Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020231007 - MEDICAL EQUIPMENT LEARNING SYSTEM

Publication Number WO/2020/231007
Publication Date 19.11.2020
International Application No. PCT/KR2020/004611
International Filing Date 06.04.2020
IPC
G16H 50/70 2018.01
GPHYSICS
16INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
50ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
70for mining of medical data, e.g. analysing previous cases of other patients
G16H 15/00 2018.01
GPHYSICS
16INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
15ICT specially adapted for medical reports, e.g. generation or transmission thereof
G16H 50/20 2018.01
GPHYSICS
16INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
50ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
20for computer-aided diagnosis, e.g. based on medical expert systems
G16H 50/80 2018.01
GPHYSICS
16INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
50ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
80for detecting, monitoring or modelling epidemics or pandemics, e.g. flu
G16H 70/00 2018.01
GPHYSICS
16INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
70ICT specially adapted for the handling or processing of medical references
Applicants
  • (주)비주얼터미놀로지 VISUAL TERMINOLOGY INC. [KR]/[KR]
Inventors
  • 최병관 CHOI, Byung Kwon
Agents
  • 김종석 KIM, Jong Seok
Priority Data
10-2019-005569813.05.2019KR
Publication Language Korean (KO)
Filing Language Korean (KO)
Designated States
Title
(EN) MEDICAL EQUIPMENT LEARNING SYSTEM
(FR) SYSTÈME D'APPRENTISSAGE D'UN ÉQUIPEMENT MÉDICAL
(KO) 의료기계 학습 시스템
Abstract
(EN)
The present invention relates to a medical equipment learning system and, particularly, to a medical artificial intelligence learning system in which medical data expressed in characters is converted into image data, and the image data is provided so as to be used for learning of a medical artificial intelligence. A medical equipment learning system according to the present invention comprises: a data extraction module (100) for collecting and then extracting character data (120) from medical data (110); a visualization module (200) for generating image data (210), which is visualization data, by using the character data (120) extracted by the data extraction module (100); a preprocessing module (300) for generating an input data set (310) for execution of machine learning on the basis of the visualization data; a learning module (400) for executing machine learning on the input data set (310) generated by the preprocessing module (300); a prediction module (500) for predicting a disease when new image data is input, on the basis of a result learned by the learning module (400); and a storage module (600) provided so as to store and check data of each module. The data extraction module (100) is a preconfigured two-dimensional or three-dimensional model. The visualization module (200) may change one or more among the color, brightness or transparency, pattern, and texture of the visualization data according to a name of a disease, severity of the disease, a degree of being acute/chronic and a degree of being malignant, various test results, function test results, and data results extracted from the equipment. The preprocessing module (300) adds a blood test result and a function test result to the image data (210) generated by the visualization module (200). Further, the preprocessing module (300) generates the input data set (310) by processing the visualization data.
(FR)
La présente invention concerne un système d'apprentissage d'un équipement médical et, en particulier, un système d'intelligence artificielle médicale par apprentissage automatique, dans lequel des données médicales exprimées en caractères sont converties en données d'image, et les données d'image sont fournies pour être utilisées aux fins d'apprentissage d'une intelligence artificielle médicale. Un système d'apprentissage d'équipement médical selon la présente invention comprend : un module d'extraction de données (100) permettant de collecter puis d'extraire des données de caractères (120) à partir de données médicales (110) ; un module de visualisation (200) permettant de générer des données d'image (210) qui sont des données de visualisation, en ayant recours aux données de caractère (120) extraites par le module d'extraction de données (100) ; un module de prétraitement (300) permettant de générer un ensemble de données d'entrée (310) aux fins d'exécution d'un apprentissage machine sur la base des données de visualisation ; un module d'apprentissage (400) permettant d'exécuter un apprentissage machine sur l'ensemble de données d'entrée (310) généré par le module de prétraitement (300) ; un module de prédiction (500) permettant de prédire une maladie lorsque de nouvelles données d'image sont entrées, sur la base d'un résultat appris par le module d'apprentissage (400) ; et un module de stockage (600) prévu pour stocker et vérifier les données de chaque module. Le module d'extraction de données (100) est un modèle bidimensionnel ou tridimensionnel préconfiguré. Le module de visualisation (200) permet de modifier la couleur et/ou la luminosité et/ou la transparence, le motif et la texture des données de visualisation selon un nom de maladie, une gravité de la maladie, un degré aigu/chronique et un degré de malignité, divers résultats de test, des résultats d'examens fonctionnels et des résultats de données extraits de l'équipement. Le module de prétraitement (300) ajoute un résultat d'examen sanguin et un résultat d'examen fonctionnel aux données d'image (210) générées par le module de visualisation (200). En outre, le module de prétraitement (300) génère l'ensemble de données d'entrée (310) en traitant les données de visualisation.
(KO)
본 발명은 의료기계 학습 시스템에 관한 것으로, 문자로 표현된 의료 데이터를 영상데이터로 바꾸고 이를 이용하여 의료 인공지능의 학습에 이용할 수 있도록 마련되어 의료용 인공 지능 학습 시스템에 관한 것이다. 본 발명에 따른 의료기계 학습시스템은 의료데이터(110)에서 문자형데이터(120)를 수집 후 추출하는 데이터추출 모듈(100); 상기 데이터추출 모듈(100)에서 추출된 문자형데이터(120)를 이용하여 시각화데이터인 이미지형데이터(210)를 생성하는 시각화 모듈(200); 상기 시각화데이터를 바탕으로 머신러닝을 실행하기 위한 입력데이터셋(310)을 생성하는 전처리 모듈(300); 상기 전처리 모듈(300)에 의해 생성된 입력데이터셋(310)에 기계학습을 실행하는 학습 모듈(400); 상기 학습 모듈(400)에서 학습된 결과를 바탕으로 새로운 이미지형데이터가 입력되었을 때 질병을 예측하는 예측 모듈(500); 및 각 모듈의 데이터를 저장하고 확인할 수 있도록 마련된 저장 모듈(600);을 포함하는 것을 특징으로 한다. 상기 데이터추출 모듈(100)은 기설정된 2차원 또는 3차원 모델인 것을 특징으로 한다. 상기 시각화 모듈(200)은 질병의 이름, 질병의 중증도, 급만성 정도 및 악성 정도, 각종 검사 결과, 기능검사 결과 및 기계에서 추출된 데이터 결과에 따라 상기 시각화데이터의 색깔, 밝기 또는 투명도, 무늬 및 텍스쳐 중 어느 하나 이상을 변경할 수 있는 것을 특징으로 한다. 상기 전처리 모듈(300)은 상기 시각화 모듈(200)에 의해 생성된 이미지형데이터(210)에 혈액 검사 결과 및 기능검사 결과를 추가하는 것을 특징으로 한다. 또한, 상기 전처리 모듈(300)은 상기 시각화데이터를 가공하여 상기 입력데이터셋(310)을 생성하는 것을 특징으로 한다.
Latest bibliographic data on file with the International Bureau