Research

Advanced Modular Phantom Platform for Noninvasive Wearable and Medical Device Validation

Fully integrated modular phantom validation platform for repeatable physiological simulation and wearable device testing.

The Optical Bio-Sensing Laboratory (OBSL), in collaboration with the Center for Remote Health Technologies and Systems (CRHTS), has developed a fully modular in vitro validation platform designed to accelerate the development, characterization, and commercialization of next-generation wearable and medical sensing technologies. This advanced phantom-based system enables rapid, repeatable evaluation of noninvasive biosensing devices across multiple anatomical form factors, including wrist-based wearables, chest patches, ring sensors, optical probes, and surgical instrumentation interfaces. By combining physiologically accurate tissue phantoms, programmable hemodynamic simulation, and automated motion control, the platform provides a highly controlled environment for validating device performance under realistic physiological and biomechanical conditions.

Arm phantom test configuration enabling controlled evaluation of wrist-based wearable sensors under realistic biomechanical conditions.

The system incorporates mechanically and optically representative tissue structures, including layered skin models, vascular networks, and dynamically controlled pulsatile flow. These features allow researchers and industry partners to test sensing technologies across a range of physiological states, skin tones, tissue compositions, and motion conditions without reliance on animal or human subject testing. A key component of the platform is the finger phantom module, which replicates the structural and vascular properties of the human digit. This system enables high-fidelity evaluation of ring-based wearables and optical sensing technologies by simulating pulsatile blood pressure dynamics within anatomically representative digital arteries.

Finger phantom module designed for high-fidelity testing of ring-based and optical biosensing technologies.

The phantom’s multi-layer construction reproduces realistic tissue elasticity and optical scattering behavior, supporting validation of photoplethysmography, force-based sensing, and multimodal biosensing architectures. The modular architecture of the test system enables rapid integration of custom fixtures to evaluate new device geometries, including smartwatches, health bands, implantable interfaces, and emerging surgical technologies. Automated actuation and programmable physiological simulation enable repeatable stress testing across dynamic conditions such as motion, pressure variation, and vascular response. By providing a standardized, repeatable validation environment, this platform reduces development risk, shortens design cycles, and lowers costs associated with early-stage human testing. Industry collaborators can leverage the system for performance benchmarking, regulatory validation studies, algorithm development, and commercialization readiness assessments. This capability positions OBSL and CRHTS as a translational hub for wearable and biomedical device innovation, supporting partners from early prototype evaluation through final commercial validation.

AI-Enabled Ultra-Sensitive Cardiac Biomarker Detection Platform

This platform enables ultra-sensitive detection of cardiac troponin I, supporting early identification of myocardial injury and acute cardiovascular events. The technology integrates aptamer-engineered DNA nanostructures with programmable molecular signal amplification strategies, including hybridization chain reaction–based assembly, to generate highly detectable biosensing responses at clinically relevant low biomarker concentrations. These amplified molecular interactions produce multimodal outputs, including optical colorimetric signals and surface-enhanced Raman scattering signatures, enabling both rapid screening and precise quantitative analysis.

Patient-derived samples are processed through an integrated molecular assay workflow, and the resulting signal patterns are interpreted using machine-learning-driven analytics to enhance diagnostic accuracy, classify disease states, and support clinical decision pathways. By uniting molecular-level sensing precision, scalable assay architecture, and intelligent data interpretation, this platform establishes a next-generation diagnostic framework that bridges high-performance laboratory biomarker detection with accessible, real-time cardiovascular risk assessment and early intervention.

Force Mapping Glove for Quantifying Astronaut Hand Biomechanics During Extravehicular Activity

Integrated sensing layout showing distributed force sensors, signal routing, and compact control electronics enabling real-time biomechanical monitoring. Designed for applications in astronaut performance, rehabilitation, human–machine interfaces, and wearable motion analytics.

This research focuses on developing a noninvasive, wearable force-mapping glove to quantify human hand biomechanics during extravehicular activities (EVAs) and other demanding operational environments

nments. Astronauts operating in pressurized space suits must generate significantly higher forces due to glove stiffness, reduced tactile feedback, and altered mechanical loading conditions. These constraints increase fatigue, reduce dexterity, and elevate the risk of musculoskeletal injury. To address this challenge, we developed a soft, distributed-sensing glove integrating silicone-embedded force-sensitive resistors (FSRs) positioned across key anatomical regions of the hand to enable real-time spatial mapping of force distribution during complex tasks.

The sensing architecture is designed to preserve natural hand motion while providing quantitative insight into grip mechanics, torque generation, and localized strain patterns. To demonstrate translational readiness and real-world deployability, the sensing system was also integrated into a commercially available off-the-shelf lacrosse glove. This validation confirmed compatibility with existing protective glove architectures and highlighted the platform’s ability to be rapidly adapted to operational equipment without requiring custom suit fabrication. Such adaptability supports accelerated testing cycles for astronaut glove design, military protective gear, industrial exosystems, and high-performance athletic equipment.

Beyond spaceflight, this wearable sensing platform serves as a versatile human performance monitoring system applicable to military training environments, rehabilitation following neurological or orthopedic injury, industrial ergonomics optimization, robotics and human-machine interface development, and athletic performance enhancement. By enabling objective quantification of biomechanical interaction forces in constrained environments, this technology supports safer equipment design, improved training protocols, and enhanced human capability in extreme operational settings.

Integrated Multimodal Chest Patch for Continuous Cardiovascular and Hemodynamic Monitoring

Early multimodal chest-mounted cardiovascular monitoring prototypes demonstrating system evolution from a large, tethered round sensing platform (black) to a compact, fully wireless wearable patch (white). This transition reflects significant advancements in system integration, user comfort, real-world deployability, and overall technology readiness level (TRL), enabling continuous ambulatory physiological monitoring outside controlled laboratory environments.

This technology enables continuous, noninvasive cardiovascular monitoring through a compact, chest-mounted multimodal patch system. The platform integrates three complementary sensing modalities, photoplethysmography (PPG), electrocardiography (ECG), and strain-gauge-based mechanical sensing, to capture a comprehensive representation of cardiovascular dynamics.
The initial prototype consists of a tethered, round multimodal patch that incorporates all sensing components. This device simultaneously acquires electrical cardiac activity (ECG), optical blood volume changes (PPG), and local mechanical deformation using strain-gauge technology. Together, these signals provide synchronized insight into cardiac timing, vascular response, and pressure-related waveform morphology.
A second-generation device advances this design into a compact, fully untethered patch. This version emphasizes improved wearability, reduced form factor, and enhanced user comfort while maintaining multimodal sensing capability. Its wireless architecture enables greater mobility and supports real-world, ambulatory monitoring without the constraints of external wiring.
By combining electrical, optical, and mechanical measurements, the system enables robust characterization of cardiovascular function. The multimodal approach improves resilience to motion artifacts and signal noise by allowing cross-validation between sensing channels. This platform supports the development of advanced algorithms to estimate physiological parameters, including heart rate, blood pressure trends, and hemodynamic changes, under both static and dynamic conditions.

Multiplexed AI-Enabled Detection Platform for Early Identification of Gestational Diabetes

AI-enabled multiplexed vertical flow diagnostic platform integrating rapid biomarker assays, optical image acquisition, and machine learning analytics to transform multidimensional biochemical signals into clinically actionable health classifications for early disease detection and personalized risk assessment.

Gestational diabetes mellitus (GDM) is a multifactorial metabolic condition that significantly impacts both maternal and fetal health, often developing without clear early symptoms. Delayed diagnosis can lead to serious complications, including preeclampsia, fetal overgrowth, and increased lifetime risk of metabolic disease. Current screening approaches rely on single-analyte testing or oral glucose tolerance procedures that are time-intensive, uncomfortable for patients, and insufficient to capture the complex biological pathways underlying disease progression.

To address this unmet clinical need, the Optical Bio-Sensing Laboratory is developing a next-generation multiplexed diagnostic platform that combines molecular sensing, nanotechnology, and artificial intelligence to enable earlier and more precise detection of gestational diabetes. The system utilizes aptamer-based recognition of multiple GDM-associated biomarkers, integrated into a vertical flow assay (VFA) architecture, enabling simultaneous measurement of biochemical signals that reflect the multidimensional nature of disease physiology.

Advanced deep learning analytics are applied to interpret complex optical and electrochemical signals, enabling the identification of subtle biomarker interactions and predictive disease signatures that are not detectable by conventional screening methods. This approach supports earlier diagnosis, improved maternal–fetal risk stratification, and increased diagnostic confidence through enhanced sensitivity and specificity.

Designed for scalable deployment in both centralized laboratories and decentralized point-of-care environments, the platform requires minimal sample volume and supports rapid, user-friendly operation. By integrating multiplexed molecular detection with AI-driven interpretation, this technology represents a transformative step toward personalized prenatal diagnostics, enabling proactive clinical decision-making and improved pregnancy outcomes across diverse healthcare settings.

Next-Generation Non-Enzymatic Continuous Glucose Monitoring

This research focuses on the development of next-generation non-enzymatic continuous glucose monitoring technologies that enable stable, reversible, and real-time tracking of glucose concentration dynamics. The sensing platform utilizes fluorescence-based molecular recognition mechanisms driven by competitive binding interactions between glucose and engineered carbohydrate ligands within a Concanavalin A–based sensing architecture. These interactions produce glucose-dependent modulation of fluorescence signal behavior, enabling quantitative monitoring across physiologically relevant concentration ranges.

By eliminating reliance on enzymatic reactions, this approach reduces signal drift, enhances long-term sensing stability, and improves measurement reliability compared to conventional enzyme-based continuous glucose monitoring systems. The sensing chemistry is designed for integration into minimally invasive implantable or wearable device formats, supporting continuous physiological monitoring with strong biochemical specificity and rapid responsiveness to metabolic fluctuations.

Advanced optical readout strategies combined with signal processing and machine-learning-enabled analytics allow robust interpretation of dynamic glucose trends in ambulatory environments. This technology establishes a scalable framework for durable, high-accuracy metabolic monitoring platforms intended for next-generation digital health applications, chronic disease management, and precision medicine.

Noninvasive Multimodal Biosensing Armband for Predictive Health Monitoring and Translational Wearable Development

Noninvasive multimodal wearable biomonitor engineered for continuous physiological monitoring and translational validation. The adjustable arm-mounted platform integrates advanced biosensing modules to enable real-time assessment of cardiovascular, metabolic, and hemodynamic function while serving as a scalable test system for the development and deployment of next-generation wearable health technologies.

This research focuses on developing a fully noninvasive, multimodal wearable biosensing platform that enables continuous, real-world physiological monitoring and predictive health intelligence. The system integrates a comprehensive suite of sensing modalities, including photoplethysmography, bioimpedance, electrocardiography, electrodermal activity, temperature sensing, inertial motion tracking, and mechanical physiological sensing within a compact and ergonomic upper-arm wearable architecture. By simultaneously capturing electrical, optical, mechanical, and autonomic signals, the platform provides a synchronized, high-resolution representation of cardiovascular, metabolic, and systemic dynamics without invasive measurements.

A major focus of this platform is the early identification and prediction of metabolic instability, including both hyperglycemic and hypoglycemic events. Rather than relying on a single biochemical marker, the system leverages multimodal physiological signatures that reflect vascular responses, autonomic nervous system activation, perfusion changes, and metabolic stress. Advanced machine learning and signal fusion algorithms analyze these complex physiological interactions to enable predictive detection of glycemic excursions before severe clinical manifestations occur, supporting next-generation approaches to metabolic health monitoring.

Integrated multimodal sensing architecture deployed on the upper arm, combining single-sided ECG, bioimpedance and electrodermal activity, optical perfusion sensing, temperature monitoring, and inertial motion tracking to enable comprehensive, noninvasive characterization of cardiovascular performance, metabolic state, and dynamic physiological response.

Beyond continuous monitoring, the platform serves as a flexible translational test system for the rapid development and validation of emerging wearable biosensing technologies, multimodal signal fusion strategies, and AI-driven physiological modeling. Its modular design enables integration of new sensing concepts and supports applications spanning diabetes management, cardiovascular risk assessment, human performance optimization, rehabilitation monitoring, and autonomous health systems for clinical, consumer, military, and aerospace environments.

Designed for scalability, comfort, and real-world deployment, this noninvasive biosensing ecosystem represents a transition from episodic measurement toward predictive, personalized physiological intelligence, enabling earlier intervention, improved situational awareness, and enhanced long-term health outcomes.

Soft Collapsible EEG Electrode for Rapid Neural Monitoring and Deployment

Soft, conformal electrode platforms developed for high-fidelity neural signal acquisition, shown alongside anatomically representative head phantom systems used for controlled validation testing and device optimization.

This research focuses on developing soft, wearable neural-monitoring technology for mobile subjects. Rigid electrodes made of metal or plastic polymers are commonly used for EEG monitoring but pose injury risks to the user during falls or collisions. High-energy impacts at or near the electrode-skin interface can cause bruises, puncture wounds, or abrasions as electrodes jab into or drag across the skin. These safety concerns prevent the use of rigid electrodes in wearable devices for long-term neural monitoring in mobile subjects, thereby limiting the ability to collect high-quality longitudinal data for diagnostic or health-monitoring purposes.

We have developed a novel soft electrode that deforms as pressure increases, collapses if a pressure limit is exceeded, and returns to its original configuration as pressure returns to normal, all while maintaining electrical conductivity and scalp contact. The electrode body and legs are made of non-toxic soft silicone polymer. Electrode tips are made of silver/silver chloride, carbon-doped soft silicone polymer, or conductive ink. As pressure is placed on the top of the electrode, the legs deform while the tips remain in contact with the scalp, allowing us to continue recording neural activity. To test these electrodes, we use a head phantom made of ballistic gelatin that simulates the electromechanical properties of the human head. Using soft electrodes for biopotential measurement protects the patient in the event of impact while still maintaining electrode-skin contact, allowing us to apply these electrodes to long-term health monitoring scenarios.

There is a wide range of commercial applications for this technology. Patients living with neurological motor diseases or injuries have increased fall risk, but clinicians still need tools to evaluate their rehabilitation progress. Sending patients home with soft wearable technology allows us to collect long-term health data without risking injury to the subject, thereby better informing therapeutic interventions. Military applications include establishing baseline neural function at the start of training, monitoring competency and stress during training, and monitoring on the battlefield to reduce the risk of physical injury and mental disorders. Athletic performance can be enhanced during practice and game day by monitoring athletes’ stress levels and adapting coping strategies to keep them in peak condition.