SEOUL, South Korea, Oct. 22, 2024 /PRNewswire/ -- Doctorpresso's study to detect depression from user-generated diary text data was published in the September 2024 issue of the Journal of Medical Internet Research (JMIR) in recognition of its excellence. JMIR noted the "new approach of Doctorpresso's digital mental health screening."
DoctorPresso Mind Diary
According to Doctorpresso (CEO Whanbo Jung), a new AI technology has been developed to detect depression by utilizing user-generated diary text data. The study suggests an innovative approach for the early diagnosis and treatment of depression, which is increasing worldwide.
Depression is a serious social problem around the world, causing a decrease in job productivity and a high burden of disability. However, existing depression screening tools have limitations in objectivity and accuracy, so new methods were needed to compensate for this. The study developed an AI algorithm to detect depression based on user's diary data obtained through Doctorpresso's own diary application and verified its usefulness.
According to the study, GPT 3.5 and GPT 4.0, AI-based large language models (LLMs), can effectively detect the risk of depression by analyzing the user's diary text. Specifically, for AI developed using LLM models, text-based depression detection accuracy of 90.2 percent and specificity of 95.5 percent were achieved. Through this, Doctorpresso has secured technology to screen for depression early based on user-generated diary data.
"Diary writing is known as a therapeutic tool that promotes emotional expression and self-reflection, and this study confirmed that such textual data can be very useful for early diagnosis of depression," said Dr. Daun Shin (Chief Medical Officer of Doctorpresso, Department of Psychiatry at Korea University Anam Hospital), who led the study. "The importance of research on qualitative digital expression as well as quantitative indicators will increase."
Doctorpresso's technology using this study is expected to contribute greatly to early detection of depression and helping better treat and prevent it.
