Identifying limitations to exercise with incremental cardiopulmonary exercise testing: a scoping review.

Cardiopulmonary exercise testing (CPET) is a comprehensive and invaluable assessment used to identify the mechanisms that limit exercise capacity. However, its interpretation remains poorly standardised. This scoping review aims to investigate which limitations to exercise are differentiated by the use of incremental CPET in literature and which criteria are used to identify them. We performed a systematic, electronic literature search of PubMed, Embase, Cochrane CENTRAL, Web of Science and Scopus. All types of publications that reported identification criteria for at least one limitation to exercise based on clinical parameters and CPET variables were eligible for inclusion. 86 publications were included, of which 57 were primary literature and 29 were secondary literature. In general, at the level of the cardiovascular system, a distinction was often made between a normal physiological limitation and a pathological one. Within the respiratory system, ventilatory limitation, commonly identified by a low breathing reserve, and gas exchange limitation, mostly identified by a high minute ventilation/carbon dioxide production slope and/or oxygen desaturation, were often described. Multiple terms were used to describe a limitation in the peripheral muscle, but all variables used to identify this limitation lacked specificity. Deconditioning was a frequently mentioned exercise limiting factor, but there was no consensus on how to identify it through CPET. There is large heterogeneity in the terminology, the classification and the identification criteria of limitations to exercise that are distinguished using incremental CPET. Standardising the interpretation of CPET is essential to establish an objective and consistent framework.

Collaboration between explainable artificial intelligence and pulmonologists improves the accuracy of pulmonary function test interpretation.

BACKGROUND: Few studies have investigated the collaborative potential between artificial intelligence (AI) and pulmonologists for diagnosing pulmonary disease. We hypothesised that the collaboration between a pulmonologist and AI with explanations (explainable AI (XAI)) is superior in diagnostic interpretation of pulmonary function tests (PFTs) than the pulmonologist without support. METHODS: The study was conducted in two phases, a monocentre study (phase 1) and a multicentre intervention study (phase 2). Each phase utilised two different sets of 24 PFT reports of patients with a clinically validated gold standard diagnosis. Each PFT was interpreted without (control) and with XAI's suggestions (intervention). Pulmonologists provided a differential diagnosis consisting of a preferential diagnosis and optionally up to three additional diagnoses. The primary end-point compared accuracy of preferential and additional diagnoses between control and intervention. Secondary end-points were the number of diagnoses in differential diagnosis, diagnostic confidence and inter-rater agreement. We also analysed how XAI influenced pulmonologists' decisions. RESULTS: In phase 1 (n=16 pulmonologists), mean preferential and differential diagnostic accuracy significantly increased by 10.4% and 9.4%, respectively, between control and intervention (p<0.001). Improvements were somewhat lower but highly significant (p<0.0001) in phase 2 (5.4% and 8.7%, respectively; n=62 pulmonologists). In both phases, the number of diagnoses in the differential diagnosis did not reduce, but diagnostic confidence and inter-rater agreement significantly increased during intervention. Pulmonologists updated their decisions with XAI's feedback and consistently improved their baseline performance if AI provided correct predictions. CONCLUSION: A collaboration between a pulmonologist and XAI is better at interpreting PFTs than individual pulmonologists reading without XAI support or XAI alone.