Artificial intelligence outperforms pulmonologists in the interpretation of pulmonary function tests.
Journal
The European respiratory journal
ISSN: 1399-3003
Titre abrégé: Eur Respir J
Pays: England
ID NLM: 8803460
Informations de publication
Date de publication:
04 2019
04 2019
Historique:
received:
30
08
2018
accepted:
25
01
2019
pubmed:
16
2
2019
medline:
29
10
2020
entrez:
16
2
2019
Statut:
epublish
Résumé
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<0.0001 for both measures).The interpretation of PFTs by pulmonologists leads to marked variations and errors. AI-based software provides more accurate interpretations and may serve as a powerful decision support tool to improve clinical practice.
Identifiants
pubmed: 30765505
pii: 13993003.01660-2018
doi: 10.1183/13993003.01660-2018
pii:
doi:
Types de publication
Comparative Study
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Investigateurs
R De Pauw
(R)
C Depuydt
(C)
C Haenebalcke
(C)
S Muyldermans
(S)
V Ringoet
(V)
D Stevens
(D)
S Bayat
(S)
J Benet
(J)
E Catho
(E)
J Claustre
(J)
A Fedi
(A)
M A Ferjani
(MA)
R Guzun
(R)
M Isnard
(M)
S Nicolas
(S)
T Pierret
(T)
C Pison
(C)
S Rouches
(S)
B Wuyam
(B)
J L Corhay
(JL)
J Guiot
(J)
K Ghysen
(K)
L Renaud
(L)
A Sibille
(A)
H De La Barriere
(H)
C Charpentier
(C)
S Corhut
(S)
K A Hamdan
(KA)
M Schlesser
(M)
G Wirtz
(G)
E Alabadan
(E)
G Birsen
(G)
P R Burgel
(PR)
A Chohra
(A)
C Hamard
(C)
B Lemarié
(B)
M N Lothe
(MN)
C Martin
(C)
A C Sainte-Marie
(AC)
L Sebane
(L)
Y Berk
(Y)
B de Brouwer
(B)
R Janssen
(R)
J Kerkhoff
(J)
A Spaanderman
(A)
M Stegers
(M)
A Termeer
(A)
I van Grimbergen
(I)
A van Veen
(A)
L van Ruitenbeek
(L)
L Vermeer
(L)
R Zaal
(R)
M Zijlker
(M)
J Aumann
(J)
K Cuppens
(K)
D Degraeve
(D)
K Demuynck
(K)
B Dieriks
(B)
K Pat
(K)
L Spaas
(L)
R Van Puijenbroek
(R)
K Weytjens
(K)
J Wynants
(J)
V Adam
(V)
B J Berendes
(BJ)
E Hardeman
(E)
P Jordens
(P)
E Munghen
(E)
K Tournoy
(K)
P Vercauter
(P)
T Alame
(T)
M Bruyneel
(M)
M Gabrovska
(M)
I Muylle
(I)
V Ninane
(V)
D Rozen
(D)
P Rummens
(P)
S Van Den Broecke
(S)
A Froidure
(A)
S Gohy
(S)
G Liistro
(G)
T Pieters
(T)
C Pilette
(C)
F Pirson
(F)
H Kerstjens
(H)
M Van den Berge
(M)
N Ten Hacken
(N)
M Duiverman
(M)
D Koster
(D)
B Vosse
(B)
L Conemans
(L)
M Maus
(M)
M Bischoff
(M)
M Rutten
(M)
D Agterhuis
(D)
R Sprooten
(R)
B Beutel
(B)
A Jerrentrup
(A)
A Klemmer
(A)
C Viniol
(C)
C Vogelmeier
(C)
H Bode
(H)
C Dooms
(C)
D Gullentops
(D)
W Janssens
(W)
K Nackaerts
(K)
D Rutens
(D)
E Wauters
(E)
W Wuyts
(W)
E Derom
(E)
S Dobbelaere
(S)
S Loof
(S)
G Serry
(G)
B Putman
(B)
L Van Acker
(L)
Y Vandeweygaerde
(Y)
M Criel
(M)
M Daenen
(M)
R Gubbelmans
(R)
S Klerkx
(S)
E Michiels
(E)
M Thomeer
(M)
A Vanhauwaert
(A)
Commentaires et corrections
Type : CommentIn
Type : CommentIn
Type : CommentIn
Type : CommentIn
Informations de copyright
Copyright ©ERS 2019.
Déclaration de conflit d'intérêts
Conflict of interest: M. Topalovic has nothing to disclose. Conflict of interest: N. Das has nothing to disclose. Conflict of interest: P-R. Burgel reports personal fees from AstraZeneca, Boehringer Ingelheim, Chiesi, Novartis, Teva and Vertex, outside the submitted work. Conflict of interest: M. Daenen has nothing to disclose. Conflict of interest: E. Derom has nothing to disclose. Conflict of interest: C. Haenebalcke reports personal fees from Novartis, Chiesi, GSK and AstraZeneca, outside the submitted work. Conflict of interest: R. Janssen has nothing to disclose. Conflict of interest: H.A.M. Kerstjens has nothing to disclose. Conflict of interest: G. Liistro has nothing to disclose. Conflict of interest: R. Louis reports grants and personal fees from GSK and Novartis, personal fees from AstraZeneca, and grants from Chiesi, outside the submitted work. Conflict of interest: V. Ninane has nothing to disclose. Conflict of interest: C. Pison has nothing to disclose. Conflict of interest: M. Schlesser has nothing to disclose. Conflict of interest: P. Vercauter has nothing to disclose. Conflict of interest: C.F. Vogelmeier reports personal fees from Almirall, Cipla, Berlin-Chemie/Menarini, CSL Behring and Teva, grants and personal fees from AstraZeneca, Boehringer Ingelheim, Chiesi, GSK, Grifols, Mundipharma, Novartis and Takeda, grants from German Federal Ministry of Education and Research (BMBF) Competence Network Asthma and COPD (ASCONET), Bayer Schering Pharma AG, MSD and Pfizer, outside the submitted work. Conflict of interest: E. Wouters reports personal fees for board membership from Nycomed and Boehringer, grants from AstraZeneca and GSK, and personal fees for lectures from AstraZeneca, GSK, Novartis and Chiesi, outside the submitted work. Conflict of interest: J. Wynants has nothing to disclose. Conflict of interest: W. Janssens has nothing to disclose.