MidMed: Towards Mixed-Type Dialogues for Medical Consultation

TL;DR

This paper introduces MidMed, a new dataset and framework for medical dialogues that help patients clarify their goals across multiple dialogue types and medical departments, improving consultation systems.

Contribution

The paper presents MidMed, a comprehensive mixed-type medical dialogue corpus and InsMed, an instruction-guided generation framework, addressing goal clarification challenges in medical consultations.

Findings

InsMed outperforms baseline models in experiments.

MidMed covers diverse dialogue types and medical departments.

The framework improves goal clarification in medical dialogues.

Abstract

Most medical dialogue systems assume that patients have clear goals (medicine querying, surgical operation querying, etc.) before medical consultation. However, in many real scenarios, due to the lack of medical knowledge, it is usually difficult for patients to determine clear goals with all necessary slots. In this paper, we identify this challenge as how to construct medical consultation dialogue systems to help patients clarify their goals. To mitigate this challenge, we propose a novel task and create a human-to-human mixed-type medical consultation dialogue corpus, termed MidMed, covering five dialogue types: task-oriented dialogue for diagnosis, recommendation, knowledge-grounded dialogue, QA, and chitchat. MidMed covers four departments (otorhinolaryngology, ophthalmology, skin, and digestive system), with 8,175 dialogues. Furthermore, we build baselines on MidMed and propose an instruction-guiding medical dialogue generation framework, termed InsMed, to address this task. Experimental results show the effectiveness of InsMed.