Enhancing Depression-Diagnosis-Oriented Chat with Psychological State Tracking

Yiyang Gu,Yougen Zhou,Qin Chen,Ningning Zhou,Jie Zhou,Aimin Zhou,Liang He

2024-03-12

Abstract:Depression-diagnosis-oriented chat aims to guide patients in self-expression to collect key symptoms for depression detection. Recent work focuses on combining task-oriented dialogue and chitchat to simulate the interview-based depression diagnosis. Whereas, these methods can not well capture the changing information, feelings, or symptoms of the patient during dialogues. Moreover, no explicit framework has been explored to guide the dialogue, which results in some useless communications that affect the experience. In this paper, we propose to integrate Psychological State Tracking (POST) within the large language model (LLM) to explicitly guide depression-diagnosis-oriented chat. Specifically, the state is adapted from a psychological theoretical model, which consists of four components, namely Stage, Information, Summary and Next. We fine-tune an LLM model to generate the dynamic psychological state, which is further used to assist response generation at each turn to simulate the psychiatrist. Experimental results on the existing benchmark show that our proposed method boosts the performance of all subtasks in depression-diagnosis-oriented chat.

Human-Computer Interaction,Artificial Intelligence,Computation and Language,Computers and Society

What problem does this paper attempt to address?

This paper aims to address the issues in depression diagnosis dialogues, especially the inability of existing methods to effectively capture changes in patients' emotions and symptoms during conversations, as well as the lack of clear guidance frameworks for dialogues. The paper proposes a Large Language Model (LLM) combined with Psychological State Tracking (POST) to guide depression diagnosis-oriented conversations more accurately. POST is based on the ABC model of Cognitive Behavioral Therapy, which includes the components of Stage, Information, Summary, and Next. The LLM is fine-tuned to generate dynamic psychological states and used for response generation in each round, simulating the dialogue strategies of psychiatrists. Experimental results show that this approach improves the performance of depression diagnosis dialogue tasks and provides explanatory power consistent with professional standards. Additionally, the paper also provides refined annotations on the D4 dataset to support psychological state tracking.

Enhancing Depression-Diagnosis-Oriented Chat with Psychological State Tracking

Towards Reliable and Empathetic Depression-Diagnosis-Oriented Chats

Multi-Modal and Multi-Task Depression Detection with Sentiment Assistance

Read, Diagnose and Chat: Towards Explainable and Interactive LLMs-Augmented Depression Detection in Social Media

InterMind: A Doctor-Patient-Family Interactive Depression Assessment System Empowered by Large Language Models

Depression Diagnosis Dialogue Simulation: Self-improving Psychiatrist with Tertiary Memory

D4: a Chinese Dialogue Dataset for Depression-Diagnosis-Oriented Chat

Multi-Task Learning for Depression Detection in Dialogs

Depression Detection on Social Media with Large Language Models

Depression Detection in Clinical Interviews with LLM-Empowered Structural Element Graph

Multi-aspect Depression Severity Assessment via Inductive Dialogue System

Leveraging ChatGPT to optimize depression intervention through explainable deep learning

Cooperative Multimodal Approach to Depression Detection in Twitter

Depression Detection On Social Media With Reinforcement Learning

Detecting depression in dyadic conversations with multimodal narratives and visualizations

Multilevel depression status detection based on fine-grained prompt learning

PsyChatbot: A Psychological Counseling Agent Towards Depressed Chinese Population Based on Cognitive Behavioural Therapy

When LLMs Meets Acoustic Landmarks: An Efficient Approach to Integrate Speech into Large Language Models for Depression Detection

Advancing Conversational Psychotherapy: Integrating Privacy, Dual-Memory, and Domain Expertise with Large Language Models

Structured Dialogue System for Mental Health: An LLM Chatbot Leveraging the PM+ Guidelines

Depression Detection Via Harvesting Social Media: A Multimodal Dictionary Learning Solution