|
Sarah Jiang
Hi! I'm a Biomedical Engineering PhD student at
Duke University,
where I work on developing digital health tools and deep learning methods to improve
health outcomes and healthcare access in the
BIG IDEAs Lab,
advised by Dr. Jessilyn Dunn.
I'm broadly interested in leveraging wearable devices and biosignals for remote monitoring
and early detection and intervention of chronic illness. I'm fortunate to be funded by the
NSF Graduate Research Fellowship.
Previously, I completed my BSE in Biomedical Engineering and Computer Science at Duke University,
where I conducted research on wearable sensor advancements and biosignal data demographics,
and developed full-stack applications for healthcare data management. I also interned at UCLA
where I was advised by Dr. Yuzhe Yang in the
Health Intelligence Lab (HAIL).
As part of HAIL, I co-led a project evaluating various self-supervised learning objectives
to build a general purpose motion foundation model and created a unified dataset of high
resolution accelerometer data for activity and disease tasks. Beyond research, I was also a
machine learning engineer at Mantis (YC W26), where I worked on developing AI models for
predictive human analytics and behavior modeling for professional athletes.
When I'm not working on a project, you can find me playing the piano, exploring local coffee shops
or buried in a good book [currently reading: The Shadow of the Gods by John Gwynne & Emily Wilson's translation of The Odyssey].
Email
/
CV
/
Scholar
/
LinkedIn
/
Github
|
|
|
Sep. 2026
|
Inertia-1 accepted at NeurIPS 2026!
[paper]
|
|
Aug. 2026
|
Started my PhD at Duke in the Big Ideas Lab!!
|
|
April 2025
|
Awarded the NSF Graduate Research Fellowship (GRFP).
|
Research
Interested in AI for multimodal biomedical data, with a focus on wearable sensing and clinical applications.
* denotes equal contribution
|
|
|
Inertia-1: An Open Exploration of Wearable Foundation Models
Z. Xu*, A. Anand*, S. Jiang*, Z. Shuai, C. Zhuang, S. Sankararaman, Y. Yang
NeurIPS (Poster), Dec 2026
arXiv
A fully open cookbook for building unified wearable motion foundation models, studying the
full lifecycle of data, sensing, objectives, scale, and transfer across 18.2M+ hours of motion
data, 115,000+ individuals, and 1,000+ trained models to enable a single motion backbone that
generalizes across body placements, devices, and downstream health tasks.
|
|
|
Demographic reporting in biosignal datasets: a comprehensive analysis of the PhysioNet open access database
S. Jiang*, P. Ashar*, M. Shandhi, J. Dunn
The Lancet Digital Health, Nov. 2024
Full Text
Comprehensive analysis of demographic reporting and biases in wearable biosignal datasets
and their implications for ML model clinical deployment.
|
|
|
Determinants of Opioid Use Disorder Relapse from the Biopsychosocial Perspective: A Systematic Review
L. Lederer*, M. Liu*, B. Chen, S. Jiang, S. Kim, D. MacKenzie, E. Ho, G. Guerreri,
A. Roghani, J. Dunn
CERSI Summit, Jan. 2024
By identifying statistically significant determinants that influence relapse, we seek to
inform the development of digital health technologies for supporting relapse prevention efforts.
Full paper under review.
|
|
|
Assessment of Cardiorespiratory Fitness and Functional Capacity: From Clinics to Real-World Settings
M. M. H. Shandhi*, H. Jeong*, S. Jiang, P. Ashar, S. Kavirajuni, A. V. Kotla,
M. Fudim, H. Pontzer, W. E. Kraus, J. Dunn
Under Review
Systematic review of SOTA wearable sensor technologies for cardiovascular health monitoring
and functional capacity assessment against gold-standard clinical measures.
|
|
Academic Service
|
Reviewer,
IEEE EMBS Conference on Healthcare Innovation - Point-of-Care Technologies (HI-POCT) 2024
|
|
Teaching
|
Graduate Teaching Assistant (BME Data Science) –
BME 704/304 Fall 2026
Head Undergraduate Teaching Assistant (Intro to Data Science) –
CS216 Spring 2023,
Fall 2023,
Spring 2024,
Fall 2024,
Spring 2025
Undergraduate Teaching Assistant (Engineering Design & Technical Communication) –
EGR 101 Fall 2022
|
|