Title:
TISSUE NODULE DETECTION AND MODEL TRAINING METHOD AND APPARATUS THEREOF, DEVICE AND SYSTEM
Document Type and Number:
WIPO Patent Application WO/2020/094026
Kind Code:
A1
Abstract:
Tissue nodule detection, a model training method and apparatus thereof, a device and a system. Said method comprises: acquiring source domain data and target domain data, the source domain data comprising a source domain image and an image annotation, and the target domain data comprising a target image (S202); performing feature extraction on the source domain image by means of a neural network model to obtain source domain sampling features, performing feature extraction on the target image by means of the neural network model, so as to obtain target sampling features, and determining a training result according to the source domain sampling features (S204); according to the source domain sampling features and the target sampling features, determining distance parameters of the source domain data and the target domain data (S206); according to the training result and the image annotation, determining a loss function value corresponding to the source domain image (S208); and determining a tissue nodule detection model on the basis of the loss function value, the distance parameters and the neural network model (S210). In this way, the detection accuracy can be improved.
Inventors:
CHENG CHEN (CN)
SUN ZHONGQIAN (CN)
CHEN ZHAO (CN)
YANG WEI (CN)
SUN ZHONGQIAN (CN)
CHEN ZHAO (CN)
YANG WEI (CN)
Application Number:
PCT/CN2019/115822
Publication Date:
May 14, 2020
Filing Date:
November 05, 2019
Export Citation:
Assignee:
TENCENT TECH SHENZHEN CO LTD (CN)
International Classes:
G06T7/00; G06N3/04; G06V10/764
Domestic Patent References:
| WO2018020277A1 | 2018-02-01 |
Foreign References:
| CN109523526A | 2019-03-26 | |||
| CN108460415A | 2018-08-28 | |||
| US20180253627A1 | 2018-09-06 |
Other References:
CHU, JINGHUI ET AL.: "Breast Cancer Diagnosis System Based on Transfer Learning and Deep Convolutional Neural Networks", LASER & OPTOELECTRONICS PROGRESS, 31 August 2018 (2018-08-31), pages 1 - 7, XP055704866, ISSN: 1006-4125, DOI: 10.3788/LOP55.081001
Attorney, Agent or Firm:
BEIJING SAN GAO YONG XIN INTELLECTUAL PROPERTY AGENCY CO., LTD. (CN)
Download PDF: