Health-LLM: Large Language Models for Health Prediction via Wearable Sensor Data
Authors:
- Yubin KimYKYubin Kim
- Xuhai XuXXXuhai Xu
- Daniel McDuffDMDaniel McDuff
- Cynthia BreazealCBCynthia Breazeal
- Hae Won ParkHPHae Won Park
Preprint
Abstract
Large language models (LLMs) are capable of many natural language tasks, yet they are far from perfect. In health applications, grounding and interpreting domain-specific and non-linguistic data is crucial. This paper investigates the capacity of LLMs to make inferences about health based on contextual information (e.g. user demographics, health knowledge) and physiological data (e.g. resting heart rate, sleep minutes). We present a comprehensive evaluation of 12 state-of-the-art LLMs with prompting and fine-tuning techniques on four public health datasets (PMData, LifeSnaps, GLOBEM and AW_FB). Our experiments cover 10 consumer health prediction tasks in mental health, activity, metabolic, and sleep assessment. Our fine-tuned model, HealthAlpaca exhibits comparable performance to much larger models (GPT-3.5, GPT-4 and Gemini-Pro), achieving the best performance in 8 out of 10 tasks. Ablation studies highlight the effectiveness of context enhancement strategies. Notably, we observe that our context enhancement can yield up to 23.8% improvement in performance. While constructing contextually rich prompts (combining user context, health knowledge and temporal information) exhibits synergistic improvement, the inclusion of health knowledge context in prompts significantly enhances overall performance.Keywords
All Related Versions
This publication has 21 references indexed in Scilit:
- Towards accurate differential diagnosis with large language modelsNature, 2025
- A survey of large language models for healthcare: from data, technology, and applications to accountability and ethicsInformation Fusion, 2025
- Toward expert-level medical question answering with large language modelsNature Medicine, 2025
- DRG-LLaMA : tuning LLaMA model to predict diagnosis-related group for hospitalized patientsnpj Digital Medicine, 2024
- Ensemble machine learning model trained on a new synthesized dataset generalizes well for stress prediction using wearable devicesJournal of Biomedical Informatics, 2023
- Multimodal LLMs for Health Grounded in Individual-Specific Data2023
- Large language models encode clinical knowledgeNature, 2023
- LifeSnaps, a 4-month multi-modal dataset capturing unobtrusive snapshots of our lives in the wildScientific Data, 2022
- Atrial Fibrillation Detection Using a Feedforward Neural NetworkJournal of Medical and Biological Engineering, 2022
- Large-Scale Assessment of a Smartwatch to Identify Atrial FibrillationNew England Journal of Medicine, 2019