1. Artificial intelligence outperforms pulmonologists in the interpretation of pulmonary function tests
- Author
-
Topalovic, Marko, Das, Nilakash, Burgel, Pierre-Regis, Daenen, Marc, Derom, Eric, Haenebalcke, Christel, Janssen, Rob, Kerstjens, Huib AM, Liistro, Giuseppe, Louis, Renaud, Ninane, Vincent, Pison, Christophe, Schlesser, Marc, Vercauter, Piet, Vogelmeier, Claus F, Wouters, Emiel, Wynants, Jokke, Janssens, Wim, De Pauw, R, Depuydt, C, Haenebalcke, C, Muyldermans, S, Ringoet, V, Stevens, D, Bayat, S, Benet, J, Catho, E, Claustre, J, Fedi, A, Ferjani, MA, Guzun, R, Isnard, M, Nicolas, S, Pierret, T, Pison, C, Rouches, S, Wuyam, B, Corhay, JL, Guiot, J, Ghysen, K, Renaud, L, Sibille, A, De La Barriere, H, Charpentier, C, Corhut, S, Hamdan, KA, Schlesser, M, Wirtz, G, Alabadan, E, Birsen, G, Burgel, PR, Chohra, A, Hamard, C, Lemarie, B, Lothe, MN, Martin, C, Sainte-Marie, AC, Sebane, L, Berk, Y, de Brouwer, B, Janssen, R, Kerkhoff, J, Spaanderman, A, Stegers, M, Termeer, A, van Grimbergen, I, van Veen, A, van Ruitenbeek, L, Vermeer, L, Zaal, R, Zijlker, M, Aumann, J, Cuppens, K, Degraeve, D, Demuynck, K, Dieriks, B, Pat, K, Spaas, L, Van Puijenbroek, R, Weytjens, K, Wynants, J, Adam, V, Berendes, BJ, Hardeman, E, Jordens, P, Munghen, E, Tournoy, K, Vercauter, P, Alame, T, Bruyneel, M, Gabrovska, M, Muylle, I, Ninane, V, Rozen, D, Rummens, P, Van den Broecke, S, Froidure, A, Gohy, S, Liistro, G, Pieters, T, Pilette, C, Pirson, F, Kerstjens, H, Van den Berge, M, Ten Hacken, N, Duiverman, M, Koster, D, Vosse, B, Conemans, L, Maus, M, Bischoff, M, Rutten, M, Agterhuis, D, Sprooten, R, Beutel, B, Jerrentrup, A, Klemmer, A, Viniol, C, Vogelmeier, C, Bode, H, Dooms, C, Gullentops, D, Janssens, W, Nackaerts, K, Rutens, D, Wauters, E, Wuyts, W, Derom, E, Dobbelaere, S, Loof, S, Serry, G, Putman, B, Van Acker, L, Vandeweygaerde, Y, Criel, M, Daenen, M, Gubbelmans, R, Klerkx, S, Michiels, E, Thomeer, M, Vanhauwaert, A, UCL - (SLuc) Service de pneumologie, Groningen Research Institute for Asthma and COPD (GRIAC), Lifestyle Medicine (LM), Hôpital Cochin [AP-HP], Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP), Centre Hospitalier Universitaire [Grenoble] (CHU), Laboratory of Fundamental and Applied Bioenergetics = Laboratoire de bioénergétique fondamentale et appliquée (LBFA), Institut National de la Santé et de la Recherche Médicale (INSERM)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019]), RS: NUTRIM - R3 - Respiratory & Age-related Health, MUMC+: MA Longziekten (3), Pulmonologie, MUMC+: MA Med Staf Spec Longziekten (9), and MUMC+: MA Med Staf Artsass Longziekten (9)
- Subjects
Pulmonary and Respiratory Medicine ,Adult ,Male ,Pulmonary function ,STRATEGIES ,Pulmonary Function Study Investigators ,Context (language use) ,DIAGNOSIS ,GUIDELINES ,[SDV.MHEP.PSR]Life Sciences [q-bio]/Human health and pathology/Pulmonology and respiratory tract ,Pulmonary function testing ,03 medical and health sciences ,0302 clinical medicine ,Clinical history ,Artificial Intelligence ,Pulmonary Medicine ,Medicine ,Humans ,030212 general & internal medicine ,Prospective Studies ,Medical diagnosis ,Pulmonologists ,Aged ,Aged, 80 and over ,Interpretation (logic) ,business.industry ,Gold standard (test) ,STANDARDIZATION ,PERFORMANCE ,Middle Aged ,3. Good health ,Respiratory Function Tests ,Clinical Practice ,030228 respiratory system ,Female ,Artificial intelligence ,business ,[SDV.MHEP]Life Sciences [q-bio]/Human health and pathology ,Software - Abstract
The interpretation of pulmonary function tests (PFTs) to diagnose respiratory diseases is built on expert opinion that relies on the recognition of patterns and the clinical context for detection of specific diseases. In this study, we aimed to explore the accuracy and interrater variability of pulmonologists when interpreting PFTs compared with artificial intelligence (AI)-based software that was developed and validated in more than 1500 historical patient cases.120 pulmonologists from 16 European hospitals evaluated 50 cases with PFT and clinical information, resulting in 6000 independent interpretations. The AI software examined the same data. American Thoracic Society/European Respiratory Society guidelines were used as the gold standard for PFT pattern interpretation. The gold standard for diagnosis was derived from clinical history, PFT and all additional tests.The pattern recognition of PFTs by pulmonologists (senior 73%, junior 27%) matched the guidelines in 74.4±5.9% of the cases (range 56-88%). The interrater variability of κ=0.67 pointed to a common agreement. Pulmonologists made correct diagnoses in 44.6±8.7% of the cases (range 24-62%) with a large interrater variability (κ=0.35). The AI-based software perfectly matched the PFT pattern interpretations (100%) and assigned a correct diagnosis in 82% of all cases (p
- Published
- 2018
- Full Text
- View/download PDF