Enabling Doctor-Centric Medical AI with LLMs through Workflow-Aligned Tasks and Benchmarks
This work addresses the need for doctor-centric AI in clinical workflows, though it is incremental by building on existing LLM applications in healthcare.
The authors tackled the problem of safely integrating large language models (LLMs) into healthcare by repositioning them as clinical assistants for doctors, rather than direct patient tools, and created DoctorFLAN, a dataset of 92,000 Q&A instances that improved open-source LLM performance in medical contexts.
The rise of large language models (LLMs) has transformed healthcare by offering clinical guidance, yet their direct deployment to patients poses safety risks due to limited domain expertise. To mitigate this, we propose repositioning LLMs as clinical assistants that collaborate with experienced physicians rather than interacting with patients directly. We conduct a two-stage inspiration-feedback survey to identify real-world needs in clinical workflows. Guided by this, we construct DoctorFLAN, a large-scale Chinese medical dataset comprising 92,000 Q&A instances across 22 clinical tasks and 27 specialties. To evaluate model performance in doctor-facing applications, we introduce DoctorFLAN-test (550 single-turn Q&A items) and DotaBench (74 multi-turn conversations). Experimental results with over ten popular LLMs demonstrate that DoctorFLAN notably improves the performance of open-source LLMs in medical contexts, facilitating their alignment with physician workflows and complementing existing patient-oriented models. This work contributes a valuable resource and framework for advancing doctor-centered medical LLM development