Multi-task question and answer driven medical entity relationship extraction method
An entity relationship, multi-task technology, applied in neural learning methods, instruments, biological neural network models, etc., can solve the problems of inaccurate extraction of medical entities, difficult to solve the problem of polysemy and nesting of electronic medical records words, etc. The effect of improving the accuracy of acquisition
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0055] Give a symptom description text: the distal end of the left vertebral artery is occluded, the diameter of the left vertebral artery is thin (physiological), and the right liver is hyperechoic—hemangioma? .
[0056] Note: "Occlusion of the distal end of the left vertebral artery" and "Slender left vertebral artery" are symptom entities, "hemangioma" is a disease entity, and the relationship between the two is "disease and examination"
[0057] The specific extraction process is as follows:
[0058] (1) The entity set in this description is {the distal end of the left vertebral artery is occluded, the entire diameter of the left vertebral artery is thin, and hemangioma}. There are two types of entities "Disease" and "Symptom". Construct two questions based on entity types: "What symptoms are in the text?" and "What diseases are in the text?"
[0059] (2) Taking the first entity as an example, construct a new text "[CLS] What are the symptoms in the text? [SEP] The dist...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
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