Title:
MEDICAL IMAGE CLASSIFICATION PROCESSING SYSTEM AND METHOD BASED ON ARTIFICIAL INTELLIGENCE
Document Type and Number:
WIPO Patent Application WO/2019/052063
Kind Code:
A1
Abstract:
A medical image classification processing system and method based on artificial intelligence. The method comprises the steps of: acquiring medical image information from a medical image information library (S21); carrying out classification processing on the medical image information based on different parts of a human body (S22); establishing a medical evaluation model of various parts of the human body based on the classified medical image information (S23); carrying out training processing on the medical evaluation model of various parts of the human body (S24), and saving medical model parameters (S25); receiving a medical examination image of a patient from a medical image collection terminal (S26); recognizing an examination part from the medical examination image of the patient, and acquiring, according to the examination part, the corresponding medical model parameters from a storage unit (S27); and sending the medical examination image and the corresponding medical model parameters of the patient to a doctor's workstation (S28). The present invention can carry out feature classification on a medical image according to various parts of a human body and can train a medical model for a doctor to carry out medical examination reference, thereby facilitating an improvement in the efficiency and accuracy of a medical examination.
Inventors:
YAO YUDONG (CN)
QIAN WEI (CN)
ZHENG BIN (CN)
MA HE (CN)
QI SHOULIANG (CN)
ZHAO MINGFANG (CN)
QIAN WEI (CN)
ZHENG BIN (CN)
MA HE (CN)
QI SHOULIANG (CN)
ZHAO MINGFANG (CN)
Application Number:
PCT/CN2017/116664
Publication Date:
March 21, 2019
Filing Date:
December 15, 2017
Export Citation:
Assignee:
ANYCHECK INFORMATION TECH CO LTD (CN)
International Classes:
G16H30/20; G06K9/62; G16H50/20
Foreign References:
CN106777953A | 2017-05-31 | |||
CN106355023A | 2017-01-25 | |||
CN104866727A | 2015-08-26 | |||
CN106228180A | 2016-12-14 |
Download PDF: