DIYHealth Suite: Dataset, Model, and Benchmark for Health Management at Home

Changshuo Liu, Wu Junran, Zhongle Xie, Wenqiao Zhang, Kaiping Zheng, Jiaqi Zhu, Qingpeng Cai, Ooi Gene Anne, Marcus Chun Jin Tan, Jianwei Yin, James Wei Luen Yip, Beng Chin Ooi
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:77028-77076, 2026.

Abstract

Generative AI is reshaping healthcare, yet most existing advances rely on hospital-grade devices, which limits their accessibility and potential for health management outside clinical settings. With the proliferation of portable devices and telemedicine, healthcare is shifting toward home-based Diagnosis-It-Yourself (DIY) care. Despite this promise, several distinctive challenges remain: (i) home-collected data are heterogeneous, exacerbated by the absence of standardized large-scale datasets; (ii) models require adaptation to variable task demands and evolving individual conditions; (iii) the broad spectrum of home care tasks lacks a unified benchmark for systematic evaluation. In this paper, we present DIYHealth Suite, a comprehensive framework designed to address these challenges through a tailored dataset, model, and benchmark. We first curate DIYHealth-900K, a large-scale multimodal dataset capturing diverse real-world home care scenarios. Building on this, we propose DIYHealthGPT, an adaptive foundation model for home-based health management, powered by the novel Hybrid Hyper Low-Rank Adaptation technique. Finally, we establish DIYHealthBench, the first benchmark to evaluate foundation models on home care tasks. Extensive experiments demonstrate that DIYHealthGPT delivers state-of-the-art performance over both general-purpose and medical-specific baselines on 11 home care tasks in both open-QA and closed-QA settings, laying the groundwork for the next generation of personalized health management at home.

Cite this Paper


BibTeX
@InProceedings{pmlr-v306-liu26br, title = {{DIYH}ealth Suite: Dataset, Model, and Benchmark for Health Management at Home}, author = {Liu, Changshuo and Junran, Wu and Xie, Zhongle and Zhang, Wenqiao and Zheng, Kaiping and Zhu, Jiaqi and Cai, Qingpeng and Anne, Ooi Gene and Tan, Marcus Chun Jin and Yin, Jianwei and Yip, James Wei Luen and Ooi, Beng Chin}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {77028--77076}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/liu26br/liu26br.pdf}, url = {https://proceedings.mlr.press/v306/liu26br.html}, abstract = {Generative AI is reshaping healthcare, yet most existing advances rely on hospital-grade devices, which limits their accessibility and potential for health management outside clinical settings. With the proliferation of portable devices and telemedicine, healthcare is shifting toward home-based Diagnosis-It-Yourself (DIY) care. Despite this promise, several distinctive challenges remain: (i) home-collected data are heterogeneous, exacerbated by the absence of standardized large-scale datasets; (ii) models require adaptation to variable task demands and evolving individual conditions; (iii) the broad spectrum of home care tasks lacks a unified benchmark for systematic evaluation. In this paper, we present DIYHealth Suite, a comprehensive framework designed to address these challenges through a tailored dataset, model, and benchmark. We first curate DIYHealth-900K, a large-scale multimodal dataset capturing diverse real-world home care scenarios. Building on this, we propose DIYHealthGPT, an adaptive foundation model for home-based health management, powered by the novel Hybrid Hyper Low-Rank Adaptation technique. Finally, we establish DIYHealthBench, the first benchmark to evaluate foundation models on home care tasks. Extensive experiments demonstrate that DIYHealthGPT delivers state-of-the-art performance over both general-purpose and medical-specific baselines on 11 home care tasks in both open-QA and closed-QA settings, laying the groundwork for the next generation of personalized health management at home.} }
Endnote
%0 Conference Paper %T DIYHealth Suite: Dataset, Model, and Benchmark for Health Management at Home %A Changshuo Liu %A Wu Junran %A Zhongle Xie %A Wenqiao Zhang %A Kaiping Zheng %A Jiaqi Zhu %A Qingpeng Cai %A Ooi Gene Anne %A Marcus Chun Jin Tan %A Jianwei Yin %A James Wei Luen Yip %A Beng Chin Ooi %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-liu26br %I PMLR %P 77028--77076 %U https://proceedings.mlr.press/v306/liu26br.html %V 306 %X Generative AI is reshaping healthcare, yet most existing advances rely on hospital-grade devices, which limits their accessibility and potential for health management outside clinical settings. With the proliferation of portable devices and telemedicine, healthcare is shifting toward home-based Diagnosis-It-Yourself (DIY) care. Despite this promise, several distinctive challenges remain: (i) home-collected data are heterogeneous, exacerbated by the absence of standardized large-scale datasets; (ii) models require adaptation to variable task demands and evolving individual conditions; (iii) the broad spectrum of home care tasks lacks a unified benchmark for systematic evaluation. In this paper, we present DIYHealth Suite, a comprehensive framework designed to address these challenges through a tailored dataset, model, and benchmark. We first curate DIYHealth-900K, a large-scale multimodal dataset capturing diverse real-world home care scenarios. Building on this, we propose DIYHealthGPT, an adaptive foundation model for home-based health management, powered by the novel Hybrid Hyper Low-Rank Adaptation technique. Finally, we establish DIYHealthBench, the first benchmark to evaluate foundation models on home care tasks. Extensive experiments demonstrate that DIYHealthGPT delivers state-of-the-art performance over both general-purpose and medical-specific baselines on 11 home care tasks in both open-QA and closed-QA settings, laying the groundwork for the next generation of personalized health management at home.
APA
Liu, C., Junran, W., Xie, Z., Zhang, W., Zheng, K., Zhu, J., Cai, Q., Anne, O.G., Tan, M.C.J., Yin, J., Yip, J.W.L. & Ooi, B.C.. (2026). DIYHealth Suite: Dataset, Model, and Benchmark for Health Management at Home. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:77028-77076 Available from https://proceedings.mlr.press/v306/liu26br.html.

Related Material