Title:
HEART DISEASE DIAGNOSIS DEVICE USING DEEP-LEARNING-BASED ELECTROCARDIOGRAM, AND METHOD THEREFOR
Document Type and Number:
WIPO Patent Application WO/2022/014942
Kind Code:
A1
Abstract:
The present invention relates to a heart disease diagnosis device using a deep-learning-based electrocardiogram, and a method therefor. The heart disease diagnosis device using a deep-learning-based electrocardiogram, according to the present invention, includes: an input unit for receiving electrocardiogram data; a feature extraction unit, which analyzes the electrocardiogram data to extract n features for diagnosing heart diseases; a determination model generation unit which generates identification codes for the n features according to the diagnosis names of the heart diseases, and which generate n determination models that determine each feature by learning the generated identification codes; a diagnosis model generation unit for generating diagnosis models that derive the diagnosis names of the heart diseases by learning each value determined by the determination models; and a control unit for outputting diagnosis results for the electrocardiogram data to be diagnosed using the determination models and the diagnosis models. According to the present invention, a deep learning algorithm is used so that the electrocardiogram data for each feature is learned, and the trained models are used so that the diagnosis names of the heart diseases are derived, and thus diagnosis accuracy is improved, and the reasons for diagnosis are presented together so that diagnosis reliability can be improved.
Inventors:
KWON JOON MYOUNG (KR)
Application Number:
PCT/KR2021/008626
Publication Date:
January 20, 2022
Filing Date:
July 07, 2021
Export Citation:
Assignee:
BODYFRIEND CO LTD (KR)
MEDICAL AI (KR)
MEDICAL AI (KR)
International Classes:
A61B5/349; A61B5/00; A61B5/363; G06N3/08; G16H50/20
Foreign References:
| KR20060117546A | 2006-11-17 | |||
| KR20190114694A | 2019-10-10 | |||
| KR101538916B1 | 2015-07-24 | |||
| KR20200041697A | 2020-04-22 | |||
| KR20200084561A | 2020-07-13 |
Download PDF: