficient data analysis to pave the wayfor rapid translation, adoption, and effective deployment ofsleep monitoring technologies.Approach:This project integrates two parallel efforts combininginnovations in hardware and data analytics: 1) enabling anadaptable and reconfigurable embedded system platformin the form factors of an adhesive patch, and 2) developingstate-of-the-art machine learning techniques incorporatingthe data-driven models necessary for improving theresilience of sleep diagnosis. The hardware system fusesmultimodal wearable sensors, combining near infraredspectroscopy (NIRS) and bioimpedance with othertraditional sensing modalities for sleep related biosignals, tocollect physiobehavioral data from multiple body locations.To ensure comfortable wearability during sleep, theadhesive patch combines novel conformable electrodeswith state-of-the-art flexible circuit integration techniques.The data analytics platform includes 1) signal processing toenable data-driven metrics for signal quality assessment fora given inference task, 2) inference models based ontransfer learning techniques and diverse datasets fordetection of sleep events and disorders, and 3) BayesianNeural Network supported sensor selection for improvingthe resilience and adaptability of sleep monitoring systems.Principal Investigators:Alper Bozkurt, PhD, James Dieffenderfer, PhD,Michael Daniele, PhD, Edgar Lobaton, PhD, Vladimir Pozdin, PhDFunding Source:NSFPostdoc/Students:Parvez Ahmmed, PhD, Alec Brewer, Kirstie Queener,Yuhan Chen, Reza Soleimani, Mauro C Victorio, KailaPeterson, Devon Martin, Ashley Dehn37Photonic sensor and circuit for sleep studies, in an adhesivebandage form factor with wireless recharging capabilitiesFabricatedconformableelectrodes withmicroperforationsWearable patch, including a bioimpedancewing board, a main board with amicrocontroller and accelerometer, and anelectrocardiography wing board. Theconformal electrodes are connected to thewearable patch for collection of physiologicalrecordings, including ECG and bioimpedance

Embedded Wearable System-Based Cough DetectionEnhanced by Out-of-Distribution RecognitionObjective:The objective of this project is to develop a wearableAutomatic Cough Detection Algorithm (ACDA) system thatmeets clinical monitoring requirements by fusing multimodalsensor data and ensures users’ privacy. The ACDA isdesigned to extract specific features from the datacollected by wearable devices and reliably pinpointcoughs. Furthermore, this algorithm is to be embedded in asmart device, enabling real-time cough detection.Approach:The ACDA developed is implemented using a convolutionalneural network (CNN) and enhanced by incorporating Out-Of-Distribution (OOD) Detection to recognize untrainedexternal environmental sounds (e.g., an engine sound). Inthe workflow, an acoustic signal is filtered on theembedded device and is recognized by ACDA in the smartdevice. Part of our cough detection efforts have also beenintegrated with the embedded hardware, in which we aretrying to minimize the amount of data to be transmittedand processed for power consumption and privacyconcerns. Furthermore, we are investigating methodologiesfor maximizing the performance of smaller models meantfor embedded systems (e.g. Type2DA Edge AI Module)using knowledge distillation, transfer learning, and relatedtechniques.Key Accomplishments:Our CNN-based ACDA achieves a sensitivity of 92.7%, aspecificity of 92.3%, and an accuracy of 92.5% using a lowsampling frequency of just 750 Hz which allows us topreserve patients' privacy by obfuscating their speechbased on our analysis. After incorporating OOD detection,the new algorithm produces trustful results when thesampling rate is greater than 750 Hz and the window size isbetween 4 - 10 seconds. At 750 Hz, the ACDA with OODdetection maintains over 80% accuracy with over 50% OODinput and outperforms the standard ACDA when the OODinput is just 15%. Using a kaiser window (size: 1024, length:256, hop length: 32) in preprocessing further boostsaccuracy by 2%. Using these pre-trained models also allowsus to achieve high performance on much smaller modelswith transfer learning techniques, achieving 85% accuracyon models with less than half of the parameters of theoriginal model.Impact:Traditional manual cough counting methods are costly andlabor-intensive. ACDA’s need for portability and extendedrecording has only recently allowed entry into non-clinicalspaces. The use of high sampling rates and high power andcomputing demands make wearable integrationchallenging. Our ACDA shows the capability of continuousmonitoring while maintaining privacy on portable devices. Principal Investigators:Edgar Lobaton, PhD, Alper Bozkurt, PhD, MichelleHernandez, PhD, James Dieffenderfer, PhDStudents:Yuhan Chen, Jeffrey Barahona, Pankaj Attri,Mayur Sanap, Maxwell NoonanFunding Source:NSFDetailed Pipeline of the ACDA with OOD DetectionOverview of a pipeline for processingon embedded device38

Funding Source:ASSIST CenterPrincipal Investigators:Edgar Lobaton, PhD, Alper Bozkurt, PhDStudents:Reza Soleimani, Geet KhatriLanguage Processing Using Cloud and EmbeddedArtificial Intelligence for Monitoring of Cognitive DeclineKey Accomplishments:The preliminary results are categorized into two groups: Incases where no augmentation is applied, our modelachieves an accuracy of 83.1% without contrastive learningand an accuracy of 85.8% when contrastive learning isincorporated on the “Cookie Theft” corpus within theDementiaBank datasets. This indicates that contrastivelearning contributes to better differentiation in therepresentation space. In instances where augmentation isemployed, the model attains an accuracy of 84.5% withoutcontrastive learning and an accuracy of 88.5% with theinclusion of contrastive learning. In comparison to thebaseline, the model's accuracy improves by a notable5.4%. The next stages include the investigation of theimpact of various augmentation techniques in order toautomatically determine the most effective augmentationapproach.Impact:The early identification of dementia is of utmost importancein slowing down cognitive decline and enhancing thequality of life for individuals affected by the condition. Ourmodeling efforts have yielded encouraging outcomes inthe diagnosis of dementia, relying solely on textual data.This has the potential to enable both healthcareprofessionals and patients to take proactive measures tomitigate cognitive decline at an early stage.Objective: The goal of this project is to create an Automatic SpeechMonitoring (ASM) system that can be worn as a wearabledevice for applications such as tracking of aphasia anddetection of patterns associated with Alzheimer’s Disease.The algorithms have the potential to be incorporated intoembedded systems or smart devices to enable real-timeand privacy-preserving monitoring.Approach:The initial stage of the system design is to set algorithmicgold standards of performance using Large LanguageModels (LLMs). These are used for comparison and toidentify key features as teacher models in knowledgedistillation frameworks. To obtain the textual information,various automatic speech recognition (ASR) methods havebeen used. These ASR employ transformer-based modelssuch as BERT and RoBERTa. The training objective for themodel consists of a classification component focused ondementia detection and employs a contrastive-based lossto enhance the distinction between various classes. Thisapproach embodies a multitask learning paradigm, whichhas demonstrated its effectiveness in improving modelaccuracy compared to training the model separately fordistinct tasks. Various forms of augmentation, includingusing synonyms and deletions, have been employed forhandling the limited availability of data and theimbalanced nature of medical datasets.Possible realization of thesystem in which some amountof the processing happens onan embedded device and therest is done at an AI-enablededge base station within theprivacy of the user’s home.39

Multifunctional Plant Wearable Sensors forAgricultural Monitoring and One HealthObjective:Plant diseases cause around 20-40% of global crop lossannually. Current disease diagnostics rely on a set ofmolecular technologies that are constrained in thelaboratory. Therefore, novel sensor technologies that cantrack the status of plant health in real-time and dissectvarious biotic/abiotic stresses are needed to detectpathogens early, prevent disease outbreak, and improveplant growth and yield. Recently, several plant wearableshave been demonstrated for continuous plant healthmonitoring. However, a multifunctional wearable sensorthat can track both biochemical (e.g., plant volatileorganic compounds) and biophysical (e.g., temperature,humidity, etc.) signals of the plant and/or surroundingenvironments with high sensitivity and specificity has rarelybeen demonstrated yet.Approach:In this project, we developed an abaxial leaf surface-attachable multimodal wearable plant sensor patch thatcan continuously and simultaneously measure leaf VOCs,leaf surface temperature/humidity, and environmentalhumidity, with high sensitivity and selectivity. The VOCsensors were composed of a hybrid network of ligand-functionalized gold-coated AgNWs (Au@AgNWs) and multi-walled carbon nanotubes (MWCNTs) embedded in ahydrophobic and nanoporous sol-gel layer. The ligand-functionalized Au@AgNWs interact with VOCs andmodulate the conductivity of MWCNTs, resulting indetectable resistive signal change from the sensor device.All sensors were patterned on a flexiblepolydimethylsiloxane (PDMS) substrate, and silver nanowire-based soft electrodes were used to connect all sensors onthe same patch. Principal Investigators:Qingshan Wei, PhD, Yong Zhu, PhD, JeanRistaino, PhD, Alper Bozkurt, PhDPostdoc/Students:Giwon Lee, PhD, Oindrila Hossain, Sina Jamalzadegan,Yuxuan Liu, Hongyu Wang, Yi ChenFunding Sources:GRIP4PSI, USDA, NSFMultifunctional plantwearable sensor deviceKey Accomplishments:This versatile platform was tested on live tomato plants inthe greenhouse for various stress monitoring applications,ranging from tracking plant water loss to early detection ofplant pathogens. A machine learning model was alsodeveloped to analyze multichannel sensor data forquantitative detection of tomato spotted wilt virus (TSWV)as early as four days after inoculation, which is even earlierthan conventional PCR tests. The model also evaluatesdifferent sensor combinations for early disease detectionand predicts minimally three sensors are required includingthe VOC sensors. These results have been recentlypublished in the peer-reviewed journal Science Advances.Impact:A multimodal plant wearable sensor patch capable ofdetecting both biochemical and biophysical parameters ofindividual plants, namely leafy VOCs, leaf surface humidity,leaf surface temperature, and environmental relativehumidity, was developed for continuous, on-plantphysiology monitoring. The biochemical sensing capability,such as VOC profiling, was added to the plant wearablesfor the first time to enable more precise disease and stressdetection in a noninvasive fashion. This sensor may findbroad applications in controlled environment agricultureand precision farming.Science Advances. Continuousmonitoring ofabiotic stresses(drought,overwatering,salinity, and nolight) of a livetomato plant.40

Figure 1.Wearablemetabolicmonitoringsystem withelectro-chemical andbiophotonicssensorFigure 4. Results ofhuman highintensity exercisetest showingrecorded tissueoxygenation,oxygenated tissuein red anddeoxygenatedtissue in blue.Multimodal Flexible Electrochemical and BiophotonicBiosensing System for Metabolic MonitoringKey Accomplishments:The platform’s sensing ability was validated againststandard benchtop equipment in laboratory settings. Theplatform's ability to continuously and in real-time monitorchanges in sweat lactate and pH during high-intensityexercise in humans has also been demonstrated throughInstitutional Review Board-approved protocols.Impact:This system allows for multimodal biosensor integration intoone wearable platform, allowing for a comprehensivephysiological picture of an individual. The tailorability of thesystem by interfacing the main module with different sensorsopens the possibility of sensing very different biomarkerswith the same hardware. The groundwork has also beenlaid in the current wearable format to be adapted into atranscutaneous format.Objective:Wearable technology has become readily adopted byconsumers but has been limited to measuring physiologicalsignals such as heart rate and motion. Being able to alsomeasure biologically relevant analytes from fluids such assweat, interstitial fluid, and blood could increase theknowledge of an individual's health and help inform relateddecisions. Therefore, our objective is to create a fullywearable multimodal biosensor platform thatsimultaneously monitors electrochemical and physiologicalsignals to provide a more comprehensive health picture.The current system measures sweat lactate, sweat pH, skintemperature, tissue oxygenation, and motion. Approach:The main module of the system is a custom, wearablepotentiostat that interfaces with a flexible electrochemicalsensor for monitoring sweat analytes and a biophotonicssensor for monitoring photoplethysmography (PPG) andtemperature, all of which are transmitted over bluetooth toan app. Moving forward, other sensors utilizing microneedlearrays to sample interstitial fluid will be used and a version ofthis system will be used with a sensor designed fortranscutaneous sensing, similar to a continuous glucosemonitor. Figure 2. Results of human high intensity exercise test showingrecorded sweat lactate and blood lactate readings.Figure 3. Results of human high-intensity exercisetest showing recorded sweat pH.Principal Investigators:Michael Daniele, PhD, Alper Bozkurt, PhD,Koji Sode, PhD, Spyridon Pavlidis, PhDFunding Sources:NSF, NBMC, SEMIStudents:Kaila Peterson, Brendan Thompson, Hannah Nissan, AngelicaAroche, Misk Hussain, Grace Maddocks, Ethan Cove41

Sensor-Integrated Microfluidic Chip to MonitorNutrient Uptake in Plants for Smart AgricultureObjective:Soil nutrient analysis is typically performed in laboratories,involving complex, time-consuming procedures. Microfluidicplatforms enable real-time monitoring of dynamicbiochemical changes, making it possible to understandplant phenotype. In conventional plant phenotypingexperiments, seeds and plants are grown in pots, fields, andagar dishes. Hence, researchers need a low-cost real-timeplatform to understand plant growth, metabolism, nutrientuptake, and survival. The high throughput of microfluidicplatforms enables them to deliver microliters of fluid tospecific outputs, making in-situ monitoring of minutespecimens possible.Approach:We present a low-cost, facile approach to fabricatingsensor-integrated microfluidic devices to study plant-nutrient interaction and dynamic nutrient uptake pathwaysusing molecularly imprinted polymer (MIP) -based electrodeprinting and polymer sheet layering. The MIP-basedtechnique offers more sensitive and accurate output. Thechip with MIP based sensor for nutrient analysis serves as aplatform technology for studying nitrate and phosphate.Incorporating the sensors in a microfluidic chip allows for thestudy of nutrient and plant microenvironments. This serves asa platform technology for a wide range of agriculture andsoil-based studies.Key Accomplishments:The printed sensors measured in-situ nitrate and phosphateconcentrations inside the growth medium as legume plantroots grew within the device. The selective sensorsdemonstrated high sensitivity and can continuously monitornutrients for approximately seven days. Overall, wedemonstrated that legume plants could be grown inuniquely fabricated microfluidic chips without hinderingtypical performance and growth. The electrochemicalnitrate and phosphate sensors, which encompass electricaland molecular-based systems, offer rapid detection ofnutrient changes.Impact:With the ability to integrate nutrient sensors, optimizedfertilization can be established, thereby maintaining asustainable agricultural environment. The microfluidicapproach provides a bridge for testing various plant-relatedphysiological and biochemical interactions beforeconducting field trials. Further, with the simplicity of thisdevice, it can be used by plant scientists for a wide rangeof real-time applications in the future, including root-pathogen interaction, drought-resistant plant selectionscreening, nutrient uptake efficiency, and monitoring thesoil microenvironment.Principal Investigators:Shekhar Bhansali, PhD, Vivek Kamat, PhDFunding Sources:NSF - PFITT and ASSIST CenterStudents:Vagheeswari Venkadesh, Lamar Burtona) Schematic illustration ofthe fabrication process; b)Illustration of germinationwells with integrated sensorsand PDMS sheets; c) View ofthe sensor-integratedmicrofluidic chip facilitatingthe growth of legume plant;d) An enlarged view oflegume roots.a) Legume plant shootand root growth in thedevice for 7 days. b)nutrient uptake trendby legume plant forapproximately 7 dayswith fresh nutrientsrecharged after dayfive (1ml nutrient MSmedia was added).42

Wearable Sensors-Based Health and ExposureTracker for Asthma and DiabetesObjective:The Health and Exposure Tracker (HET) system is a uniquehardware that has a modular structure to test and evaluatevarious ASSIST technologies (from electrodes and opticaldevices to ultra-low power electronics and sensors) to drivethe vision of ASSIST for correlated sensing of health andexposure. In addition to physiological sensors(electrocardiography, pulse oximetry, photoplethys-mography) and behavioral sensors (inertial measurementunits), this system also supports environmental sensors(ozone, VOC, ambient temperature, and relative humidity)and novel biochemical sensors (lactate, glucose, and pH)for a more comprehensive and correlated analysis. Noexisting system can provide the multi-sensing capability ofHET prototypes as well as an open platform where the next-generation sensing devices can be integrated.Approach:The HET-engineered system is composed of a modulararchitecture with various electrophysiological, biophotonic,inertial, potentiostatic, amperometric, and environmentalsensor front ends connected to a system-on-chipcombining a microcontroller with a Bluetooth Low Energytransceiver. This circuit architecture has been packaged inthe form factors of a wristband, chest patch, and a flexiblepatch that can be attached to body locations. Using HET,sweat is collected through a zero-power osmotic pumpingscheme connected to a screen-printed enzymatic sensor.Collecting sweat and analyzing the glucose and lactateconcentrations help assess metabolic state and supportdiet management for diabetic patients to provide auto-mated and actionable feedback. The wound monitoringsystem monitors uric acid levels to ultimately track healing.The modular structure of the HET-engineered systemenables it to be powered by inductively rechargeablelithium batteries or energy harvesting structures such asflexible solar cells or ASSIST’s thermoelectric generators.Key Accomplishments:HET systems have been tested in various clinical experimentsrelated to asthma exacerbation prediction, sweat analysis,and wound sensing. This past year the HET prototypes werebrought to a TRL 5/6 level while also lowering the powerconsumption to sub-milliwatt levels.Impact:The HET system strategically targets asthma as the medicalcondition of interest due to its high prevalence anddependence on environmental factors. HET brings theunique potential of continuously measuring local ozoneconcentration around the patient while also assessing heartrate, heart rate variability, respiratory rate, arterialoxygenation, and coughing frequency. This is combinedwith the next breakthrough in wearables with the analysis ofbiochemical markers such as sweat and wound fluidsanalysis for diabetic management. These capabilities couldallow medical professionals to track asthma exacerbations,diabetic metabolism, and wounds remotely and enableadvanced data analytics to generate automatedfeedback for the patient.Principal Investigators:Alper Bozkurt, PhD, James Dieffenderfer, PhD, Michael Daniele, PhD,Edgar Lobaton, PhD, Orlin Velev, PhD, Vladimir Pozdin, PhD,Michelle Hernandez, PhD, Michael Dickey, PhD, Veena Misra. PhD Students:Brendan Thompson, Kaila Peterson, AlecBrewer, Maxwell Noonan, Devon Martin,Yi Chen, Natalie SmitFunding Sources:ASSIST Center, NSF43

Amay Bandodkar15, 26, 32Jordan Besnoff26Shekhar Bhansali13, 42Ben Calhoun24Michael Daniele10, 14, 17, 21, 23,31, 37, 41, 43James Dieffenderfer20, 22, 36,37, 38, 43Michael Dickey8, 14, 43Wei Gao4, 6, 29Jason Franz21, 36Sujit Ghosh36Tushar Ghosh34Katarina Haley36Michelle Hernandez38, 43Lilian Hsiao32Helen Huang34Adam Jacks36Yaoyao Jia20, 25Xiaoning Jiang18Vivek Kamat13, 42Cassandra Kwon29Bongmook Lee4, 16Edgar Lobaton36-39, 43Amanda Mills6, 29, 33Shevaun Neupert36Ömer Oralkan25Spyros Pavlidis10, 41Brian Pietrosimone21Vladimir Pozdin17, 37,43Ram Rajagopalan7Clive Randall7David Ricketts26Jean Ristaino40David Roberts20Shad Roundy5, 9Nitin Sharma18Koji Sode10, 23, 41Trevor Tilly14Susan Trolier-McKinstry5, 9AbrahamVazquez-Guardado12Orlin Velev14, 32, 43Qingshan Wei40Douglas Werner28F. Yalcin Yamaner25Yong Zhu30, 40Veena Misra4, 6, 16, 29,36, 43ASSIST Co-DirectorsASSIST Co-DirectorsPrincipalPrincipalInvestigatorsInvestigatorsand Index of Featured Research Projectsand Index of Featured Research Projects Alper Bozkurt10, 14, 17, 20,22, 32, 34, 36-41,4344

45Sponsored research for exclusive relationships,priority IP access + negotiable pricing/IP termsCollaborations for external funding (e.g., DOE, DOD,DARPA, ARPA, NIH, SBIR)Endowments, joint publications+resource sharingDirecting core research efforts at ASSISTMultiplied research dollarsAccess to low-overhead “enhancement projects”Early exposure to cutting-edge researchAccess to world-class experts + medical practitioners Priority IP Access, including non-exclusive, internaluse licensesShared patenting costsFree workshops, training seminars+symposiaStudent engagement: mentorship, internships, seniordesign projectsFree ECE career fair accessDiscounted rates for access to facilities + equipmentMembersPartnersASSIST Membership BenefitsMembers and Partners Past and PresentContact us today!www.assistcenter.org@assist-centerLet’s explore opportunitiesfor collaboration.Ravi Chilukuri

assistcenter.org