Model training method for depression detection, detection system, program, and storage medium
By constructing an artificial intelligence model based on small molecule compounds and using high-performance liquid chromatography-mass spectrometry detection technology, the problem of low accuracy in depression detection in existing technologies has been solved, achieving efficient and accurate depression risk assessment, which is suitable for clinical testing.
Patent Information
- Application Number
- CN202410430169.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-10
- Publication Date
- 2026-03-20
- Estimated Expiration
- 2044-04-10
AI Technical Summary
Existing technologies for detecting depression suffer from low diagnostic accuracy, cumbersome signal acquisition, and difficulty in clinical application. There is a lack of efficient biomarker detection methods and artificial intelligence-assisted diagnostic technologies.
An artificial intelligence model was constructed using small molecule compounds such as 5-hydroxy-an-aminobenzoic acid, methionine, γ-aminobutyric acid, 3,4-dihydroxyphenylacetic acid, 2-pyrolinic acid, quinolinic acid, kynurenine, 5-hydroxytryptophan, 3-hydroxykynurenine, 3-hydroxy-an-aminobenzoic acid, dopamine, norepinephrine, and homocysteine as biomarkers. The AI model was trained with large sample data, and high-performance liquid chromatography and mass spectrometry were used to determine the concentration of compounds in the sample to assess the risk of depression.
It achieves high-precision and high-speed depression detection, with an AUC greater than 0.95 and an accuracy rate of over 90%. It simplifies the detection process, facilitates application in detection systems, and is suitable for use as an independent data processing unit or storage medium.