A method and system for constructing a roche corpus for depression screening and a storage medium
By constructing the Rorschach corpus system, the problems of poor correlation of multimodal data and difficulty in version management in Rorschach test research were solved, realizing efficient integrated management of data and support for deep learning, and improving the standardization and efficiency of research.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- NAT UNIV OF DEFENSE TECH
- Filing Date
- 2026-02-12
- Publication Date
- 2026-06-02
AI Technical Summary
Existing Rorschach test research suffers from poor correlation of multimodal data (speech, text, video, and encoding), difficulty in version management, low collaboration efficiency, non-standard metadata recording, and difficulty in supporting deep learning model training.
We construct a Rorschach corpus system for depression screening. By collecting and preprocessing multimodal data, we establish a core data model for structured storage, provide data association and retrieval modules, support deep learning model integration, and enable cross-modal queries and anonymized export.
It achieves high correlation and integrated management of multimodal data, improves the standardization and reproducibility of research, enhances collaboration and analysis efficiency, directly supports deep learning research, adapts to different coding systems, and has flexibility and scalability.
Smart Images

Figure CN122132424A_ABST