TY - JOUR
T1 - Collaboration between explainable artificial intelligence and pulmonologists improves the accuracy of pulmonary function test interpretation
AU - Das, Nilakash
AU - Happaerts, Sofie
AU - Gyselinck, Iwein
AU - Staes, Michael
AU - Derom, Eric
AU - Brusselle, Guy
AU - Burgos, Felip
AU - Contoli, Marco
AU - Dinh-Xuan, Anh Tuan
AU - Franssen, Frits M. E.
AU - Gonem, Sherif
AU - Greening, Neil
AU - Haenebalcke, Christel
AU - Man, William D-C.
AU - Moises, Jorge
AU - Peche, Rudi
AU - Poberezhets, Vitalii
AU - Quint, Jennifer K.
AU - Steiner, Michael C.
AU - Vanderhelst, Eef
AU - Abdo, Mustafa
AU - Topalovic, Marko
AU - Janssens, Wim
PY - 2023/5/1
Y1 - 2023/5/1
N2 - Background Few studies have investigated the collaborative potential between artificial intelligence (AI) and pulmonologists for diagnosing pulmonary disease. We hypothesised that the collaboration between 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 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.
AB - Background Few studies have investigated the collaborative potential between artificial intelligence (AI) and pulmonologists for diagnosing pulmonary disease. We hypothesised that the collaboration between 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 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.
KW - BLACK-BOX
KW - DIAGNOSIS
U2 - 10.1183/13993003.01720-2022
DO - 10.1183/13993003.01720-2022
M3 - Article
C2 - 37080566
SN - 0903-1936
VL - 61
JO - European Respiratory Journal
JF - European Respiratory Journal
IS - 5
M1 - 2201720
ER -