Exploring the Capabilities of a Language Model-Only Approach for Depression Detection in Text Data

Sadeghi M, Egger B, Agahi R, Richer R, Capito K, Rupp L, Gmelch LM, Berking M, Eskofier B (2023)

Publication Language: English

Publication Type: Conference contribution, Original article

Publication year: 2023

Publisher: IEEE

Pages Range: 5

Conference Proceedings Title: IEEE EMBS International Conference on Biomedical and Health Informatics (BHI)

Event location: Pittsburgh, PA, USA US

ISBN: 979-8-3503-1050-4

URI: https://ieeexplore.ieee.org/document/10313367

DOI: 10.1109/BHI58575.2023.10313367

Abstract

Depression is a prevalent and debilitating mental health condition that requires accurate and efficient detection for timely and effective treatment. In this study, we utilized the E-DAIC (Extended Distress Analysis Interview Corpus-Wizard-of-Oz) dataset, an extended version of the DAIC-WOZ dataset, which consists of semi-clinical interviews conducted by an animated virtual interviewer called Ellie, controlled by a human interviewer in another room. With 275 participants, the E-DAIC dataset represents a valuable resource for investigating depression detection methods. Our aim is to predict PHQ-8 scores through text analysis. Leveraging state-of-the-art speech processing, LLM-based text summarization, and a specialized depression detection module, we demonstrate the transformative potential of language data analysis in enhancing depression screening. By overcoming the limitations of manual feature extraction methods, our automated techniques provide a more efficient and effective means of evaluating depression. In our evaluation, we achieve robust accuracy on the development set of the E-DAIC dataset, with a Mean Absolute Error (MAE) of 3.65 in estimating PHQ-8 scores from recorded interviews. This remarkable performance highlights the efficacy of our approach in automatically predicting depression severity. Our research contributes to the growing evidence supporting the use of LLMs in mental health assessment, showcasing the role of innovative technologies in advancing patient care for depression.

Authors with CRIS profile

Misha Sadeghi Sonderforschungsbereich 1483/1 Empathokinästhetische Sensorik – Sensortechniken und Datenanalyseverfahren zur empathokinästhetischen Modellbildung und Zustandsbestimmung (EmpkinS) Bernhard Egger Professur für Cognitive Computer Vision (Stiftungsprofessur) Robert Richer Machine Learning and Data Analytics Lab Klara Capito Sonderforschungsbereich 1483/1 Empathokinästhetische Sensorik – Sensortechniken und Datenanalyseverfahren zur empathokinästhetischen Modellbildung und Zustandsbestimmung (EmpkinS) Lydia Rupp Lehrstuhl für Klinische Psychologie und Psychotherapie (KliPs) Lena Marie Gmelch Lehrstuhl für Klinische Psychologie und Psychotherapie (KliPs) Matthias Berking Lehrstuhl für Klinische Psychologie und Psychotherapie (KliPs) Björn Eskofier Machine Learning and Data Analytics Lab

Additional Organisation(s)

Sonderforschungsbereich 1483/1 Empathokinästhetische Sensorik – Sensortechniken und Datenanalyseverfahren zur empathokinästhetischen Modellbildung und Zustandsbestimmung (EmpkinS)

Related research project(s)

Empathokinästhetische Sensorik für Biofeedback bei depressiven Patienten (SFB 1483 EmpkinS D02) Empatho-Kinaesthetic Sensor Technology (SFB 1483 EmpkinS) July 1, 2021 - June 30, 2025

How to cite

APA:

Sadeghi, M., Egger, B., Agahi, R., Richer, R., Capito, K., Rupp, L.,... Eskofier, B. (2023). Exploring the Capabilities of a Language Model-Only Approach for Depression Detection in Text Data. In IEEE EMBS International Conference on Biomedical and Health Informatics (BHI) (pp. 5). Pittsburgh, PA, USA, US: IEEE.

MLA:

Sadeghi, Misha, et al. "Exploring the Capabilities of a Language Model-Only Approach for Depression Detection in Text Data." Proceedings of the IEEE EMBS International Conference on Biomedical and Health Informatics (BHI) 2023, Pittsburgh, PA, USA IEEE, 2023. 5.

BibTeX: Download