Diagnostic test accuracy study of a commercially available deep learning algorithm for ischemic lesion detection on brain MRIs in suspected stroke patients from a non-comprehensive stroke center
Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt
Standard
Diagnostic test accuracy study of a commercially available deep learning algorithm for ischemic lesion detection on brain MRIs in suspected stroke patients from a non-comprehensive stroke center. / Krag, Christian H.; Müller, Felix C.; Gandrup, Karen L.; Raaschou, Henriette; Andersen, Michael B.; Brejnebøl, Mathias W.; Sagar, Malini V.; Bojsen, Jonas A.; Rasmussen, Benjamin S.; Graumann, Ole; Nielsen, Mads; Kruuse, Christina; Boesen, Mikael.
I: European Journal of Radiology, Bind 168, 111126, 2023.Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - Diagnostic test accuracy study of a commercially available deep learning algorithm for ischemic lesion detection on brain MRIs in suspected stroke patients from a non-comprehensive stroke center
AU - Krag, Christian H.
AU - Müller, Felix C.
AU - Gandrup, Karen L.
AU - Raaschou, Henriette
AU - Andersen, Michael B.
AU - Brejnebøl, Mathias W.
AU - Sagar, Malini V.
AU - Bojsen, Jonas A.
AU - Rasmussen, Benjamin S.
AU - Graumann, Ole
AU - Nielsen, Mads
AU - Kruuse, Christina
AU - Boesen, Mikael
N1 - Publisher Copyright: © 2023
PY - 2023
Y1 - 2023
N2 - Purpose: To estimate the ability of a commercially available artificial intelligence (AI) tool to detect acute brain ischemia on Magnetic Resonance Imaging (MRI), compared to an experienced neuroradiologist. Methods: We retrospectively included 1030 patients with brain MRI, suspected of stroke from January 6th, 2020 to 1st of April 2022, based on these criteria: Age ≥ 18 years, symptoms within four weeks before the scan. The neuroradiologist reinterpreted the MRI scans and subclassified ischemic lesions for reference. We excluded scans with interpretation difficulties due to artifacts or missing sequences. Four MRI scanner models from the same vendor were used. The first 800 patients were included consecutively, remaining enriched for less frequent lesions. The index test was a CE-approved AI tool (Apollo version 2.1.1 by Cerebriu). Results: The final analysis cohort comprised 995 patients (mean age 69 years, 53 % female). A case-based analysis for detecting acute ischemic lesions showed a sensitivity of 89 % (95 % CI: 85 %–91 %) and specificity of 90 % (95 % CI: 87 %–92 %). We found no significant difference in sensitivity or specificity based on sex, age, or comorbidities. Specificity was reduced in cases with DWI artifacts. Multivariate analysis showed that increasing ischemic lesion size and fragmented lesions were independently associated with higher sensitivity, while non-acute lesion ages lowered sensitivity. Conclusions: The AI tool exhibits high sensitivity and specificity in detecting acute ischemic lesions on MRI compared to an experienced neuroradiologist. While sensitivity depends on the ischemic lesions' characteristics, specificity depends on the image quality.
AB - Purpose: To estimate the ability of a commercially available artificial intelligence (AI) tool to detect acute brain ischemia on Magnetic Resonance Imaging (MRI), compared to an experienced neuroradiologist. Methods: We retrospectively included 1030 patients with brain MRI, suspected of stroke from January 6th, 2020 to 1st of April 2022, based on these criteria: Age ≥ 18 years, symptoms within four weeks before the scan. The neuroradiologist reinterpreted the MRI scans and subclassified ischemic lesions for reference. We excluded scans with interpretation difficulties due to artifacts or missing sequences. Four MRI scanner models from the same vendor were used. The first 800 patients were included consecutively, remaining enriched for less frequent lesions. The index test was a CE-approved AI tool (Apollo version 2.1.1 by Cerebriu). Results: The final analysis cohort comprised 995 patients (mean age 69 years, 53 % female). A case-based analysis for detecting acute ischemic lesions showed a sensitivity of 89 % (95 % CI: 85 %–91 %) and specificity of 90 % (95 % CI: 87 %–92 %). We found no significant difference in sensitivity or specificity based on sex, age, or comorbidities. Specificity was reduced in cases with DWI artifacts. Multivariate analysis showed that increasing ischemic lesion size and fragmented lesions were independently associated with higher sensitivity, while non-acute lesion ages lowered sensitivity. Conclusions: The AI tool exhibits high sensitivity and specificity in detecting acute ischemic lesions on MRI compared to an experienced neuroradiologist. While sensitivity depends on the ischemic lesions' characteristics, specificity depends on the image quality.
KW - Artificial Intelligence
KW - Diagnostic Accuracy
KW - Diffusion Weighted Imaging
KW - Ischemic Stroke
KW - Magnetic Resonance Imaging
KW - Neuroradiology
UR - http://www.scopus.com/inward/record.url?scp=85173215282&partnerID=8YFLogxK
U2 - 10.1016/j.ejrad.2023.111126
DO - 10.1016/j.ejrad.2023.111126
M3 - Journal article
C2 - 37804650
AN - SCOPUS:85173215282
VL - 168
JO - European Journal of Radiology
JF - European Journal of Radiology
SN - 0720-048X
M1 - 111126
ER -
ID: 369927646