News & Events

Publication: Measuring multi-site pulse transit time with an AI-enabled mmWave radar

Authors: Jiangyifei Zhu, Kuang Yuan, Akrash Prabhakara, Yunzhi Li, Gongwei Wang, Kelly Michaelsen, Justin Chan & Swarun Kumar Abstract Pulse Transit Time (PTT) is a measure of arterial stiffness and a physiological marker associated with cardiovascular function, with an inverse relationship to diastolic blood pressure (DBP). We present an AI-enabled mmWave system for contactless multi-site PTT measurement using a single radar. By leveraging radar beamforming and deep learning algorithms our system simultaneously measures PTT and estimates diastolic blood pressure at multiple sites. The system was evaluated across three physiological pathways – heart-to-radial artery, heart-to-carotid artery, and mastoid area-to-radial artery – achieving correlation coefficients of 0.75–0.86 compared to contact-based reference sensors for measuring PTT. Furthermore, the system demonstrated correlation coefficients of 0.90–0.91 for estimating DBP, and achieved a mean error of -0.62–0.06 mmHg and standard deviation…

Continue ReadingPublication: Measuring multi-site pulse transit time with an AI-enabled mmWave radar

Upcoming Webinar – Best Practices for Digital Phenotyping Research in Aging Populations

Zoom Registration: https://umass-amherst.zoom.us/meeting/register/DOlht6o5Q3OkTNaeWzNF4A Abstract: Digital phenotyping is transforming aging research by enabling high-frequency, real-world measurement of cognition, behavior, symptoms, and context through smartphones, wearables, and passive sensing technologies. This talk will review how digital health tools can complement traditional clinic-based assessments by capturing intraindividual variability, diurnal patterns, environmental influences, and subtle changes in cognitive and functional performance that may signal risk for neurodegenerative disease. Using examples from studies of healthy aging, MCI, Alzheimer’s disease risk, dementia caregiving, and related clinical populations, the talk will highlight best practices for designing digital phenotyping protocols, balancing participant burden with data richness, maximizing adherence, integrating active cognitive assessments with passive data streams and biomarkers, and applying analytic approaches that distinguish within-person change from between-person differences. The session will emphasize opportunities for digital phenotyping to improve early detection, clinical trial…

Continue ReadingUpcoming Webinar – Best Practices for Digital Phenotyping Research in Aging Populations

Past Webinar – TRIALCHAT: Developing Agentic AI Chatbot Tools with Older Adults and Caregivers to Encourage Participation in ADRD Clinical Trials

https://www.youtube.com/watch?v=L70oKKvsgrA Abstract: Hype around the potential of generative AI tools to transform healthcare is at an all time high, but their design and utility is often not user or patient-centered. This is particularly true for generalized large language models that have limited UI/UX features, may hallucinate or give incorrect medical/healthcare advice, or may lack conversational clarity and specificity. In fact, these tools are rarely designed in partnership with the end-user or patients they intend to serve. As part of a2 Collective Pilot award in partnership with MassAITC, S-3 Research and California State Fullerton have been developing “TRIALCHAT", a multiagentic AI tool with the goal of navigating older adults and their caregivers to resources related to Alzheimer’s Disease and clinical research participation opportunities. Lessons learned from a technology design and development process that involved rapid prototyping,…

Continue ReadingPast Webinar – TRIALCHAT: Developing Agentic AI Chatbot Tools with Older Adults and Caregivers to Encourage Participation in ADRD Clinical Trials

Publication: An Explainable Transformer Model for Pain Intensity Assessment Using Multi-Modal Facial Sequential Images

Authors: Xian Du, Meysam Safarzadeh, Maoqin Zhu, Shishir Prasad, Sudeshna Das, Joohyun Chung Abstract Pain monitoring and assessment traditionally rely on subjective methods such as self-reports and caregiver evaluations, which can be costly and often inaccurate due to their inherent subjectivity and reliance on the individual's communication skills. Many objective methods have been introduced to address these issues, primarily utilizing single or multiple wearable sensor modalities. However, these approaches face challenges in home care settings, particularly concerning continuous wearability and discomfort, especially among elderly users. An alternative solution is using patient monitoring tools such as various imaging modalities to detect pain-related facial expressions. In this paper, we developed a new transformer model to extract pain-related features from facial expressions captured through three imaging modalities—RGB, thermal, and depth across sequential images. This method can leverage the…

Continue ReadingPublication: An Explainable Transformer Model for Pain Intensity Assessment Using Multi-Modal Facial Sequential Images

Past Webinar – From AI Validation to Evidence: Proving Readiness for Real-World Aging Care Systems

https://www.youtube.com/watch?v=IIxuaUpqhIM Abstract: Artificial intelligence is increasingly being deployed to support aging populations through applications such as remote monitoring, fall detection, cognitive assessment, and caregiver support. While many AI systems demonstrate strong performance in pilot studies, only a small fraction successfully translate into safe, reliable, and scalable real-world deployments. A key challenge lies in the gap between validation and real-world evidence. Traditional evaluation approaches emphasize model performance, but aging care environments introduce additional complexities—including multimorbidity, heterogeneous living conditions, and safety-critical decision contexts—that are often not adequately captured during pilot validation. This webinar presents a structured approach to moving from AI validation to evidence-driven deployment readiness for aging care systems. The session will explore lifecycle-based evaluation methods spanning data, model development, pilot testing, real-world deployment, and continuous monitoring. It will highlight approaches for assessing fairness across aging…

Continue ReadingPast Webinar – From AI Validation to Evidence: Proving Readiness for Real-World Aging Care Systems

Grant Funding: NIH R01: mHealth Technologies for Assessing Hearing Disorders

Project Abstract Sound Sensitivity and Tolerance (SST) disorders represent an urgent public health challenge in hearing disorders, affecting over 350,000 new patients annually in the United States with an economic burden exceeding $14 billion. While SST frequently co-occurs with conditions like autism spectrum disorder, PTSD, and migraine, current clinical assessment methods remain largely subjective and inaccessible, particularly for underserved populations. Recent research shows that Auditory Pupillary Response (APR) provides a promising objective biomarker for SST, as affected individuals exhibit distinct pupillary reactions to sound stimuli. However, measuring APR currently requires expensive clinical equipment and controlled laboratory settings, creating significant barriers to widespread adoption. This project aims to transform SST assessment through AudioSight, an innovative smartphone-based system that enables accessible, objective measurement of APR. Building on our preliminary data showing clear APR differentiation between SST patients…

Continue ReadingGrant Funding: NIH R01: mHealth Technologies for Assessing Hearing Disorders