Internal structure of a disease name and its application for ICD coding.

Area of Science:

• Medical Informatics
• Health Information Management
• Computational Linguistics

Background:

• International Classification of Diseases (ICD) coding is complex and prone to variability based on coder expertise.
• Existing coding systems, like the Japanese Standard Disease-Code Master, can lead to post-coordination issues, altering original ICD codes.
• These alterations complicate accurate disease classification and data analysis.

Purpose of the Study:

• To propose and evaluate a novel approach for correcting ICD codes by analyzing the internal structure of disease names.
• To enhance the precision and reliability of ICD coding through automated analysis.
• To address post-coordination problems inherent in current disease coding practices.

Main Methods:

• Development of an internal structure analyzer for disease names.
• Utilizing the analyzed internal structures to correct potentially erroneous ICD codes.
• Evaluation of the analyzer's accuracy in assisting precise ICD-coding tasks.

Main Results:

• The developed internal structure analyzer achieved a high accuracy rate of 83.7%.
• The analyzed output proved beneficial for improving the precision of ICD-coding.
• The proposed method demonstrated potential in mitigating post-coordination problems.

Conclusions:

• Analyzing internal disease name structures offers a promising solution for enhancing ICD-coding accuracy.
• The developed analyzer can serve as a valuable tool for medical coders and health information professionals.
• This approach has the potential to improve the integrity of medical coding and subsequent data analysis.