Psychological crisis detection based on behavioral data: a new approach to non-contact measurement.

ABSTRACT:

Background

Timely identification of psychological crisis reduces the risk of self-harm or suicide and maintains mental health. Previous methods of identifying psychological crisis have relied on contact methods such as annual psychological surveys and voluntary reports from peers and teachers to identify students in psychological crisis. However, these methods have the disadvantages of time lag and limited scope. This study aims to propose a contactless measurement method that can detect psychological crises in a timely manner. Based on multi-dimensional data, this study constructed a feature set of psychological crises through digital phenotyping, established prediction models of multiple machine learning algorithms, and verified their effectiveness.

Methods

We collected be

SUBMITTER: Lin J 

PROVIDER: S-EPMC12696908 | biostudies-literature | 2025 Dec

REPOSITORIES: biostudies-literature