부산대학교 도서관

Objective: To study the phenotypic characteristics of the combination of disease and syndrome of COPD. Methods: Structured clinical EMRs of inpatients in the respiratory department of the First Affiliated Hospital of Henan University of Chinese Medicine for a total of 10 years from 2010 to 2019 were collected. Patients with COPD in Western medicine diagnosis on admission or discharge were used as the study subjects. A Heterogeneous Medical Record network based on COPD patients was constructed. Then, the K-means clustering algorithm was used to divide the supplemented patient medical record feature matrix into different modules (ie, subtypes), and enrichment analysis was performed on the phenotypes characteristics of different modules to determine the typical clinical phenotypes characteristics of different modules. Results: After the analysis of different phenotypic characteristics of COPD patients, it was found that the phenotypic characteristics of COPD patients were diverse. The prominent manifestation is that COPD patients often have diseases and symptoms other than respiratory system, and more than 70% of them have 2-6 different diseases. Through cluster study and analysis of each subtype of COPD patients revealed that each module has its own unique core characteristics. Conclusions: After the analysis of different phenotypic characteristics of COPD patients, it was found that the phenotypic characteristics of COPD patients were diverse, and each subtype has its own unique core characteristics.