Abstract
Introduction: Diagnosis of occult atrial fibrillation (AF) is difficult as it is often asymptomatic, leading to under-detection. Current diagnostic tests have variable limitations in feasibility and accuracy. Machine learning is gaining greater traction for clinical decision-making and may help facilitate the detection of undiagnosed AF when applied to magnetic resonance imaging (MRI). We hypothesize that a machine learning algorithm increases the accurate classification of MRIs of stroke patients into those due to AF versus large artery atherosclerosis. Methods: Stroke aetiology for each patient was determined by a review of medical records and investigations. Patients with either AF or large artery atherosclerosis were included. Patients were randomly divided into the training and validation groups (4:1). A 3D convolutional neural network (ConvNeXt) was developed to train and validate the algorithm. After training, the models were evaluated using common metrics for binary classification. Results: A total of 235 patients were analysed (97 with AF, 138 without AF). The mean age of the sample was 71.1 (SD 14.2), and 35% were female. The best discriminative performance was obtained in the 5th fold of cross-validation (AUC-ROC 0.88), and the overall model performance was 0.81 ± 0.05. The best performing metrics were precision (0.84 ± 0.08) and the F1-score (0.77 ± 0.06). Conclusion: Our machine learning algorithm has reasonable classification power in categorizing stroke patients into those with and without underlying AF. Testing in external validation datasets is critical to confirm these results.
Journal Section:
Research Article
Plain Language Summary
Atrial fibrillation (AF) is an abnormal heart rhythm which is a common cause of stroke. AF is associated with certain findings on brain imaging which can provide clues to the diagnosis. This paper showed the use of artificial intelligence, when applied to MRI brain imaging, can assist in the differentiating AF from another common cause of stroke, large artery atherosclerosis.
Journal Section:
Research Article
References
1.
Schnabel
RB
, Haeusler
KG
, Healey
JS
, Freedman
B
, Boriani
G
, Brachmann
J
, et al. Searching for atrial fibrillation poststroke: a white paper of the AF-SCREEN international collaboration
. Circulation
. 2019
;140
(22
):1834
–50
. 2.
Reiffel
JA
, Verma
A
, Kowey
PR
, Halperin
JL
, Gersh
BJ
, Wachter
R
, et al. Incidence of previously undiagnosed atrial fibrillation using insertable cardiac monitors in a high-risk population: the REVEAL AF study
. JAMA Cardiol
. 2017
;2
(10
):1120
–7
. 3.
McGrath
ER
, Kapral
MK
, Fang
J
, Eikelboom
JW
, Conghaile
A
, Canavan
M
, et al. Association of atrial fibrillation with mortality and disability after ischemic stroke
. Neurology
. 2013
;81
(9
):825
–32
. 4.
Kalarus
Z
, Mairesse
GH
, Sokal
A
, Boriani
G
, Sredniawa
B
, Casado-Arroyo
R
, et al. Searching for atrial fibrillation: looking harder, looking longer, and in increasingly sophisticated ways. An EHRA position paper
. Europace
. 2023
;25
(1
):185
–98
. 5.
Yushan
B
, Tan
BYQ
, Ngiam
NJ
, Chan
BPL
, Luen
TH
, Sharma
VK
, et al. Association between bilateral infarcts pattern and detection of occult atrial fibrillation in embolic stroke of undetermined source (ESUS) patients with insertable cardiac monitor (ICM)
. J Stroke Cerebrovasc Dis
. 2019
;28
(9
):2448
–52
. 6.
Sharobeam
A
, Churilov
L
, Parsons
M
, Donnan
GA
, Davis
SM
, Yan
B
. Patterns of infarction on MRI in patients with acute ischemic stroke and cardio-embolism: a systematic review and meta-analysis
. Front Neurol
. 2020
;11
:606521
. 7.
Chung
JW
, Park
SH
, Kim
N
, Kim
WJ
, Park
JH
, Ko
Y
, et al. Trial of ORG 10172 in Acute Stroke Treatment (TOAST) classification and vascular territory of ischemic stroke lesions diagnosed by diffusion-weighted imaging
. J Am Heart Assoc
. 2014
;3
(4
):e001119
. 8.
Mayasi
Y
, Helenius
J
, McManus
DD
, Goddeau
RP
Jr, Jun-O’Connell
AH
, Moonis
M
, et al. Atrial fibrillation is associated with anterior predominant white matter lesions in patients presenting with embolic stroke
. J Neurol Neurosurg Psychiatry
. 2018
;89
(1
):6
–13
. 9.
Akhtar
T
, Shahjouei
S
, Zand
R
. Etiologies of simultaneous cerebral infarcts in multiple arterial territories: a simple literature-based pooled analysis
. Neurol India
. 2019
;67
(3
):692
–5
. 10.
Pierik
R
, Algra
A
, van Dijk
E
, Erasmus
ME
, van Gelder
IC
, Koudstaal
PJ
, et al. Distribution of cardioembolic stroke: a cohort study
. Cerebrovasc Dis
. 2020
;49
(1
):97
–104
. 11.
Fisher
CM
. The arterial lesions underlying lacunes
. Acta Neuropathol
. 1968
;12
(1
):1
–15
. 12.
Müller
K
, Courtois
G
, Ursini
MV
, Schwaninger
M
. New insight into the pathogenesis of cerebral small-vessel diseases
. Stroke
. 2017
;48
(2
):520
–7
. 13.
García-Carmona
JA
, Conesa-García
E
, Vidal-Mena
D
, González-Morales
M
, Ramos-Arenas
V
, Sánchez-Vizcaíno-Buendía
C
, et al. Increased plasma levels of N-terminal pro-B-type natriuretic peptide as biomarker for the diagnosis of cardioembolic ischaemic stroke
. Neurología
. 2024
;39
(6
):496
–504
. 14.
15.
Esteva
A
, Robicquet
A
, Ramsundar
B
, Kuleshov
V
, DePristo
M
, Chou
K
, et al. A guide to deep learning in healthcare
. Nat Med
. 2019
;25
(1
):24
–9
. 16.
Ho
KC
, Speier
W
, Zhang
H
, Scalzo
F
, El-Saden
S
, Arnold
CW
. A machine learning approach for classifying ischemic stroke onset time from imaging
. IEEE Trans Med Imaging
. 2019
;38
(7
):1666
–76
. 17.
Lee
H
, Lee
EJ
, Ham
S
, Lee
HB
, Lee
JS
, Kwon
SU
, et al. Machine learning approach to identify stroke within 4.5 hours
. Stroke
. 2020
;51
(3
):860
–6
. 18.
Kim
B-K
, Park
S
, Han
M-K
, Hong
J-H
, Lee
D-I
, Yum
KS
. Deep learning for prediction of mechanism in acute ischemic stroke using brain diffusion magnetic resonance image
. J Neurocrit Care
. 2023
;16
(2
):85
–93
. 19.
Meschia
JF
, Barrett
KM
, Chukwudelunzu
F
, Brown
WM
, Case
LD
, Kissela
BM
, et al. Interobserver agreement in the trial of org 10172 in acute stroke treatment classification of stroke based on retrospective medical record review
. J Stroke Cerebrovasc Dis
. 2006
;15
(6
):266
–72
. 20.
Marnane
M
, Duggan
CA
, Sheehan
OC
, Merwick
A
, Hannon
N
, Curtin
D
, et al. Stroke subtype classification to mechanism-specific and undetermined categories by TOAST, A-S-C-O, and causative classification system: direct comparison in the North Dublin population stroke study
. Stroke
. 2010
;41
(8
):1579
–86
. 21.
Ay
H
, Benner
T
, Arsava
EM
, Furie
KL
, Singhal
AB
, Jensen
MB
, et al. A computerized algorithm for etiologic classification of ischemic stroke: the Causative Classification of Stroke System
. Stroke
. 2007
;38
(11
):2979
–84
. 22.
Liu Hm
Z
, Wu
C-Y
, Feichtenhofer
C
, Darrell
T
, Xie
S
. A ConvNet for the 2020s
. In: IEEE/CVF conference on computer vision and pattern recognition (CVPR)
; New Orleans, LA, USA
; 2022
; p. 11966
–76
.23.
Selvaraju
RR
, Cogswell
M
, Das
A
, Vedantam
R
, Parikh
D
, Batra
D
. Grad-CAM: visual explanations from deep networks via gradient-based localization
. Int J Comput Vis
. 2020
;128
(2
):336
–59
. 24.
Wei
J
, Wang
Q
, Li
Z
, Wang
S
, Zhou
SK
, Cui
S
. Shallow feature matters for weakly supervised object localization
. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition; 2021
.25.
Cullen
MW
, Kim
S
, Piccini
JP
, Ansell
JE
, Fonarow
GC
, Hylek
EM
, et al. Risks and benefits of anticoagulation in atrial fibrillation: insights from the outcomes registry for better informed treatment of atrial fibrillation (ORBIT-AF) registry
. Circ Cardiovasc Qual Outcomes
. 2013
;6
(4
):461
–9
. 26.
Halliday
A
, Harrison
M
, Hayter
E
, Kong
X
, Mansfield
A
, Marro
J
, et al. 10-year stroke prevention after successful carotid endarterectomy for asymptomatic stenosis (ACST-1): a multicentre randomised trial
. The Lancet
. 2010
;376
(9746
):1074
–84
. 27.
Kamel
H
, Navi
BB
, Parikh
NS
, Merkler
AE
, Okin
PM
, Devereux
RB
, et al. Machine learning prediction of stroke mechanism in embolic strokes of undetermined source
. Stroke
. 2020
;51
(9
):e203
–10
. 28.
Yuan
N
, Stein
NR
, Duffy
G
, Sandhu
RK
, Chugh
SS
, Chen
PS
, et al. Deep learning evaluation of echocardiograms to identify occult atrial fibrillation
. NPJ Digit Med
. 2024
;7
(1
):96
. 29.
Ming
C
, Lee
GJW
, Teo
YH
, Teo
YN
, Toh
EMS
, Li
TYW
, et al. Machine learning modeling to predict atrial fibrillation detection in embolic stroke of undetermined source patients
. J Pers Med
. 2024
;14
(5
):534
. 30.
Arsava
EM
, Helenius
J
, Avery
R
, Sorgun
MH
, Kim
G-M
, Pontes-Neto
OM
, et al. Assessment of the predictive validity of etiologic stroke classification
. JAMA Neurol
. 2017
;74
(4
):419
–26
. 31.
Bernstein
RA
, Kamel
H
, Granger
CB
, Piccini
JP
, Sethi
PP
, Katz
JM
, et al. Effect of long-term continuous cardiac monitoring vs usual care on detection of atrial fibrillation in patients with stroke attributed to large- or small-vessel disease: the STROKE-AF randomized clinical trial
. JAMA
. 2021
;325
(21
):2169
–77
. 32.
Rubiera
M
, Aires
A
, Antonenko
K
, Lemeret
S
, Nolte
CH
, Putaala
J
, et al. European Stroke Organisation (ESO) guideline on screening for subclinical atrial fibrillation after stroke or transient ischaemic attack of undetermined origin
. Eur Stroke J
. 2022
;7
(3
):VI
. 33.
Kleindorfer
DO
, Towfighi
A
, Chaturvedi
S
, Cockroft
KM
, Gutierrez
J
, Lombardi-Hill
D
, et al. 2021 guideline for the prevention of stroke in patients with stroke and transient ischemic attack: a guideline from the American heart association/American stroke association
. Stroke
. 2021
;52
(7
):e364
–467
. © 2025 S. Karger AG, Basel
Copyright / Drug Dosage / Disclaimer
Copyright: All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing from the publisher.
Drug Dosage: The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any changes in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug.
Disclaimer: The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publishers and the editor(s). The appearance of advertisements or/and product references in the publication is not a warranty, endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements.
2025
You do not currently have access to this content.
