Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Oct 22, 2025

DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning

DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning

Published on: May 10, 2024

Automatic ICD-10 Coding and Training System: Deep Neural Network Based on Supervised Learning.

Pei-Fu Chen^1,2, Ssu-Ming Wang¹, Wei-Chih Liao¹

¹Graduate Institute of Biomedical Electronics and Bioinformatics, National Taiwan University, Taipei, Taiwan.

JMIR Medical Informatics

|August 31, 2021

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Classifying American Society of Anesthesiologists Physical Status With a Low-Rank-Adapted Large Language Model: Development and Validation Study.

Journal of medical Internet research·2026

Same author

Effects of Heat Adaptation Behaviors on Resting Heart Rate Response to Summer Temperatures in Older Adults: Wearable Device Panel Study.

JMIR mHealth and uHealth·2025

Same author

Using Wearable Device and Machine Learning to Predict Mood Symptoms in Bipolar Disorder: Development and Usability Study.

JMIR medical informatics·2025

Same author

Clinical and Economic Evaluation of a Real-Time Chest X-Ray Computer-Aided Detection System for Misplaced Endotracheal and Nasogastric Tubes and Pneumothorax in Emergency and Critical Care Settings: Protocol for a Cluster Randomized Controlled Trial.

JMIR research protocols·2025

Same author

Predicting 30-Day Postoperative Mortality and American Society of Anesthesiologists Physical Status Using Retrieval-Augmented Large Language Models: Development and Validation Study.

Journal of medical Internet research·2025

Same author

Estimating sample means and standard deviations from the log-normal distribution using medians and quartiles: evaluating reporting requirements for primary and secondary endpoints of meta-analyses in anesthesiology.

Canadian journal of anaesthesia = Journal canadien d'anesthesie·2025

Same journal

Selecting, Scaling, and Measuring the Value of Ambient AI in a Nonacademic Health System: Multiphase Pilot Study.

JMIR medical informatics·2026

Same journal

Prediction of Early Hospital Admission (≤24 Hours) After Stroke Using Machine Learning and Deep Learning: Multicenter Study From China.

JMIR medical informatics·2026

Same journal

Assessing the Feasibility and Acceptability of Implementing a Preclinic Vital Signs Assessment in Primary Care: Cross-Sectional Pilot Study.

JMIR medical informatics·2026

Same journal

Candidate Passive Sensor Suite Technologies for Tactical Combat Casualty Care Environments: Comparative Assessment Study.

JMIR medical informatics·2026

Same journal

Relevance of the uMap Collaborative Platform as Support for Choropleth Mapping: A Traffic‒Light Statistical Signal Atlas of All-Cause Mortality-First French Lockdown.

JMIR medical informatics·2026

Same journal

Ambient AI Scribe Implementation in an Ambulatory Setting in a Single Medical Group: Prospective Study.

JMIR medical informatics·2026

See all related articles

This summary is machine-generated.

This study developed a deep learning model for International Classification of Diseases (ICD)-10 auto-coding, significantly improving coding accuracy (F1-score). While the model enhanced performance, it did not reduce the time spent by human coders.

Area of Science:

Medical Informatics
Artificial Intelligence in Healthcare
Natural Language Processing

Background:

International Classification of Diseases (ICD) codes are crucial for medical systems and billing.
Manual ICD coding is labor-intensive, time-consuming, and complex, especially after the ICD-9 to ICD-10 transition.
Deep learning and NLP approaches are being explored to assist disease coders.

Purpose of the Study:

To construct a deep learning model for automated ICD-10 coding using free-text medical notes.
To enhance accuracy and reduce human effort in the disease coding process.
To develop a tool for automatic diagnosis and procedure code determination.

Main Methods:

Utilized National Taiwan University Hospital diagnosis records and NLP techniques (GloVe, Word2Vec, ELMo, BERT, attention RNN).

Keywords:

International Classification of Diseases Recurrent Neural Network deep learning natural language processing text classification

More Related Videos

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images

Published on: October 27, 2023

Related Experiment Videos

Last Updated: Oct 22, 2025

DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning

DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning

Published on: May 10, 2024

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images

Published on: October 27, 2023

Implemented a deep neural network architecture for ICD-10 auto-coding.

Introduced an attention mechanism for keyword extraction and visualization to aid training.

Main Results:

Achieved F1-scores of 0.715 for ICD-10-CM and 0.618 for PCS using BERT embeddings with a GRU model.
The model, integrated into a web service, increased the median F1-score for coders from 0.832 to 0.922 (P<.05).
Coding time was not significantly reduced despite the improved accuracy.

Conclusions:

The proposed deep learning model significantly enhances ICD-10 coding accuracy.
The model aids in keyword extraction and visualization for training purposes.
The model improves F1-score but does not decrease the time required for manual coding.