Abstract
Abstract
The integration of Artificial Intelligence (AI) into healthcare and medicine promises benefits such as precision diagnosis, accurate patient monitoring, and reducing clinician workload. These benefits are particularly salient for asthma and chronic obstructive pulmonary disease (COPD), two of the most prevalent respiratory conditions. As AI techniques continue to burgeon, and the body of research into using these for respiratory medicine grows, it is important to review and assess the literature. This work offers a comprehensive systematic review of 121 articles which apply AI techniques to diagnose and monitor asthma and COPD based on the themes of disease detection; identification of risk factors; disease quantification and modelling; and exacerbation prediction. Prior surveys consider AI in medicine more broadly, lack a systematic methodology, and do not critical appraises or assess risk of bias. This systematic review offers a focused, methodical assessment specific to AI applications for diagnosing and monitoring asthma and COPD, and contains a quality appraisal of literature, complete with a risk of bias assessment. Finally, we offer a roadmap for the integration of AI in respiratory healthcare, with recommendations for future research and clinical translation, informed by gaps identified in the body of literature.