Researchers developed a deep learning model achieving high accuracy in automatic ICD-11 coding of Chinese electronic medical records.
Key Details
- 1The new LA-TextCNN-BiLSTM model leverages MC-BERT and a label attention mechanism.
- 2Evaluated on real-world EMRs, it achieved 83.86% accuracy, 75.82% macro-F1, and 82.83% micro-F1 scores.
- 3Label attention helps the model focus on diagnostically relevant text, improving performance over traditional methods.
- 4Automated ICD-11 coding can streamline hospital workflows and reduce manual coding errors, especially in complex Chinese clinical language settings.
- 5Research was conducted by Peking Union Medical College Hospital & WHO Family of International Classification Collaborating Center.
Why It Matters
Automating ICD-11 coding with advanced NLP can greatly reduce the labor and error burden in extracting structured diagnostic data, which has wide-reaching implications for radiology AI, clinical research, and healthcare efficiency—particularly in linguistically complex environments like Chinese EMRs.
Source
EurekAlert
Related News
•EurekAlert
UVA's $4M NIH Grant Backs AI to Optimize Cardiac MRI Therapies
UVA secures $4 million NIH funding to develop GenStrain, an AI model using MRI data to personalize cardiac pacing therapies.
•EurekAlert
Machine Learning Models Predict Cancer Risk with Routine Health Data
Researchers developed machine learning models using routinely collected health data to predict individual risk for 20 common cancers, outperforming traditional methods for several cancer types.
•EurekAlert
Duke Receives $25M to Lead National Trustworthy AI Hub for Dementia and Aging
Duke University has launched a $25 million national center to make AI for aging and dementia clinically robust and trustworthy.