Title:
METHOD AND SYSTEM FOR AUTOMATICALLY RECOGNIZING IMAGE OF PRIMARY TUMOR OF NASOPHARYNGEAL CARCINOMA
Document Type and Number:
WIPO Patent Application WO/2022/001237
Kind Code:
A1
Abstract:
A method and system for automatically recognizing an image of a primary tumor of nasopharyngeal carcinoma, based on a CT three-dimensional image of a test subject, supplemented by an MR series of three-dimensional images, constituting one multimodal input for a semantic segmentation network model, and implementing the recognition of a primary tumor of nasopharyngeal carcinoma on CT. The combination of the CT three-dimensional image with the MR series of three-dimensional images effectively increases the quality of input data, learns global information and detailed information of a high-resolution image, effectively increases the prognosis accuracy and generalization capability of the semantic segmentation network model, and provides flexibility at an input end and at an output end, thus effectively increasing the work efficiency of a health professional.
Inventors:
WEI JUN (CN)
ZHU DEMING (CN)
XIE PEILIANG (CN)
ZHU DEMING (CN)
XIE PEILIANG (CN)
Application Number:
PCT/CN2021/083154
Publication Date:
January 06, 2022
Filing Date:
March 26, 2021
Export Citation:
Assignee:
PERCEPTION VISION MEDICAL TECH CO LTD (CN)
PERCEPTION VISION DATA TECH CO LTD (CN)
PERCEPTION VISION DATA TECH CO LTD (CN)
International Classes:
G06T7/11
Foreign References:
| CN108257134A | 2018-07-06 | |||
| CN111091560A | 2020-05-01 | |||
| CN111784706A | 2020-10-16 | |||
| CN1952980A | 2007-04-25 | |||
| CN110211079A | 2019-09-06 | |||
| US8588498B2 | 2013-11-19 |
Attorney, Agent or Firm:
BEIJING XINGZHI XIANGDA INTELLECTUAL PROPERTY AGENCY CO., LTD. (CN)
Download PDF: