Health-LLM: Large Language Models for Health Prediction via Wearable Sensor Data

Yubin Kim, Xuhai Xu, Daniel McDuff, Cynthia Breazeal, Hae Won Park
Proceedings of the fifth Conference on Health, Inference, and Learning, PMLR 248:522-539, 2024.

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 \textbf{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 \textbf{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.

Cite this Paper

BibTeX
@InProceedings{pmlr-v248-kim24b, title = {Health-LLM: Large Language Models for Health Prediction via Wearable Sensor Data}, author = {Kim, Yubin and Xu, Xuhai and McDuff, Daniel and Breazeal, Cynthia and Park, Hae Won}, booktitle = {Proceedings of the fifth Conference on Health, Inference, and Learning}, pages = {522--539}, year = {2024}, editor = {Pollard, Tom and Choi, Edward and Singhal, Pankhuri and Hughes, Michael and Sizikova, Elena and Mortazavi, Bobak and Chen, Irene and Wang, Fei and Sarker, Tasmie and McDermott, Matthew and Ghassemi, Marzyeh}, volume = {248}, series = {Proceedings of Machine Learning Research}, month = {27--28 Jun}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v248/main/assets/kim24b/kim24b.pdf}, url = {https://proceedings.mlr.press/v248/kim24b.html}, 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 \textbf{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 \textbf{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.} }
Endnote
%0 Conference Paper %T Health-LLM: Large Language Models for Health Prediction via Wearable Sensor Data %A Yubin Kim %A Xuhai Xu %A Daniel McDuff %A Cynthia Breazeal %A Hae Won Park %B Proceedings of the fifth Conference on Health, Inference, and Learning %C Proceedings of Machine Learning Research %D 2024 %E Tom Pollard %E Edward Choi %E Pankhuri Singhal %E Michael Hughes %E Elena Sizikova %E Bobak Mortazavi %E Irene Chen %E Fei Wang %E Tasmie Sarker %E Matthew McDermott %E Marzyeh Ghassemi %F pmlr-v248-kim24b %I PMLR %P 522--539 %U https://proceedings.mlr.press/v248/kim24b.html %V 248 %X 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 \textbf{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 \textbf{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.
APA
Kim, Y., Xu, X., McDuff, D., Breazeal, C. & Park, H.W.. (2024). Health-LLM: Large Language Models for Health Prediction via Wearable Sensor Data. Proceedings of the fifth Conference on Health, Inference, and Learning, in Proceedings of Machine Learning Research 248:522-539 Available from https://proceedings.mlr.press/v248/kim24b.html.

Related Material