Table of Contents
Fetching ...

Breaking the Stigma! Unobtrusively Probe Symptoms in Depression Disorder Diagnosis Dialogue

Jieming Cao, Chen Huang, Yanan Zhang, Ruibo Deng, Jincheng Zhang, Wenqiang Lei

TL;DR

Stigma surrounding depression hinders open dialogue and accurate diagnosis. The authors introduce UPSD4, a two-module dialogue framework comprising an Unobtrusive Probing Module and a Conversational Diagnosis Module, to unobtrusively elicit symptom information and apply ICD-11–based criteria for diagnosis. The approach uses discreet probing strategies and a two-stage strategy selection guided by psychology theory, enabling dynamic adaptation to users' stigma levels. Experiments on stigma-aware and non-stigma user simulators show UPSD4 improves symptom disclosure and diagnostic accuracy over baselines. This work advances accessible, user-friendly depression screening, with potential for real-world deployment under ethical constraints.

Abstract

Stigma has emerged as one of the major obstacles to effectively diagnosing depression, as it prevents users from open conversations about their struggles. This requires advanced questioning skills to carefully probe the presence of specific symptoms in an unobtrusive manner. While recent efforts have been made on depression-diagnosis-oriented dialogue systems, they largely ignore this problem, ultimately hampering their practical utility. To this end, we propose a novel and effective method, UPSD$^{4}$, developing a series of strategies to promote a sense of unobtrusiveness within the dialogue system and assessing depression disorder by probing symptoms. We experimentally show that UPSD$^{4}$ demonstrates a significant improvement over current baselines, including unobtrusiveness evaluation of dialogue content and diagnostic accuracy. We believe our work contributes to developing more accessible and user-friendly tools for addressing the widespread need for depression diagnosis.

Breaking the Stigma! Unobtrusively Probe Symptoms in Depression Disorder Diagnosis Dialogue

TL;DR

Stigma surrounding depression hinders open dialogue and accurate diagnosis. The authors introduce UPSD4, a two-module dialogue framework comprising an Unobtrusive Probing Module and a Conversational Diagnosis Module, to unobtrusively elicit symptom information and apply ICD-11–based criteria for diagnosis. The approach uses discreet probing strategies and a two-stage strategy selection guided by psychology theory, enabling dynamic adaptation to users' stigma levels. Experiments on stigma-aware and non-stigma user simulators show UPSD4 improves symptom disclosure and diagnostic accuracy over baselines. This work advances accessible, user-friendly depression screening, with potential for real-world deployment under ethical constraints.

Abstract

Stigma has emerged as one of the major obstacles to effectively diagnosing depression, as it prevents users from open conversations about their struggles. This requires advanced questioning skills to carefully probe the presence of specific symptoms in an unobtrusive manner. While recent efforts have been made on depression-diagnosis-oriented dialogue systems, they largely ignore this problem, ultimately hampering their practical utility. To this end, we propose a novel and effective method, UPSD, developing a series of strategies to promote a sense of unobtrusiveness within the dialogue system and assessing depression disorder by probing symptoms. We experimentally show that UPSD demonstrates a significant improvement over current baselines, including unobtrusiveness evaluation of dialogue content and diagnostic accuracy. We believe our work contributes to developing more accessible and user-friendly tools for addressing the widespread need for depression diagnosis.
Paper Structure (23 sections, 4 figures, 19 tables)

This paper contains 23 sections, 4 figures, 19 tables.

Figures (4)

  • Figure 1: An example of a comparison of the existing method with unobtrusively probing symptoms in depression diagnosis is when the user exhibits Depression Mood and next turns to probe Suicide Tendency.
  • Figure 2: A taxonomy of related work in depression-diagnosis-oriented dialogue. We highlight the need for unobtrusiveness in stigma-aware diagnosis.
  • Figure 3: USPD$^4$ contains two interconnected modules: Unobtrusive Probing Module (UPM) and Conversational Diagnosis Module (CDM). The UPM, guided by our probing strategies, promotes a sense of unobtrusiveness within the dialogue system and cultivates questioning skills. Meanwhile, the CDM leverages established diagnostic criteria to evaluate potential symptoms in them.
  • Figure 4: The distribution of coarse-grained strategies (left) and fine-grained Questioning Skill strategies (right) across different user simulators. UPSD$^{4}$ dynamically adjusts its unobtrusive probing strategies to accommodate users with varying levels of stigma sensitivity.