RelCon: Relative Contrastive Learning for a Motion Foundation Model for Wearable Data

MA Xu, J Narain, G Darnell, H Hallgrimsson… - arxiv preprint arxiv …, 2024 - arxiv.org
We present RelCon, a novel self-supervised* Rel* ative* Con* trastive learning approach
that uses a learnable distance measure in combination with a softened contrastive loss for …

Memorize and rank: Elevating large language models for clinical diagnosis prediction

MD Ma, X Wang, Y **ao, A Cuturrufo, VS Nori… - arxiv preprint arxiv …, 2025 - arxiv.org
Clinical diagnosis prediction models, when provided with a patient's medical history, aim to
detect potential diseases early, facilitating timely intervention and improving prognostic …

MEDS-Tab: Automated tabularization and baseline methods for MEDS datasets

N Oufattole, T Bergamaschi, A Kolo, H Jeong… - arxiv preprint arxiv …, 2024 - arxiv.org
Effective, reliable, and scalable development of machine learning (ML) solutions for
structured electronic health record (EHR) data requires the ability to reliably generate high …

Pulse-PPG: An Open-Source Field-Trained PPG Foundation Model for Wearable Applications Across Lab and Field Settings

M Saha, MA Xu, W Mao, S Neupane, JM Rehg… - arxiv preprint arxiv …, 2025 - arxiv.org
Photoplethysmography (PPG)-based foundation models are gaining traction due to the
widespread use of PPG in biosignal monitoring and their potential to generalize across …

Foundation Model of Electronic Medical Records for Adaptive Risk Estimation

P Renc, MK Grzeszczyk, N Oufattole, D Goode… - arxiv preprint arxiv …, 2025 - arxiv.org
We developed the Enhanced Transformer for Health Outcome Simulation (ETHOS), an AI
model that tokenizes patient health timelines (PHTs) from EHRs. ETHOS predicts future …