Method for extracting clinical medical information based on neural network

A technology of clinical medicine and neural network, applied in the field of clinical medical information extraction based on neural network, can solve problems such as high cost and achieve the effect of solving the field adaptation

Inactive Publication Date: 2018-06-19
XI AN JIAOTONG UNIV
View PDF4 Cites 26 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, labeling medical data takes a lot of time fo

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
  • Method for extracting clinical medical information based on neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0026] Embodiment 1: It is mainly used to extract time series information in clinical medical texts. Both the training data and the test data are selected from the bowel cancer text data in the THYME corpus.

[0027] Step 1: Firstly, word segmentation is performed on the training text and the test text, and the training text obtained after word segmentation is marked with BIO tags. Among them, B represents that this word is the beginning of an information sequence, I represents that this word is an intermediate word forming an information sequence, and O represents that this word does not belong to any information sequence.

[0028] Step 2: Construct the corresponding initial character vector table for 24 English letters and other common characters, and use the biomedical articles in the PubMed database as the corpus to construct the initial word vector. Based on the text after word segmentation in step 1, the initial word vector corresponding to each word and the initial cha...

Embodiment 2

[0042] Embodiment 2: It is mainly used to extract event information in clinical medical texts, such as craniotomy, bleed, cancer. In order to demonstrate the ability of the present invention to adapt to the field, the training text is selected from the intestinal cancer text data in the THYME corpus, and the test text is selected from THYME Brain cancer text data in database. The training text is about bowel cancer, and the test text is about brain cancer, so it is necessary to solve the problem of domain adaptation. For example, Chemotherapy (chemotherapy) is an event in both the source domain and the target domain, while craniotomy (craniotomy) is an event only in the target domain.

[0043] Step 1: First, use the Stanford natural language processing tool to perform word segmentation on the training text and test text, and mark the training text obtained after word segmentation with BIO tags. Among them, B represents that this word is the beginning of an event sequence, I r...

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 method for extracting clinical medical information based on a neural network. Generally, a medical text has much professional vocabulary, uncommon vocabulary and time expressions composed of numbers and characters, but character vectors obtained by a convolutional neural network may contain the morphological information of words, so the problem can be solved very well. Further, a bidirectional LSTM can capture context information well. In addition, the method of using the neural network avoids a process of artificially designing features in machine learning, and can well solve domain adaptation. The method of the invention achieves good results in different data fields, and can extract information with practical values and research significance from massive medical data efficiently and accurately.

Description

technical field [0001] The invention relates to the field of biomedical text natural language processing, in particular to a method for extracting clinical medical information based on a neural network. Background technique [0002] In the context of big data and "smart medical care", text mining and information extraction technology in the medical field has become the "top priority" that researchers have focused on in recent years. Extracting medical entity information in medical texts, such as time and events, is one of the important tasks in medical big data processing. However, unstructured medical text data expressed in natural language has the characteristics of huge data volume, complex structure, and fast generation speed. It is very difficult for relevant researchers to quickly and accurately obtain valuable knowledge and information from a large amount of text. . Therefore, how to efficiently and intelligently extract medical knowledge information with practical ...

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
IPC IPC(8): G16H50/70G06F17/27
CPCG06F40/279G06F40/289
Inventor 李辰王轩龙雨李质婧
Owner XI AN JIAOTONG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap