TY - JOUR
T1 - Diagnostic performance of an algorithm for automated large vessel occlusion detection on CT angiography
AU - MR CLEAN Registry and PRESTO investigators
AU - Luijten, Sven P.R.
AU - Wolff, Lennard
AU - Duvekot, Martijne H.C.
AU - Van Doormaal, Pieter Jan
AU - Moudrous, Walid
AU - Kerkhoff, Henk
AU - Lycklama a Nijeholt, Geert J.
AU - Bokkers, Reinoud P.H.
AU - Yo, Lonneke S.F.
AU - Hofmeijer, Jeannette
AU - van Zwam, Wim H.
AU - van Es, Adriaan C.G.M.
AU - Dippel, Diederik W.J.
AU - Roozenbeek, Bob
AU - van der Lugt, Aad
AU - van Oostenbrugge, Robert J.
AU - Boiten, Jelis
AU - Majoie, Charles B.L.M.
AU - Roos, Yvo B.W.E.M.
AU - Vos, Jan Albert
AU - Jansen, Ivo G.H.
AU - Mulder, Maxim J.H.L.
AU - Goldhoorn, Robert Jan B.
AU - Compagne, Kars C.J.
AU - Kappelhof, Manon
AU - Schonewille, Wouter J.
AU - Coutinho, Jonathan M.
AU - Wermer, Marieke J.H.
AU - van Walderveen, Marianne A.A.
AU - Staals, Julie
AU - Martens, Jasper M.
AU - Emmer, Bart J.
AU - de Bruijn, Sebastiaan F.
AU - van Dijk, Lukas C.
AU - van der Worp, Bart
AU - Lo, Rob H.
AU - van Dijk, Ewoud J.
AU - Boogaarts, Hieronymus D.
AU - de Kort, Paul L.M.
AU - van Tuijl, Julia
AU - Peluso, Jo J.P.
AU - van den Berg, Jan S.P.
AU - van Hasselt, Boudewijn A.A.M.
AU - Aerden, Leo A.M.
AU - Sprengers, Marieke E.S.
AU - van den Berg, René
AU - Beenen, Ludo F.M.
AU - Roosendaal, Stefan D.
AU - Bot, Joost
AU - Berkhemer, Olvert A.
N1 - Funding Information: Funding The MR CLEAN Registry was partly funded by TWIN Foundation, Erasmus MC University Medical Center, Maastricht University Medical Center, and Amsterdam University Medical Center. PRESTO was funded by BeterKeten Collaboration and Theia Foundation (Zilveren Kruis). Funding Information: Competing interests WHvZ reports grants from Stryker and Cerenovus, all paid to the institution. DWJD reports funding from the Dutch Heart Foundation, Brain Foundation Netherlands, The Netherlands Organisation for Health Research and Development, Health Holland Top Sector Life Sciences and Health, and unrestricted grants from Penumbra, Stryker, Medtronic, Thrombolytic Science, LLC, and Cerenovus, all paid to the institution. AvdL reports grants from Penumbra, Stryker, Cerenovus, and Medtronic, all paid to the institution. Publisher Copyright: © Author(s) (or their employer(s)) 2022.
PY - 2022/8
Y1 - 2022/8
N2 - Background Machine learning algorithms hold the potential to contribute to fast and accurate detection of large vessel occlusion (LVO) in patients with suspected acute ischemic stroke. We assessed the diagnostic performance of an automated LVO detection algorithm on CT angiography (CTA). Methods Data from the MR CLEAN Registry and PRESTO were used including patients with and without LVO. CTA data were analyzed by the algorithm for detection and localization of LVO (intracranial internal carotid artery (ICA)/ICA terminus (ICA-T), M1, or M2). Assessments done by expert neuroradiologists were used as reference. Diagnostic performance was assessed for detection of LVO and per occlusion location by means of sensitivity, specificity, and area under the curve (AUC). Results We analyzed CTAs of 1110 patients from the MR CLEAN Registry (median age (IQR) 71 years (60–80); 584 men; 1110 with LVO) and of 646 patients from PRESTO (median age (IQR) 73 years (62–82); 358 men; 141 with and 505 without LVO). For detection of LVO, the algorithm yielded a sensitivity of 89% in the MR CLEAN Registry and a sensitivity of 72%, specificity of 78%, and AUC of 0.75 in PRESTO. Sensitivity per occlusion location was 88% for ICA/ICA-T, 94% for M1, and 72% for M2 occlusion in the MR CLEAN Registry, and 80% for ICA/ICA-T, 95% for M1, and 49% for M2 occlusion in PRESTO. Conclusion The algorithm provided a high detection rate for proximal LVO, but performance varied significantly by occlusion location. Detection of M2 occlusion needs further improvement.
AB - Background Machine learning algorithms hold the potential to contribute to fast and accurate detection of large vessel occlusion (LVO) in patients with suspected acute ischemic stroke. We assessed the diagnostic performance of an automated LVO detection algorithm on CT angiography (CTA). Methods Data from the MR CLEAN Registry and PRESTO were used including patients with and without LVO. CTA data were analyzed by the algorithm for detection and localization of LVO (intracranial internal carotid artery (ICA)/ICA terminus (ICA-T), M1, or M2). Assessments done by expert neuroradiologists were used as reference. Diagnostic performance was assessed for detection of LVO and per occlusion location by means of sensitivity, specificity, and area under the curve (AUC). Results We analyzed CTAs of 1110 patients from the MR CLEAN Registry (median age (IQR) 71 years (60–80); 584 men; 1110 with LVO) and of 646 patients from PRESTO (median age (IQR) 73 years (62–82); 358 men; 141 with and 505 without LVO). For detection of LVO, the algorithm yielded a sensitivity of 89% in the MR CLEAN Registry and a sensitivity of 72%, specificity of 78%, and AUC of 0.75 in PRESTO. Sensitivity per occlusion location was 88% for ICA/ICA-T, 94% for M1, and 72% for M2 occlusion in the MR CLEAN Registry, and 80% for ICA/ICA-T, 95% for M1, and 49% for M2 occlusion in PRESTO. Conclusion The algorithm provided a high detection rate for proximal LVO, but performance varied significantly by occlusion location. Detection of M2 occlusion needs further improvement.
UR - http://www.scopus.com/inward/record.url?scp=85134435308&partnerID=8YFLogxK
U2 - https://doi.org/10.1136/neurintsurg-2021-017842
DO - https://doi.org/10.1136/neurintsurg-2021-017842
M3 - Article
C2 - 34413245
SN - 1759-8478
VL - 14
SP - 794
EP - 798
JO - Journal of NeuroInterventional Surgery
JF - Journal of NeuroInterventional Surgery
IS - 8
ER -