This conference presentation was prepared for SPIE BiOS, 2024.

Performance of paper-based rapid tests is limited by their poor limit of detection and binary response. We have developed a low-cost end-user device that overcomes the classic limitations of rapid tests. The Arduino-based hand-held device excites rapid test gold nanoparticles with modulated 532nm LED while recording their thermal wave response with low-cost thermometer ICs. Validation studies on COVID-19 antibodies and THC (psychoactive constituent of cannabis) in saliva demonstrates ability of the innovation in quantifying analyte concentrations and enhancing the limit of detection by over an order of magnitude.

We present a rapid, portable optical system for label-free detection of COVID-19. Raman spectra from an entire liquid drop of saliva supernatant can be obtained within 6 minutes, and the sample is classified as COVID-19 positive or negative using artificial intelligence (AI). 293 COVID negative and 49 COVID positive saliva supernatant samples were analyzed. Positive samples were from hospitalized patients (non-critical and critical) and non-hospitalized testing clinic volunteers (symptomatic and asymptomatic). Our Raman/AI system has an 82% accuracy detecting people with COVID-19 of any severity with any symptom presentation, and 89% accuracy when detecting COVID-19 in hospitalized patients alone. Rapid label-free analysis of biofluids for viruses could provide a low-cost screening solution that could be adapted to respond to viral mutations. This could be invaluable for future pandemics and for reducing infections in hospitals, care homes and workplaces.

Healthcare disparities revealed during the COVID-19 pandemic must be addressed. A simple way to enable care to reach remote or low-resource communities is through inexpensive, rapid, sensitive, and portable biosensors. Here, we discuss work on the development and testing of a lens-free holographic microscopy (LFHM) and agglutination assay-based biosensor capable of rapid and sensitive point-of-care detection of SARS-CoV-2 virus. Additionally, we discuss recent efforts to further improve this sensor by using colloidal gold nanorods and machine learning approaches.

Current diagnostic tumor biopsies are invasive and can disrupt and spread the tumor. Liquid biopsies are inadequate for early-stage detection, resulting in lower survival rates and poorer prognoses. Raman spectroscopy can detect many cancers by identifying subtle cancer-associated metabolites in circulating biofluids. This study investigates preanalytical variables affecting Raman biofluid measurements in head and neck cancer patients, namely spatially correlated changes caused by Marangoni and capillary flow, aiming to streamline testing by reducing sample acquisition time and human intervention. This method is fully automated, providing a high-throughput assay for large-scale screening, paving the way for widely available, sensitive cancer detection.

Measuring blood flow on the head is an indirect yet crucial way to access or quantify brain activity non-invasively, useful for diagnosing traumatic brain injuries by looking at the changes of blood flow. However, measuring the cerebral blood flow (CBF) remains challenging, principally due to the ability to reach and collect signal from the brain. In this talk, I will present our efforts in building an affordable, compact, and portable laser device for real-time Cerebral Blood Flow (CBF) monitoring. Our technique draws upon the speckle visibility spectroscopy technique, using the spatial ensemble of the speckle field to evaluate blood flow dynamics.

A compact, low-cost (<$150), wearable device that continuously monitors peripheral perfusion via laser speckle flow index (LSFI) was used to measure changes in blood flow in human subjects during a set of physiological challenges. One group (A) of subjects underwent arm occlusion followed by exercise (n=3) and a second group (B) of subjects underwent arm occlusion, cold stimulation, and exercise (n=4). Cold stimulation alone did not cause significant changes from baseline LSFI. The rate of increase of the LSFI signal over the 5-minute exercise period was 3 times larger in group A compared to group B. Although cold stimulation alone did not appear to cause vasoconstriction in the wrist, it did impact the rate at which perfusion increases during exercise. The low-cost device consistently monitored changes in hemodynamics in all subjects, including both increases and decreases in blood flow caused by occlusion and sympathetic nervous system stimulation.

Speckle contrast optical spectroscopy (SCOS) allows for simultaneous monitoring of blood flow and volume changes within each cardiac pulse. SCOS measurements were collected from 10 subjects at two time points and the blood flow and volume pulse waveforms (PWFs) were extracted. The eXtreme Gradient Boosting model was trained on an individual subject’s first measurement and predicted BP for that subject’s second measurement. A model trained on features extracted from both flow and volume PWFs was compared to a model trained only on features extracted from blood volume PWFs. With the addition of blood flow information, BP estimation was significantly improved.

Recently we developed the open-source FlexNIRS: a battery-operated, wireless, wearable oximeter whose self-calibrating geometry allows measurements of oxygen saturation in tissue. The first implementation of the device operating at 100 Hz has been validated and is enrolled in several measurement campaigns across different research laboratories. A recent firmware upgrade provides 266 Hz sampling rate, and hardware modifications provide improved form factor, wearability, and multi-modal acquisition. The new version is currently adopted in multiple clinical measurement campaigns focusing on pulsatile component analysis. We will present the instrument performance, its recent and future upgrades, and the applications where the device is currently in use.

To combat the unmet need of early detection and monitoring of postpartum hemorrhage (PPH), the leading cause of maternal death, a wearable short-wave infrared (SWIR) photoplethysmography (PPG) sensor was developed for real-time detection and monitoring of hemodilution, a compensatory response to PPH. The device uses wavelengths at 900 and 1300nm, ideal for monitoring hemoglobin and water contributions in blood, respectively. These wavelengths minimally absorb melanin, which has been shown to bias PPG results in pulse oximetry. High signal-to-noise PPG signals have been acquired from both the 900nm and 1300nm channels in human volunteers. Further, the hemodilution sensing capabilities were tested by flowing sheep blood through a pulsatile flow phantom. The ratio of the peak amplitudes of the PPG waveforms from both LEDs showed a linear correlation, demonstrating feasibility to detect hemodilution and provide a low-cost, fully wireless, continuous monitoring tool for early diagnosis and monitoring of PPH.

High precision polarimetry typically employs polarization modulation to enhance sensitivity. Common modulators include electro-optic, acousto-optic, and magneto-optic types. Faraday cells have proven successful as high-performance modulators for measuring glucose concentrations. However, these modulators require high voltage or high current, hindering wearable or implantable device implementation. Liquid crystals (LCs) were suggested as alternatives but introduce impurities and have limited functionality for broadband light required in glucose measurement. This study proposes digital polarization modulation using LC as shutters to detect glucose in physiological concentrations. The absence of power-consuming elements allows for potential miniaturization as a wearable or implantable device.

The accurate measurement of blood hematocrit levels in ocular vessels holds significant clinical value by saving time, minimizing discomfort, and reducing the use of single-use cuvettes, which generates significant biohazardous waste. We present a novel approach for non-invasive measurement of blood hematocrit levels in retinal vasculature using a modified full-field fundus imaging system centered around two isosbestic wavelengths. To aid in the selection of bandpass filters, we developed a detailed Monte-Carlo simulation of our imaging system applied to retinal vasculature. We also present this model as a tool to analyze the images coming from the Fundus camera, which helps us understand what tissue and system parameters affect our in-vivo measurements. We present the analysis methods employed for data from a clinical trial data and demonstrate the trends between the optical densities of the collected images and our Monte Carlo simulations.

Chronic-pain (CP) affects at least 10% of the world’s population, and costs our society $1T. There is currently no effective approach to treating CP. To this end, there is an urgent need in the pain-research community to enable monitoring pain-related biomarkers, e.g. circulating DNAs, cytokines, endocannabinoids, over a long-period of time (> 3 months). Here, we show how implant incorporated with photonic nanomaterials known as upconversion nanoparticles (UCNP) are particularly suited for such purposes. We will argue that the unique optical properties of UCNP make possible a simpler wearable readout electronics with low power consumption, thus greatly prolonging operation time.

Pulse oximeters overestimate the oxygen saturation in people with higher melanin concentration limiting their ability to get care or coverage. Our research has identified the sources of error to be the relationship between oxygen saturation and red-to-infrared ratio and insufficient information about the contribution of skin pigmentation to the measurements. Our design addresses these sources of errors. Due to the racial inequities in data from existing clinical trials, we will initially use precisely designed optical phantoms, followed by in vitro tests. In this talk, we will be presenting the design, the challenges, and our approach to addressing them.

Continuous monitoring of pH and pressure is a common procedure followed by physicians to acquire a clinical picture in gastroesophageal reflux diseases but also the presence of the bile in the gastroesophageal apparatus has been demonstrated to provide important information associated to the damaging effects of the biliary salts on the esophageal mucosa. Results and performances of optical fiber sensors for the simultaneous measurement of bile and pH with the optical fibers will be described together with all the necessary steps followed to arrive to clinical applications. The activity related to the integration of the bile and pH sensors in a single catheter and to the development of a single catheter for the simultaneous measurement of all the three important parameters (pH, bile and pressure) will be fully discussed.