Related Experiment Video
Updated: Dec 11, 2025
04:17
DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning
Published on: May 10, 2024
1.3K
Medical Information Extraction in the Age of Deep Learning.
1Jena University Language & Information Engineering (JULIE) Lab, Friedrich-Schiller-Universität Jena, Jena, Germany.
Yearbook of Medical Informatics
|August 22, 2020
Summary
Deep learning methods now dominate medical information extraction, outperforming traditional machine learning. This shift impacts various medical informatics fields, requiring adaptive strategies for challenges like limited data.
Area of Science:
- Medical Informatics
- Natural Language Processing
- Artificial Intelligence
- Deep Learning
Background:
- The field of Natural Language Processing (NLP) has shifted from symbolic methods to Deep Learning (DL) based on distributed representations.
- This trend is increasingly reflected in medical NLP, including medical Information Extraction (IE).
Purpose of the Study:
- To survey recent developments in medical Information Extraction (IE) over the past three years.
- To focus on the methodological paradigm shift from standard Machine Learning (ML) to Deep Neural Networks (DNNs).
- To describe DNN applications in named entity recognition and relation extraction for diseases and drugs.
Main Methods:
- Literature search from 2017 to early 2020.
- Covered publications from medicine, medical informatics, NLP, and AI communities.
Main Results:
- Deep Learning (DL) based approaches significantly outperform non-DL methods in medical IE.
- Overwhelming experimental evidence supports the superiority of DL methods.
- Challenges remain due to small, access-limited corpora and specialized medical language.
Conclusions:
- The paradigm shift from ML to DNNs fundamentally changes medical NLP and Information Extraction.
- This transformation is expected to influence broader areas of medical informatics beyond NLP.
