A probabilistic Model for COPD Diagnosis and Phenotyping Using Bayesian Networks

Introduction: This research was meant to provide a model for COPDdiagnosis and to classify the cases into phenotypes; General COPD,Chronic bronchitis, Emphysema, and the Asthmatic COPD using aBayesian Network (BN).Methods: The model was constructed through developing the BayesianNetwork structure and instantiating the parameters for each of thevariables. In order to validate the achieved results, the same data set wasapplied to a neural network application using the Levenberge- Marquardtalgorithm. Furthermore, a card Diag, a C++ application that enablesgraphical classification of COPD into phenotypes and depicts therelationships of COPD phenotypes was developed.Results: The results showed that a Bayesian Network can be successfullyapplied to develop a probabilistic model for diagnosis and classificationof COPD cases into the corresponding phenotypes.Conclusions: A model that classifies COPD cases into phenotypes ofgeneral COPD, Chronic bronchitis, Emphysema, and Asthmatic COPDwas successfully developed. Moreover, the achieved results also helped torepresent graphical representations of COPD phenotypes and explainedhow the phenotypes relate to each other. It was also observed that COPDis mostly associated with people aged 40 years or older. Overall, smokingis the major cause of COPD.

Keywords:

Bayesian networks , COPD Diagnosis , COPD Phenotypes , Noisy-OR CPD