Title:
DEEP LEARNING METHOD AND DEVICE FOR ANALYSIS OF HIGH-DIMENSIONAL MEDICAL DATA
Document Type and Number:
WIPO Patent Application WO/2018/103538
Kind Code:
A1
Abstract:
Disclosed in the present application are a deep learning method and device for analysis of high-dimensional medical data. The method comprises: reading high-dimensional medical data and, by analyzing data properties, preprocessing the high-dimensional medical data into a data format acceptable to a feature extraction model; selecting different feature extraction models according to different preprocessing manners, and inputting the preprocessed data into the selected feature extraction model for feature extraction to obtain a corresponding feature vector; and inputting the corresponding feature vector into a recurrent neural network model for training to obtain a final deep learning model used for medical data analysis. The method of the present invention can enhance the accuracy of analyzing high-dimensional medical data based on a deep learning model. Also provided in the present invention is a deep learning device for analysis of high-dimensional medical data, comprising: a preprocessing module, a feature extraction module, and a model training module.
Inventors:
ZHANG RONGGUO (CN)
CHEN KUAN (CN)
CHEN KUAN (CN)
Application Number:
PCT/CN2017/112606
Publication Date:
June 14, 2018
Filing Date:
November 23, 2017
Export Citation:
Assignee:
INFERVISION (CN)
International Classes:
G06F19/00
Foreign References:
CN106097335A | 2016-11-09 | |||
CN105760507A | 2016-07-13 | |||
CN106096616A | 2016-11-09 | |||
US5852684A | 1998-12-22 |
Attorney, Agent or Firm:
GE CHENG & CO., LTD. (CN)
Download PDF: