Clinical data mining of phenotypic network in angina pectoris of coronary heart disease

Area of Science:

• Computational Medicine
• Traditional Chinese Medicine
• Network Analysis

Background:

• Coronary Heart Disease (CHD) is a major health concern in China, leading to significant morbidity and mortality.
• Traditional Chinese Medicine (TCM) diagnosis for CHD has historically relied on experienced practitioners.
• Objective diagnosis of CHD based on phenotypes in TCM requires advanced analytical methods.

Purpose of the Study:

• To develop and validate novel algorithms for the objective diagnosis of Coronary Heart Disease (CHD) within the framework of Traditional Chinese Medicine (TCM).
• To establish a standardized scale for TCM syndromes to enable automated patient diagnosis based on phenotypic data.
• To investigate the alterations in core TCM syndromes when CHD co-occurs with other diseases and to characterize distinct phenotypic spectra.

Main Methods:

• Proposed four Mutual Information (MI)-based association algorithms for analyzing CHD phenotype networks.
• Developed a syndrome scale for automated diagnosis generation from patient phenotypes.
• Compared changes in core syndromes for CHD combined with other diseases, analyzing phenotype spectra.

Main Results:

• Successfully developed and applied MI-based algorithms to analyze CHD phenotype networks.
• Established a functional syndrome scale for automated TCM diagnosis of CHD.
• Identified and presented distinct phenotype spectra and changes in core syndromes associated with co-morbidities of CHD.

Conclusions:

• The proposed MI-based algorithms and syndrome scale offer a novel, objective approach to CHD diagnosis in TCM.
• Automated diagnosis based on phenotypes can improve the consistency and accessibility of TCM for CHD.
• Understanding syndrome alterations in co-morbidities provides insights into complex disease patterns within TCM framework.