EUVIMEDEUVIMED entdecken
Crossref

CausalSymptom: Learning Causal Disentangled Representation for Depression Severity Estimation on Transcribed Clinical Interviews

Mingzheng Li, Xiao Sun, Xinke Wang, Fengqi Cui, Xun Yang

IEEE Transactions on Affective Computing · 2026-01 · 17 (1) · 277-289

DOI: 10.1109/taffc.2025.3613376 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Affective Computing, 2026-01. Verfasst von Mingzheng Li und 4 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 69 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Anhui Provincial Key Research and Development Project, Advanced Computing resources provided by the Supercomputing Center of the USTC.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Depression is one of the most prevalent mental disorders in the world. Although some researchers have applied deep learning to assess depression, understanding its pathogenesis remains challenging. Psychological research highlights the inherent correlations among several depression symptoms. However, existing methods often ignore causal connections, treating symptoms independently and relying on local relationships without identifying valid causal paths. In this paper, we propose a novel framework, CausalSymptom, based on a variational autoencoder, to predict the severity of depression and explore its pathogenesis. The framework incorporates causal representation learning to model latent symptom states. Crucially, we provide theoretical evidence that these symptom variables can be disentangled from others by using additional information as supervision. Once these representations are learned, we present a new framework that includes a novel module, the Graph Causal Effect Network (GCEN), for the estimation of depression. This module learns a causal graph representation based on a causal total effect mechanism, which integrates causal influences across all paths, leading to a more accurate estimation of depression severity. Our model achieves competitive performance on the DAIC-WOZ and E-DAIC datasets. Furthermore, the experimental study empirically shows that our model offers insight into the pathogenesis of depression.

Förderung

  • National Natural Science Foundation of China
  • Anhui Provincial Key Research and Development Project
  • Advanced Computing resources provided by the Supercomputing Center of the USTC

Zitieren

APA

Li, M., Sun, X., Wang, X., Cui, F., & Yang, X. (2026). CausalSymptom: Learning Causal Disentangled Representation for Depression Severity Estimation on Transcribed Clinical Interviews. IEEE Transactions on Affective Computing, 17(1), 277-289. https://doi.org/10.1109/taffc.2025.3613376

Vancouver

Li M, Sun X, Wang X, Cui F, Yang X. CausalSymptom: Learning Causal Disentangled Representation for Depression Severity Estimation on Transcribed Clinical Interviews. IEEE Transactions on Affective Computing. 2026;17(1):277-289. doi:10.1109/taffc.2025.3613376

BibTeX
@article{Li2026,
  author = {Li, Mingzheng and Sun, Xiao and Wang, Xinke and Cui, Fengqi and Yang, Xun},
  title = {CausalSymptom: Learning Causal Disentangled Representation for Depression Severity Estimation on Transcribed Clinical Interviews},
  journal = {IEEE Transactions on Affective Computing},
  year = {2026},
  volume = {17},
  number = {1},
  pages = {277--289},
  doi = {10.1109/taffc.2025.3613376}
}
Kostenlos starten →