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

Keywords: Large Language Model, Large Vision Language Model, Healthcare, Diagnosis-It-Yourself, Home Care

Abstract:

Generative AI is reshaping healthcare by enhancing multimodal data interpretation, clinical insight generation, and personalized decision support. However, existing advances remain tightly coupled with hospital-grade devices, restricting accessibility and use for anytime, anywhere health management in non-clinical settings. With the proliferation of wearables, mobile sensors, and telemedicine, healthcare is
shifting toward the home, giving rise to the emerging field of Diagnosis-It-Yourself (DIY) at home, i.e., home 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 highly variable task demands and dynamically 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, which integrates expert mixtures with hypernetwork-driven modulation to balance cross-task generalization and instance-level personalization. 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 AI-driven, personalized, and scalable health management at home.

Primary Area: foundation or frontier models, including LLMs

Submission Number: 10866