Development of a Depression Severity Classification Model Through Natural Language Processing by Psychiatry Clinical Texts

Read full text Summarise with VIRAT
Authors: Myung-Gwan Kim, ChanYoung Song, Dawoon Wang, 한현욱
Cited by: 0
References: 17

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Background: Depression has surged in Korea, with 933,481 patients in 2021, a 35.1% increase since 2017.Globally, 5% of adults suffer from depression, resulting in over 700,000 suicides annually.However, Korea has only 29.5 mental health workers per 100,000 people, below the OECD average of 97.1.This underscores the need for improved mental illness classification support systems.Medical artificial intelligence (AI) technology can help alleviate the burden of insufficient mental health resources and treatment overload. Methods:We developed an AI model for a clinical decision support system to determine depression severity using natural language data from patients in the Bundang CHA University Hospital Clinical Data Warehouse between 2018 and 2022.Among these, 169 patients had mild depressive episodes, and 460 had moderate depressive episodes.The control group used natural language datasets from the AI Hub platform.The final dataset included 460 patients with moderate depression, 169 with mild depression, and 123,690 normal conversation sessions.Results: Various algorithms were applied to diagnose depression severity based on reported symptoms, and XGBoost achieved the highest accuracy: 99.7% accuracy, 99.6% precision, 99.7% recall, and a 99.6% F1 score.The area under the curve was close to 1. Conclusion: Utilizing advanced AI and natural language processing technology in psychiatry can significantly aid in the precise, personalized assessment of depression severity based on patients' expressed symptoms.

Fields of Study & Subjects

Mental Health via Writing Artificial Intelligence in Healthcare Machine Learning in Healthcare

Publication details

Journal / source
Digital Health Research
Volume
4
Issue
1
Publication date
1 January 2026
Language
EN