Research Portfolio

Dr. Dickerson’s research focuses on developing scalable, automated methods to extract clinically meaningful variables from electronic health records using large language models (LLMs). This work addresses a fundamental bottleneck in clinical research: the reliance on labor‑intensive manual chart review to create usable datasets.

Beginning with a deeply annotated cohort of patients with complex breast cancer care, his group has developed hybrid embedding‑ and LLM‑based extraction pipelines capable of abstracting important variables for cancer research, including dates of diagnosis, treatment initiation, and discontinuation, and death; the specific anti‑cancer therapies administered; the clinical rationale for starting and stopping treatments; and tumor genomic characteristics. The current pipeline outperforms trained research coordinators in abstracting all of these variables. 

Current pipelines achieve approximately 90% precision and recall, with an explicit lab goal of reaching 97% performance across most variables. Ongoing work evaluates whether this approach can generalize with similarly high accuracy across diverse clinical domains and document types. If successful, this research will enable large‑scale, automated abstraction of high‑fidelity clinical data, transforming unstructured patient charts into research‑ready datasets and substantially expanding the feasibility and scope of observational, health services, and translational research across medicine.

PI: James Dickerson, Clinical Assistant Professor, Medicine - Oncology

The PROACT (Preventive Risk Outreach And Cascade Testing) Program, funded through a $3.5 million gift from Bright Pink, aims to revolutionize the approach to cascade genetic testing within families at risk for cancer. Cascade testing is essential for identifying individuals who may be at risk of developing cancer, but studies show that many people with a family history of cancer risk do not undergo genetic testing, often because family members don’t communicate their risk or due to logistical and insurance barriers. PROACT will address these challenges by providing a user-friendly artificial intelligence-powered online platform where individuals who have tested positive for cancer-related genetic variants can invite relatives to learn about their own cancer risk. The PROACT Program is led by Drs Allison Kurian and Jennifer Caswell-Jin of Stanford, in collaboration with Drs Steven Katz and Larry An of the University of Michigan. The Center for Digital Health’s pilot award was instrumental in laying the groundwork for PROACT and the major gift that funded it. In particular, it enabled the team to establish the necessary infrastructure, through collaboration with the Stanford Center for Clinical Research, to initiate pilot testing of a digital tool for cascade testing in Stanford’s cancer genetics clinic. Please see proactprogram.org for more information.

PIs: Allison Kurian, Professor of Medicine (Oncology) and of Epidemiology and Population Health & Jennifer Caswell-Jin, Assistant Professor of Medicine (Oncology)

Quantitative Digitography (QDG) technology and its AI-enhanced algorithm transform care by measuring validated metrics of all PD motor symptoms in real time, assisting with medication adherence, and optimizing individualized treatment plans. QDG metrics are validated with the current clinical standard rating scale, the MDS-UPDRS III score, and all sub scores. The technology can differentiate off and on medication, off and on DBS, and detects people with Parkinson’s (PWP’s) responses to adjustments of therapy. QDG enables Health Care Providers (HCPs) to see daily patient data and adjust medications in real-time and drastically reduces the disparities that PWP face every day in the current standard of care. The major goals of this project are to develop user and HCP training modules, so that the data coming into the web service from remote use of QDG monitoring is reliable and the HCP can understand the outputs in order to make therapeutic decisions.

PI: Helen Bronte-Stewart, John E. Cahill Family Professor, Professor of Neurology and Neurological Sciences (Adult Neurology) and, by courtesy, of Neurosurgery

A biosensor capable of continuously measuring specific molecules in vivo would provide a valuable window into patients’ health status and their response to therapeutics. Unfortunately, continuous, real-time molecular measurement is currently limited to a handful of analytes (i.e. glucose and oxygen) and these sensors cannot be generalized to measure other analytes. In this work, we will describe a biosensor technology that can be generalized to measure a wide range of biomolecules in living subjects. To achieve this, we develop novel reagents (molecular switches) that change its structure upon binding to its target analyte and emit light or produce an electrochemical signal. Our real-time biosensor requires no exogenous reagents and can be readily reconfigured to measure different target analytes by exchanging the molecular switches in a modular manner. Importantly, we will discuss methods for generating the molecular switches which are at the heart of this biosensor technology.

PI: Hyongsok Tom Soh, W. M. Keck Foundation Professor of Electrical Engineering, Professor of Radiology (Diagnostic Sciences Laboratory) of Bioengineering and, by courtesy, of Chemical Engineering

Shuno is a digital hearing health project that (1) enables parents to monitor their children's hearing health daily at home for free using the digital Ling-6 Sound Test, and (2) uses machine learning approaches to reduce the time needed to achieve accurate insights from the Ling-6 Sound Test. Led by a team that previously reduced vision test prediction errors by 74% with the Stanford Acuity Test, Shuno addresses critical gaps in pediatric hearing monitoring caused by limited specialist access and inconsistent at-home testing. By making professional-grade hearing assessment accessible to families, Shuno aims to ensure early detection of changes and better outcomes for the thousands of children with cochlear implants.

PI: Christopher Piech, Associate Professor (Teaching) of Computer Science

Real-time melatonin measurement allows access to individual circadian rhythms, which is critical for personalized medicine and preventing neurodegeneration in the aging population. Current lab-based melatonin measurement methods often result in a time lag between sample collection and assessment, posing challenges for timely melatonin-based treatments. The objective of our project is to bring together expertise in sensors, materials, wireless communication, and physiological studies to develop a real-time melatonin sensor operable at home, exploring the clinical significance of such a sensor and providing a digital solution for health monitoring in the aging population.

PIs: Ada Poon, Associate Professor of Electrical Engineering & Makoto Kawai, Clinical Professor, Psychiatry and Behavioral Sciences - Sleep Medicine

We have been developing genomic foundation models capable of zero-shot prediction of variant effects across coding and noncoding regions of the human genome. We first aim to specialize genomic foundation models for problems in human genetics and genomics, including fine-tuning on population-scale variation data, integrating functional genomics annotations, and incorporating information from clinical variant databases. These adaptations will enable improved variant effect prediction for rare disease diagnosis and polygenic risk scoring, prioritization of causal variants in association studies, and interpretation of noncoding mutations in regulatory regions. Beyond variant interpretation, we are developing models with generative capabilities that open new avenues for therapeutic design: we plan to engineer synthetic elements with programmable cell-type-specific activity and compose more complex genetic circuits, such as logic-gated gene expression systems or multi-input biosensors, for precision gene therapies that respond to disease-relevant cellular states.

PI: Brian Hie, Assistant Professor of Chemical Engineering

A compelling body of evidence strongly suggests that shingles vaccination prevents or delays dementia. The Geldsetzer group has used natural experiments that take advantage of the date of birth-based eligibility rules for shingles vaccination across various countries, allowing for a comparison of individuals who differ in ages by just a few weeks but have a large difference in their probability of ever receiving the shingles vaccination. This approach avoids the common bias concern in observational studies that those who decide to get vaccinated differ in their health behaviors and motivation from those who opt against vaccination. Published in leading journals, including Nature, JAMA, and Cell, the Geldsetzer group has found large protective effects for dementia in a host of different countries and for a series of dementia-related outcomes (mild cognitive impairment, dementia diagnoses, and deaths due to dementia). It may well be the case that shingles vaccination does not only have a protective effect for dementia but contributes to neuroimmune health in older age more generally. As a next step in research, the Geldsetzer group seeks to conduct a large-scale clinical trial using the old, off-patent, live-attenuated shingles vaccine to conclusively test the effect of shingles vaccination for reducing age-related cognitive decline and preventing dementia. He is currently seeking funding from foundations and philanthropy for this ambitious project.

PI: Pascal Geldsetzer, Assistant Professor of Medicine (Primary Care and Population Health) and, by courtesy, of Epidemiology and Population Health

Heart rhythm abnormalities are common and a leading cause of sudden death in children. Early detection and characterization of arrhythmias are crucial for developing effective treatments. Wearable technologies have become key in modern healthcare and have demonstrated success in detecting arrhythmias in adults. However, there is limited data on their utility for children. This study aims to assess the accuracy of Apple Watch tracings in children, evaluate whether the Apple Watch detects more clinically significant arrhythmias than standard heart monitors, and develop a user- friendly smartphone application tailored to children. The app would consolidate arrhythmia data and allow clinicians real-time access through a web-based interface. The goal is to make this innovative, pediatric-specific tool universally accessible, enhancing arrhythmia detection for children worldwide.

PIs: Scott Ceresnak, Professor of Pediatrics (Cardiology) & Henry Chubb, Clinical Associate Professor, Pediatrics - Cardiology

We will integrate our crowd- and AI-powered ASD detection mechanism into GuessWhat, a gamified mobile platform developed by our lab. This will provide families around the world with a cheap, fast, and reliable tool to identify ASD (and, eventually, other pediatric mental health conditions) and receive the standard of care. To this end, with the present proposal we will 1) use GuessWhat to obtain a large library of videos of Bangladeshi children at risk for ASD, 2) recruit a local crowd workforce to label the resulting data and process it through our enhanced AI models, and 3) Train advanced AI for automatic feature extraction that speeds detection and expands reach to a global population of children with risk for DD while reducing bias.

PIs: Dennis Wall, Professor of Pediatrics (Clinical Informatics), of Biomedical Data Science and, by courtesy, of Psychiatry and Behavioral Sciences & Gary Darmstadt, Professor (Teaching) of Pediatrics (Neonatology) and, by courtesy, of Obstetrics and Gynecology

Excessive smartphone usage among adolescents is linked to negative outcomes such as poor sleep quality, increased anxiety, and decreased academic performance. By partnering with the one sec application, which has over 2.5 million users, the study aims to investigate digital interventions and correlate them with measurable health outcomes such as sleep quality, physical activity, and mental well-being. This pilot study will not only measure the impact of these interventions on health metrics but also identify the most effective types, providing essential data to support future research and funding applications aimed at improving adolescent health and well-being.

PIs: Paul Schmiedmayer, Lead Artificial Intelligence, Assistant Director of Digital Health, & Research Engineer, School of Medicine - MDRP'S - Biodesign Program & Emily Fox, Professor of Statistics and of Computer Science

Atrial fibrillation is the most common heart rhythm disorder and increases the risk of dementia and stroke, which can decrease quality of life and patient longevity in our aging population. While anticoagulation medications can reduce the risk of these complications, nonprescription rates of this medication can be as high as 50%, and traditional techniques to identify actionable reasons for reducing this disparity are expensive to undertake and difficult to scale. In this proposal, we aim to use proprietary and open-source large language models to identify reasons for anticoagulation nonprescription from clinical notes at Stanford and other geographically diverse health systems, understand differences by patient sociodemographic characteristics, and extract actionable insights at the health system and public health level to reduce this disparity and improve healthy aging.

PIs: Sulaiman Somani, Fellow in Medicine, Cardiovascular Medicine & Tina Hernandez-Boussard, Professor of Medicine (Biomedical Informatics), of Biomedical Data Science, of Surgery and, by courtesy, of Epidemiology and Population Health

The My Heart Counts (MHC) smartphone app has broad reach with >100K users and continues to publish high-impact digital health research. We propose to integrate a pre-trained large language model to automatically generate coaching interventions aimed at increasing physical activity, which we have demonstrated to be equally efficacious to expert-curated coaching prompts in our preliminary data, in order to revolutionize the MHC app to serve as an autonomous digital health platform. With the integration of the trained large language model, we will perform numerous fully-digital clinical trials of physical activity in different disease contexts, with the aim of providing an evidence base of digital health interventions in improving human health and longevity.

PIs: Euan Ashley, Roger and Joelle Burnell Professor of Genomics and Precision Health, Arthur L. Bloomfield Professor of Medicine and Professor of Genetics, of Biomedical Data Science and, by courtesy, of Pathology & Daniel Kim, Assistant Professor of Medicine (University of Washington) 

Atherosclerosis of vessels is commonly implicated with poor cardiovascular outcomes. Conventionally, atherosclerosis is only quantified in the coronary arteries, which has led to clear evidence that early diagnostics can lead to beneficial therapeutic or lifestyle interventions. However, atherosclerosis of other large vessels, such as the aorta, which is commonly visualized in 20+ million routine abdominal computed tomography (CT) scan, is less well studied. In this project, we propose developing an automated pipeline to opportunistically measure aortic atherosclerosis in abdominal CT scans for use as a novel biomarker to predict future major cardiovascular events.

PI: Akshay Chaudhari, Assistant Professor (Research) of Radiology (Integrative Biomedical Imaging Informatics at Stanford) and of Biomedical Data Science

We study the use of modern AI-based speech recognition to automatically score human speech recognition tests and identify patients with hearing loss who are at risk for cognitive impairment. To be more specific, we apply a machine-learning model to the additional data obtained by automatic scoring to identify meaningful information for characterizing patient performance, and detect those at risk for disorders like cognitive impairment. This proposal extends conventional speech in noise measures to conversational speech, allowing us to characterize the effect of noise and cognitive impairments on a subject’s ability to enjoy the noisy world we live in.

PIs: Matthew Fitzgerald, Associate Professor of Otolaryngology - Head & Neck Surgery (OHNS) & Malcolm Slaney, Adjunct Professor, Music

Contrary to popular opinion, most homeless persons own a Smartphone or tablet and are amenable to using these devices to receive a range of care services. Despite this, we know little in the homeless population about (1) which individuals engage with digital health resources, and (2) the potential for device data in this population to facilitate real-time monitoring of health status. The current proposal will seek to answer these questions, with a broader goal of establishing a research program on digital phenotyping in homeless persons.

PIs: Daniel Blonigen, Associate Professor of Psychiatry and Behavioral Sciences (Public Mental Health and Population Sciences) & Donna Zulman, Professor of Medicine (Primary Care and Population Health)

In order to prevent central line associated blood stream infections (CLABSIs), current practice is for nurses to visually inspect and document the central line insertion site once per shift, a time-intensive and subjective process which could potentially be improved using artificial intelligence (AI), saving time for other direct patient care activities. Our collaborative team previously trained a Medical Foundational Model (MedFM) with an image-text understanding using 4.5 million image-text pairs, with a general understanding of medical images and terminology surpassing human expert-level performance in various medical disciplines, including infectious diseases and dermatology. Building on these successes, we aim to leverage MedFM’s advanced capabilities and explore its performance for surveillance and prevention of catheter-related bloodstream infections evaluating the catheter insertion site.

PIs: Jorge Salinas, Assistant Professor of Medicine (Infectious Diseases) & Olivier Gevaert, Associate Professor of Medicine (Biomedical Informatics) and of Biomedical Data Science

This project aims to assess the impact of ambient AI scribe technology on patient care equity and privacy at Stanford Health Care, focusing on diverse patient groups and exploring patient perspectives. By employing a combination of quantitative analyses and qualitative methods, the study will identify potential disparities in the use of the AI tool and gather insights into patient autonomy and preferences. The ultimate goal is to inform the responsible implementation of AI-assisted documentation in healthcare settings, ensuring technologies like these enhance patient care across diverse patient groups.

PIs: Shreya Shah, Clinical Associate Professor, Medicine - Primary Care and Population Health & Danton Char, Associate Professor of Anesthesiology, Perioperative and Pain Medicine (Pediatric)

Lung cancer is the leading cause of cancer death in the U.S. Current treatments such as targeted therapy and immunotherapy benefit only a subset of patients, and existing biomarkers do not predict an individual’s response accurately. This project aims to refine a multimodal vision-language foundation AI model trained on millions of pathology images and texts to better predict which patients will respond to these therapies. By integrating clinical notes and pathology data, this work seeks to personalize treatment and improve outcomes for patients with lung cancer.

PI: Ruijiang Li, Associate Professor of Radiation Oncology (Radiation Physics)

Atherosclerotic cardiovascular disease (ASCVD) is the leading cause of mortality in the United States, and despite strong guideline recommendations from the ACC/AHA to start statin therapy and lower low-density lipoprotein cholesterol (LDL-C) to <70 mg/dL in patients with ASCVD, statin initiation rates, statin adherence, and LDL-C goal attainment remain poor. In primary prevention, coronary artery calcium (CAC) is a powerful tool for ASCVD risk stratification and visualization of plaque serves as strong motivator for healthier lifestyle choices and medication adherence. We aim to determine the impact of an AI-enabled quantification and visualization of incidental CAC on non-gated chest CT scans on initiation of lipid-lowering therapy and LDL-C goal achievement in ASCVD patients.

PIs: Fahim Abbasi, Clinical Assistant Professor, Medicine - Cardiovascular Medicine & Fatima Rodriguez, Associate Professor of Medicine (Cardiovascular Medicine)

Cardiovascular disease (CVD) remains the leading cause of morbidity and mortality worldwide, with prevention efforts often hindered by a lack of patient understanding of risk factors and medical options. Risk calculators such as CHA2DS2-VASC, PREVENT, and PCE are widely used to guide treatment and medication prescriptions. Large language models (LLMs) have emerged as transformative tools in healthcare, offering conversational, patient-facing solutions that can simplify complex medical information. While LLMs show potential in improving patient education, their application in collecting cardiovascular risk factors and presenting tailored risk assessments, along with evidence-based medical options, remains underexplored. A personalized, interactive tool that leverages LLMs to assess cardiovascular risk and present medical information could empower patients to make informed decisions and facilitate shared decision-making with clinicians, ultimately improving prevention and management of CVD. This study aims to develop and evaluate a patient-facing tool powered by LLMs to: 1. Collect cardiovascular risk factors such as age, gender, race/ethnicity, history of diabetes, blood pressure, total cholesterol, LDL, HDL-C levels, and smoking status along with medical and lifestyle treatments that they are pursuing for these conditions in an interactive and user-friendly manner. 2. Provide personalized risk assessments, including calculated scores such as the PREVENT and PCE in phase 1 and CHA2DS2-VASC in phase 2. 3. Present evidence-based medical options to guide patients in reducing their cardiovascular risk and improving health outcomes.

PIs: Rajesh Dash, Associate Professor of Medicine (Cardiovascular Medicine), Vijaya Parameswaran, Clinical/Systems Scientist, Cardiology & Rupan Bose, Clinical Assistant Professor, Medicine - Cardiovascular Medicine

Tumor burden (TB) is a well-established prognostic factor in relapsed/refractory large B cell lymphoma (R/R LBCL) treated with CAR-T therapy, and metabolic tumor volume (MTV) derived from 18F-FDG PET-CT offers a comprehensive measure of total metabolically active disease. In this study of a large cohort, we aimed to establish a clinically meaningful risk stratification cutpoint for baseline MTV and applied findings from a discovery cohort to an independent validation cohort. We also propose 12-month toxicity-free progression-free survival (TFPFS12) as a composite endpoint to integrate treatment efficacy and high- grade toxicity into a unified clinical outcome measure.

PI: Saurabh Dahiya, Associate Professor of Medicine (Blood and Marrow Transplantation and Cellular Therapy)

In the first stage of the project, we tackled inconsistent ISR reporting across clinical trials by conducting a literature review, analyzing 500 ISR images, and identifying a key barrier, nonstandard morphology terminology and grading across clinical trials. We conducted a qualitative survey of ISRs by three dermatologists which demonstrated 100% concordance in classifying ISR types, confirming that consistent morphology-based reporting is achievable. Building on this, the goals for the next phase of the project are to 1) develop a Visual ISR Morphology Atlas to guide consistent ISR reporting and 2) Evaluate a Training Survey to assess how the atlas improves accuracy and consistency among clinical trialists. The goal is to deliver validated, image-based tools to standardize ISR classification and improve data quality across clinical trials.

PI: Kavita Sarin, Professor of Dermatology

Severe weather events and other environmental emergencies can threaten health and wellbeing by disrupting clinic-based care. Digital health technologies may provide a lifeline to preserve access to critical services during these events. The Veterans Affairs (VA) Health Care System—the largest integrated healthcare system in the U.S.—operates over 1,100 outpatient clinics to serve more than 9 million enrolled veterans. Given the geographic diversity of these clinics, the VA offers a unique opportunity to study the impact of leveraging digital health technologies during weather events that disrupt routine health care. Our long-term goal is to evaluate the impact of leveraging digital health technologies during disruptive weather events, and to identify and disseminate best practices throughout VA’s national network of outpatient clinics.

PI: Donna Zulman, Professor of Medicine (Primary Care and Population Health)

The design and approval of clinical trials is a complex, resource-intensive process that requires alignment between sponsors, investigators, and regulatory agencies such as the U.S. Food and Drug Administration (FDA). A major challenge in clinical development is anticipating and addressing regulatory feedback—such as requests for additional data, revisions to eligibility criteria, or clarifications of study endpoints—before submission. Currently, much of this learning occurs reactively after initial review cycles, leading to costly delays and redundant amendments. Our project aims to develop an AI agent that learns from historical FDA documentation and Citeline’s clinical trial records to proactively guide trial designers in creating submissions that are scientifically robust and more likely to pass review efficiently.

PI: James Zou, Associate Professor of Biomedical Data Science and, by courtesy, of Computer Science and of Electrical Engineering

We will develop a team of AI agents that can work together to accelerate drug discovery and development. The agents will have diverse expertise ranging from target identification, modality selection, safety analysis and more. We will evaluate this virtual team's ability to prioritize targets.

PI: James Zou, Associate Professor of Biomedical Data Science and, by courtesy, of Computer Science and of Electrical Engineering

Heart disease is a major cause of death in the United States, often striking without warning and causing significant suffering. The underlying condition, called atherosclerosis, develops silently over many years, providing a valuable opportunity for early intervention. Detecting atherosclerosis early could slow or even reverse its progression, reducing the risk of heart disease events. Currently, the common method for detecting silent atherosclerosis is an extra, out-of-pocket scan. Dr. Chaudhari and Dr. Dudum’s project aims to use artificial intelligence to create a new system that identifies atherosclerosis on abdominal scans that are routinely collected in hospitals for a variety of clinical reasons. This approach potentially provides a more accessible and personalized way to assess heart disease risk. By studying thousands of existing scans, their team hopes to improve how physicians prevent and treat heart disease, making better cardiovascular care available to everyone.

PIs: Akshay Chaudhari, Assistant Professor (Research) of Radiology (Integrative Biomedical Imaging Informatics at Stanford) and of Biomedical Data Science & Ramzi Dudum, Member (Postdoc), Cardiovascular Institute

Dr. Alison Okamura and Dr. Sun Kim are working on a smartphone-based solution for early detection of diabetic peripheral neuropathy (DPN), a common complication in people living with diabetes. DPN is a type of nerve damage, affecting touch sensitivity. Current tests often detect DPN at an advanced stage. Their goal is to create a simple, accessible screening method using a smartphone app. By measuring vibration perception with their app, the team aims to identify individuals at risk for DPN before it becomes severe. This could help prevent complications like infections and amputations, providing valuable information to both patients and healthcare providers.

PIs: Allison Okamura, Richard W. Weiland Professor in the School of Engineering and Professor of Mechanical Engineering & Sun H. Kim, Associate Professor of Medicine (Endocrinology)

There is a severe shortage of mental healthcare for youth globally. This shortage may be especially acute for conflict-affected youth, who have high rates of PTSD, depression, and other mental health challenges. The Youth Readiness Intervention is an evidence-based behavioral solution, delivered by trained community health workers (CHWs), to this global health challenge. We propose adapting and digitizing the YRI training for CHWs in Nairobi, Kenya. We will then pilot test the digitized version of the training with Kenyan CHWs, who will deliver the YRI in-person to Somali refugee youth in the Nairobi informal settlements. This project has the potential to vastly change the reach the evidence-based YRI intervention, as well as provide a blueprint for digitization of training components of other prevention interventions.

PIs: Clea Sarnquist, Clinical Professor, Pediatrics - Infectious DiseasesClinical Professor (By courtesy), Epidemiology and Population Health & Michael Baiocchi, Associate Professor of Epidemiology and Population Health and, by courtesy, of Statistics and of Medicine (Stanford Prevention Research Center)

Adult and pediatric patients routinely develop breathing obstructions under anesthesia for surgery or procedures. Fortunately, clinicians can relieve breathing obstructions by tilting the chin or other maneuvers to prevent low oxygen levels—but the breathing obstruction must be detected early. Current monitoring methods, like pulse oximetry and exhaled CO2 monitoring, have limitations as they either detect only the late downstream effects of breathing obstructions or prove unreliable for non-intubated patients. As a result, clinicians often resort to periodically listening to the patient's neck with a stethoscope to gauge respiratory airflow. Unfortunately, due to the intermittent nature of these breathing "spot checks," breathing obstructions can go unnoticed until oxygen levels drop significantly. To address this critical gap in patient safety, we've developed a continuous breath-to-breath monitor that visually displays real-time respiratory airflow in the trachea. Using this device in combination with machine learning, we plan to develop a smart system to provide early warnings of obstructed breathing, enabling clinicians to intervene promptly to prevent episodes of low oxygen levels and enhance patient safety during (and after) anesthesia.

PIs: Matthew Muffly, Clinical Associate Professor, Anesthesiology, Perioperative and Pain Medicine & Chris Chafe, Duca Family Professor - Music

Imagine being able to check your eyesight at home without the hassle of frequent trips to the eye doctor, especially for conditions like age-related macular degeneration and diabetic retinopathy. Dr. Sepah’s team is working on a solution using a virtual reality box that attaches to your smartphone. This affordable device will measure Best Corrected Visual Acuity (BCVA), a crucial indicator of visual function, and transmit real-time data to healthcare providers. By integrating this technology into telemedicine platforms, this project aims to make it easier for patients and doctors, potentially reducing the need for in-person visits and ensuring quicker treatment for common eye diseases.

PI: Yasir Sepah, Assistant Professor (Research) of Ophthalmology