Abstract
The invention relates to computer-based methods and systems for automated detection of symptoms of cognitive impairment in a person for diagnostic purposes, in particular for automated detection of depressive disorders. The proposed method for automated detection of depressive disorders includes: (i) obtaining a set of images captured by a speaking subject and a set of images captured by a listening subject with a video camera; (ii) obtaining a set of voice sounds of the subject with a microphone; (iii) obtaining the sets of images captured by the subject and the sets of voice sounds with a processor connected to a video camera and a microphone, distinguishing the sets of images captured by the subject obtained when the subject speaks and the subject listens with a processor; (iv) generating image data including the frequency of the subject looking down, the length of the subject's lip corner pulled down, or the length of the subject's lip corner pulled to the side, with a processor based on the obtained sets of capture images; further, image data is generated separately for a speaking subject and a listening subject; (v) generating voice sound data including voice tone, voice volume, and the number of words spoken by the subject in 10 seconds by the processor based on the obtained voice sounds; (vi) transmitting a memory storing data on the threshold value of the subject's downward gaze frequency, the length of the subject's lip corner pulled down, or the length of the subject's lip corner pulled sideways, voice tone, voice volume, and the number of words spoken in 10 seconds by the processor; (vii) comparing the obtained image data and voice sound data with the data on the threshold value of the image and voice sound stored in the memory; further, the compared image data includes the subject's downward gaze frequency, the length of the subject's lip corner pulled down, or the length of the subject's lip corner pulled sideways; and comparing the voice sound data, which includes the tone of voice, the volume of voice and the number of words spoken by the subject in 10 seconds, with the processor; (viii) determining with the processor whether the value of the acquired image and voice sound data exceeds the value of the image and voice sound threshold stored in the memory, if so, sending a notification about the detected depressive disorders using the notification means with the processor.
| Translated title of the contribution | A depressive disorder detection method and a system |
|---|---|
| Original language | Latvian |
| Patent number | 15708 |
| IPC | A61B 5/103,G06K 7/00 |
| Priority date | 21/11/22 |
| Publication status | Published - 20 Apr 2023 |
OECD Field of Science
- 1.2 Computer and Information Sciences
- 5.1 Psychology and Cognitive Sciences
Fingerprint
Dive into the research topics of 'A depressive disorder detection method and a system'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver