Stepping Up the Personalized Approach in COPD with Machine Learning

Abstract

There is increasing interest in the application of artificial intelligence (AI) and machine learning (ML) in all fields of medicine to facilitate greater personalisation of management. ML could be the next step of personalized medicine in chronic obstructive pulmonary disease (COPD) by giving the exact risk (risk for exacerbation, death, etc.) of every patient (based on his/her parameters like lung function, clinical data, demographics, previous exacerbations, etc.), thus providing a prognosis/risk for the specific patient based on individual characteristics (individual approach). ML algorithm might utilise some traditional risk factors along with some others that may be location-specific (e.g. the risk of exacerbation thatmay be related to ambient pollution but that could vary massively between different countries, or between different regions of a particular country). This is a step forward from the commonly used assignment of patients to a specific group for which prognosis/risk data are available (group approach).