Does the magic of BERT apply to medical code assignment? A quantitative study

Unsupervised pretraining is an integral part of many natural language processing systems, and transfer learning with language models has achieved remarkable results in downstream tasks. In the clinical application of medical code assignment, diagnosis and procedure codes are inferred from lengthy cl...

Full description

Saved in:
Bibliographic Details
Published in:Computers in biology and medicine Vol. 139; p. 104998
Main Authors: Ji, Shaoxiong, Hölttä, Matti, Marttinen, Pekka
Format: Journal Article
Language:English
Published: United States Elsevier Ltd 01.12.2021
Elsevier Limited
Subjects:
BERT
Building codes
Codes
Empirical analysis
Information management
Internal Medicine
Knowledge
Language
Machine learning
Medical code assignment
Medical coding
Natural Language Processing
Neural networks
Other
Pretrained language models
Quantitative analysis
Quantitative study
Semantics
Transfer learning
Medical code assignment
BERT
Quantitative study
Pretrained language models
ISSN:0010-4825, 1879-0534, 1879-0534
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Unsupervised pretraining is an integral part of many natural language processing systems, and transfer learning with language models has achieved remarkable results in downstream tasks. In the clinical application of medical code assignment, diagnosis and procedure codes are inferred from lengthy clinical notes such as hospital discharge summaries. However, it is not clear if pretrained models are useful for medical code prediction without further architecture engineering. This paper conducts a comprehensive quantitative analysis of various contextualized language models' performances, pretrained in different domains, for medical code assignment from clinical notes. We propose a hierarchical fine-tuning architecture to capture interactions between distant words and adopt label-wise attention to exploit label information. Contrary to current trends, we demonstrate that a carefully trained classical CNN outperforms attention-based models on a MIMIC-III subset with frequent codes. Our empirical findings suggest directions for building robust medical code assignment models. •A study on knowledge transfer via mixed-domain and task-adaptive language model.•A thorough comparative study to answer the research questions.•A hierarchical BERT architecture with a label attention mechanism.•Vanilla CNN model with appropriate training can improve the predictive performance.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:0010-4825
1879-0534
1879-0534
DOI:10.1016/j.compbiomed.2021.104998