본문 바로가기
  • Home

CCTV-Based Behavioral Indicator Analysis System for Mental Health Evaluation in Protective Wards

  • Journal of The Korea Society of Computer and Information
  • Abbr : JKSCI
  • 2024, 29(12), pp.87-98
  • Publisher : The Korean Society Of Computer And Information
  • Research Area : Engineering > Computer Science
  • Received : November 14, 2024
  • Accepted : December 5, 2024
  • Published : December 31, 2024

Hee-Cheol Kim 1 Kang-Hee Lee 2 Yourack Lee 3 Hyun-Ghang Jeong 3 Hyun-Soo Choi 4

1(주)지오비전
2강원대학교
3고려대학교 의과대학, 고려대학교 구로병원
4서울과학기술대학교

Accredited

ABSTRACT

This study proposes a non-invasive system that collects patients' daily behavioral indicators to evaluate and predict mental health status. By extracting indicators such as travel distance, activity radius, time spent alone, and duration of contact with others, the system aims to detect changes in psychological states, including depression, anxiety, and suicide risk. To collect these behavioral indicators, the system utilizes object identification and tracking based on the Bird’s Eye View method to ensure consistent tracking and obtain location information across multiple camera perspectives. Based on the extracted location information, we generated behavioral indicators and analyzed their correlations with suicide risk and mental health measures. Our analysis revealed statistically significant relationships between the behavioral indicators and mental health measures. This indicates that behavioral indicators can objectively assess patients' mental health status. By enabling the evaluation of patients' mental health through this system, we expect to contribute to regular patient management and improvements in mental health care.

Citation status

* References for papers published after 2023 are currently being built.

This paper was written with support from the National Research Foundation of Korea.