Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Medical named entity recognition model training method, recognition method and federal learning system

A technology of named entity recognition and model training, which is applied in neural learning methods, character and pattern recognition, and biological neural network models. Data Leakage Risks, Good Controls, and Expected Effects

Active Publication Date: 2022-03-11
北京智源人工智能研究院
View PDF5 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In order to solve the problems of low medical data sharing and model learning that do not consider the localization of medical vocabulary descriptions in the prior art, the present invention provides the following technical solutions

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Medical named entity recognition model training method, recognition method and federal learning system
  • Medical named entity recognition model training method, recognition method and federal learning system
  • Medical named entity recognition model training method, recognition method and federal learning system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0058] Such as figure 1 As shown, the first aspect of the present invention provides a medical named entity recognition model training method. The method may be performed by a medical institution node, and the medical institution node may refer to an institution performing model training based on a distributed server in the federated learning system, or may refer to the distributed server itself in the federated learning system, specifically including:

[0059] S101. Receive a global model for identifying medical named entities sent by a central server.

[0060] The overall federated learning framework of the present invention is for example figure 2 As shown, including central server and distributed server. Among them, the distributed server corresponds to the medical named entity recognition system distributed in various medical institutions, and is used for local model training and prediction, while the central server is responsible for initial model training, model deli...

Embodiment 2

[0083] Such as Figure 4 As shown, the second aspect of the present invention provides a medical named entity recognition model training method. The method can be executed by the central server, and specifically includes:

[0084] S201. Distribute the global model for identifying medical named entities to each medical institution node in the federated learning system, so that each of the medical institution nodes trains the global model based on their respective local medical text annotation data, and calculates The corresponding gradient data.

[0085] S202. Receive gradient data respectively sent by each of the medical institution nodes, and train the global model based on each of the gradient data to obtain a new global model.

[0086] S203. If the new global model is currently converged, distribute the converged global model to each of the medical institution nodes in the federated learning system, so that each of the medical institution nodes is based on their correspon...

Embodiment 3

[0096] Such as Figure 5 As shown, the present invention provides a localized medical named entity recognition model training method in a third aspect, the method is executed by a medical institution node, and specifically includes:

[0097] S301. Obtain medical data.

[0098] In S301, the medical institution node can perform data cleaning, invalid data screening, and data format standardization processing on the medical data based on preset preprocessing rules, so as to further improve the execution efficiency and accuracy of S302.

[0099] S302. Input the medical data into a localized medical named entity recognition model, so that the localized medical named entity recognition model outputs a medical named entity recognition result corresponding to the medical data.

[0100] Wherein, the localized medical named entity recognition model is obtained in advance based on the medical named entity recognition model training method described in the first aspect.

[0101] After S...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a medical named entity recognition model training method, a medical named entity recognition method and a federal learning system. The medical named entity recognition model training method comprises the following steps: receiving a global model which is sent by a central server and is used for recognizing a medical named entity; training a global model based on local medical text annotation data, and calculating to obtain corresponding gradient data; sending the gradient data to a central server to enable the central server to train a global model based on each gradient data received by the federal learning system to obtain a new global model, and if the new global model is currently converged, distributing the converged global model; receiving a converged global model; and performing localized fine tuning processing on the converged global model based on a local prompt template to form a localized medical named entity recognition model. According to the technical scheme, privacy protection of the medical data and local personalization of the medical named entity recognition model are realized.

Description

technical field [0001] The invention relates to the technical field of natural language processing, in particular to a medical named entity recognition model training method, a medical named entity recognition method and a federated learning system. Background technique [0002] Named Entity Recognition (MNER) in the medical field is the basis for the construction of medical knowledge graphs and medical big data. It is an important basis for the intelligent analysis of medical records and the realization of medical intelligence. It is of great value for applications such as medical intelligence and auxiliary diagnosis. [0003] Existing medical named entity recognition technologies usually rely on large-scale labeling data. However, due to the privacy of medical data, it is difficult and costly to obtain a large amount of medical named entity labeling data. At present, it is possible to produce high-quality labeled medical professionals. Very scarce; and data privacy risks ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F40/295G06F40/186G06K9/62G06N3/08
CPCG06F40/295G06F40/186G06N3/08G06F18/214
Inventor 安波
Owner 北京智源人工智能研究院
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com