Collaboration between explainable artificial intelligence and pulmonologists improves the accuracy of pulmonary function test interpretation
European Respiratory Journal · 2023-05 · 61 (5) · 2201720
DOI: 10.1183/13993003.01720-2022 ↗Kurz gefasst
Fachartikel in der Zeitschrift European Respiratory Journal, online erschienen am 20.04.2023. Verfasst von Nilakash Das und 22 weiteren Autor:innen. Beteiligte Einrichtung: Laboratory of Respiratory Diseases and Thoracic Surgery, Department of Chronic Diseases Metabolism and Ageing, KU Leuven, Leuven, Belgium. Hauptthemen laut MEDLINE: Artificial Intelligence, Lung Diseases. Die Arbeit wurde bisher 49-mal zitiert (Crossref) und nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Agentschap Innoveren en Ondernemen, Fonds Wetenschappelijk Onderzoek.
Abstract (Originalsprache)
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
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Laboratory of Respiratory Diseases and Thoracic Surgery, Department of Chronic Diseases Metabolism and Ageing, KU Leuven, Leuven, Belgium.
- Clinical Department of Respiratory Diseases, University Hospitals Leuven, Leuven, Belgium.
- UZ Gent, University of Ghent, Ghent, Belgium.
- Department of Pulmonary Medicine, Hospital Clinic, Institut d'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), University of Barcelona, Barcelona, Spain.
- Department of Translational Medicine, University of Ferrara, Ferrara, Italy.
- Service de Physiologie-Explorations Fonctionnelles, AP-HP, Hôpital Cochin, Université Paris Cité, Paris, France.
ORCID-Profile: Sofie Happaerts, Iwein Gyselinck, Eric Derom, Guy Brusselle, Felip Burgos, Anh Tuan Dinh-Xuan, Frits M E Franssen, William D-C Man, Jorge Moisés, Vitalii Poberezhets, Jennifer K Quint, Eef Vanderhelst, Wim Janssens
Förderung
- Agentschap Innoveren en Ondernemen
- Fonds Wetenschappelijk Onderzoek
Zitieren
APA
Das, N., Happaerts, S., Gyselinck, I., Staes, M., Derom, E., Brusselle, G., Burgos, F., Contoli, M., Dinh-Xuan, A. T., Franssen, F. M. E., Gonem, S., Greening, N., Haenebalcke, C., Man, W. D., Moisés, J., Peché, R., Poberezhets, V., Quint, J. K., Steiner, M. C., …, & Janssens, W. (2023). Collaboration between explainable artificial intelligence and pulmonologists improves the accuracy of pulmonary function test interpretation. European Respiratory Journal, 61(5), 2201720. https://doi.org/10.1183/13993003.01720-2022
Vancouver
Das N, Happaerts S, Gyselinck I, Staes M, Derom E, Brusselle G, et al. Collaboration between explainable artificial intelligence and pulmonologists improves the accuracy of pulmonary function test interpretation. European Respiratory Journal. 2023;61(5):2201720. doi:10.1183/13993003.01720-2022
BibTeX
@article{Das2023,
author = {Das, Nilakash and Happaerts, Sofie and Gyselinck, Iwein and Staes, Michael and Derom, Eric and Brusselle, Guy and Burgos, Felip and Contoli, Marco and Dinh-Xuan, Anh Tuan and Franssen, Frits M E and Gonem, Sherif and Greening, Neil and Haenebalcke, Christel and Man, William D-C and Moisés, Jorge and Peché, Rudi and Poberezhets, Vitalii and Quint, Jennifer K and Steiner, Michael C and Vanderhelst, Eef and Abdo, Mustafa and Topalovic, Marko and Janssens, Wim},
title = {Collaboration between explainable artificial intelligence and pulmonologists improves the accuracy of pulmonary function test interpretation},
journal = {European Respiratory Journal},
year = {2023},
volume = {61},
number = {5},
pages = {2201720},
doi = {10.1183/13993003.01720-2022}
}
EUVIMED