OpenMHC: Accelerating the Science of Wearable Foundation Models
Narayan Schuetz, Yuze Bai, Lianggang Pan, Edgar Eggert, Favour Nerrise, Juan Delgado-SanMartin, Max Rosenblattl, Milana Gurbanova, Mohammad Asadi, Anders Johnson, Paul Schmiedmayer, Dennis Wang, Allan Lawrie, Daniel Seung Kim, Xin Liu, Akshay Paruchuri, Ehsan Adeli, Euan Ashley, Kelly W. Zhang
Why It Matters
What makes this one worth your time
This work addresses the critical lack of publicly available wearable health datasets, enabling researchers to advance the field of wearable health AI more effectively.
OpenMHC democratizes wearable health AI research with a comprehensive dataset and open-source models.
Summary
The paper presents OpenMHC, an extensive open-access dataset for wearable health data, along with open-source implementations of wearable foundation models and a unified benchmark for evaluating these models.
Key contributions
- Release of the largest open-access wearable health dataset to date.
- Open-source implementations of recent wearable foundation models.
- A unified benchmark for evaluating wearable health models across multiple tasks.
Notable insights
- The dataset includes over 60 million hours of data, which is a significant scale for training and benchmarking models.
- The introduction of a unified benchmark allows for standardized comparisons across various model types, enhancing reproducibility in research.
Possible limitations
- Not stated in the abstract.
Abstract
arXiv:2607.16235v2 Announce Type: replace-cross Abstract: Mobile and wearable devices offer an unprecedented opportunity for continuous, passive health monitoring and active health coaching. However, the largest wearable datasets are not publicly available for research, and leading wearable foundation models trained on such datasets are rarely open-weight or come with reproducible training code. To accelerate open science in wearable health, we release OpenMyHeartCounts (OpenMHC), the largest and most comprehensive open-access wearable health dataset to date, alongside open-source implementations of recent wearable foundation models. OpenMHC, derived from over a decade of data collected through the My Heart Counts study app, includes >60 million hours of wearable data across 19 sensor channels (e.g., step count, heart rate, sleep, workouts) and up to 169 linked variables, including health, lifestyle, mood, and behavior from 11,894 consenting participants. Furthermore, we introduce a unified, open benchmark that enables standardized comparison of wearable health models across three tracks: health and behavior downstream prediction, multivariate data imputation, and time-series forecasting. We benchmark classical methods alongside recent wearable and multivariate time series foundation models. By open-sourcing data, code, and model weights at this unprecedented scale, we aim to democratize wearable health AI research and enable the community to drive open progress in this domain.