LATA – Label attention transformer architectures for ICD-10 coding of unstructured clinical notes

Veena Mayya, Sowmya S. Kamath, Vijayan Sugumaran

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Effective code assignment for patient clinical records in a hospital plays a significant role in the process of standardizing medical records, mainly for streamlining clinical care delivery, billing, and managing insurance claims. The current practice employed is manual coding, usually carried out by trained medical coders, making the process subjective, error-prone, inexact, and time-consuming. To alleviate this cost-intensive process, intelligent coding systems built on patients’ structured electronic medical records are critical. Classification of medical diagnostic codes, like ICD-10, is widely employed to categorize patients’ clinical conditions and associated diagnoses. In this work, we present a neural model LATA, built on Label Attention Transformer Architectures for automatic assignment of ICD-10 codes. Our work is benchmarked on the CodiEsp dataset, a dataset for automatic clinical coding systems for multilingual medical documents, used in the eHealth CLEF 2020-Multilingual Information Extraction Shared Task. The experimental results reveal that the proposed LATA variants outperform their basic BERT counterparts by 33-49% in terms of standard metrics like precision, recall, F1-score and mean average precision. The label attention mechanism also enables direct extraction of textual evidence in medical documents that map to the clinical ICD-10 diagnostic codes.

Original languageEnglish
Title of host publication2021 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2021
EditorsJennifer Hallinan, Madhu Chetty, Gonzalo Ruz Heredia, Adrian Shatte, Suryani Lim
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781665401128
DOIs
Publication statusPublished - 2021
Event2021 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2021 - Virtual, Online, Australia
Duration: 13-10-202115-10-2021

Publication series

Name2021 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2021

Conference

Conference2021 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2021
Country/TerritoryAustralia
CityVirtual, Online
Period13-10-2115-10-21

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Information Systems
  • Computational Mathematics
  • Health Informatics

Fingerprint

Dive into the research topics of 'LATA – Label attention transformer architectures for ICD-10 coding of unstructured clinical notes'. Together they form a unique fingerprint.

Cite this