<!–
permalink: / title: “Pragya Sharma” layout: single author_profile: true classes: wide excerpt: “AI + wearables + clinical time-series for digital health.” redirect_from:
- /about/
/about.html
I am an interdisciplinary researcher working at the intersection of noninvasive sensing, signal processing, and artificial intelligence. My work focuses on translating continuous physiological data from bedside monitors and wearable devices into clinically meaningful insights that enable earlier detection of health changes and more personalized patient care.
Currently: Instructor & Schmidt AI Fellow in the Windreich Department of Artificial Intelligence and Human Health, Icahn School of Medicine at Mount Sinai, applying AI to wearables, clinical, and imaging data.
At a glance
I am an interdisciplinary researcher working at the intersection of noninvasive sensing, signal processing, and artificial intelligence. My work focuses on translating continuous physiological data from bedside monitors and wearable devices into clinically meaningful insights that enable earlier detection of health changes and more personalized patient care.
PhD: Wearable RF Sensing for Cardiorespiratory Monitoring
I received my Ph.D. in Electrical and Computer Engineering from Cornell University, where I developed a novel wearable radio-frequency sensor for simultaneous respiratory and cardiac monitoring. The system enabled estimation of respiratory rate, respiratory volume, and heart rate using near-field RF sensing.
This technology was validated through controlled human-subject studies and clinical collaboration at the Weill Cornell Medical Sleep Center, demonstrating robustness across irregular breathing patterns and real-world sleep conditions.
Digital Health Industry: Remote Patient Monitoring
At Biofourmis, a digital health startup, I led research and development efforts in remote patient monitoring systems deployed in real-world care settings.
My work included:
- Developing a deep-learning model for adult sleep apnea detection using pulse rate and oxygen saturation
- Improving wearable-based respiratory rate estimation algorithm
- Improving SpO₂ alert systems to reduce false alarms and optimize clinical alert burden
These efforts supported scalable hospital-at-home and chronic disease monitoring programs.
Academic Medicine: AI for Complex Clinical Populations
Currently, in collaboration with clinicians across rheumatology, pediatrics, and cardiology, I apply AI methods to wearable, clinical, and imaging datasets. Ongoing projects include:
- Predicting disease activity in rheumatoid arthritis using wearable devices
- Outcome prediction in transient tachypnea of the newborn
- Echocardiogram-based dysfunction prediction in pediatrics
These projects extend AI development to clinically complex populations where physiology is dynamic and labeled data are limited.
Vision
My long-term goal is to integrate continuous sensing and AI into scalable, real-world clinical solutions that advance precision health monitoring across both acute and chronic care settings.