Extracting Significant Comorbid Diseases from MeSH Index of PubMed

Area of Science:

• Medical Informatics
• Computational Biology
• Bioinformatics

Background:

• Understanding disease associations and comorbidities is crucial for effective patient care.
• Comorbidity and multimorbidity, though distinct, are often intertwined in clinical studies.
• Genetic or molecular factors can underlie comorbid conditions.

Purpose of the Study:

• To explore text mining for insights into disease associations and comorbidities.
• To present a protocol for analyzing interrelationships between health conditions.
• To leverage biological vocabulary resources for enhanced analysis.

Main Methods:

• Utilizing text mining techniques to analyze patient health information.
• Applying association rule learning to identify comorbidity patterns.
• Employing biological vocabulary resources such as the Unified Medical Language System (UMLS) and Medical Subject Headings (MeSH).

Main Results:

• Text mining can reveal complex disease associations and comorbidities.
• Association rules offer a method for novel knowledge prediction.
• The proposed protocol facilitates extensive analysis of disease interrelationships.

Conclusions:

• Text mining provides a robust framework for understanding disease associations.
• Integrating diverse data sources and resources like UMLS/MeSH enhances analytical capabilities.
• The presented protocol supports a more rational approach to studying comorbidity and multimorbidity.