Abstract
Chronic Obstructive Pulmonary Disease (COPD) is among the major global concerns of mortality and morbidity, affecting millions of people worldwide. It is a progressive pulmonary disease including symptoms like chest tightness, wheezing, coughing, and shortness of breath. COPD is a multifactorial respiratory condition influenced by various risk factors. These risk factors include tobacco smoking, environmental air pollution, occupational exposures, occupational exposures, genetic factors, and other comorbidities. Another factor that appears to play important role in the development and recurrence of COPD is lung microbiota dysbiosis. This dysbiosis is believed to contribute to chronic inflammation, impaired host defense mechanisms, increased mucus production, and treatment response in COPD patients. To delve deeper, the research in the field is oriented toward understanding the association between COPD and microbial Dysbiosis. The biomarkers associated with the microbiome are being used for diagnosis of the disease. This study brings forth to readers the application of Machine learning (ML) and Deep learning (DL) tools in the detection of the disease by extracting meaningful information from clinically relevant COPD data generated by various diagnostic techniques such as CT scans, spirometry, acute exacerbations, and several other COPD risk factors. Although ML and DL techniques have been applied extensively in the literature for the prediction of COPD readmission, microbial dysbiosis data has not been used for this prediction. The focus of this study is to highlight the latest research related to microbial dysbiosis in COPD and explore the possibility of applying AI tools for novel diagnostics and therapeutic strategies.
Keywords: COPD, Readmission, Machine Learning, Microbial Dysbiosis.