Mining Patterns of Disease Progression: A Topic-Model-Based Approach.

Area of Science:

• Medical Informatics
• Computational Biology
• Data Mining

Background:

• Understanding disease progression is vital for effective clinical decision-making.
• Electronic medical records (EMR) contain valuable data but are often noisy and unorganized.
• Extracting meaningful disease patterns typically requires significant manual effort from medical professionals.

Purpose of the Study:

• To develop an automated approach for discovering disease progression patterns from EMR data.
• To improve traditional sequential pattern mining (SPM) algorithms by incorporating topic modeling.
• To reduce the noise and manual effort involved in identifying clinical patterns.

Main Methods:

• A novel topic-model-based sequential pattern mining (SPM) approach was developed.
• Diagnosis sequences were filtered and grouped based on automatically detected clinical topics.
• The method was applied to real-world electronic medical record (EMR) data.

Main Results:

• The proposed approach successfully identified meaningful disease progression patterns.
• The method demonstrated a significant reduction in noise compared to traditional SPM.
• Clinical topics helped in filtering and grouping diagnosis sequences effectively.

Conclusions:

• The topic-model-based SPM approach enhances the discovery of disease progression patterns from EMR data.
• This method offers a less noisy and more efficient way to identify clinically relevant patterns.
• It reduces the need for manual data manipulation by medical experts.