Automated Detection and Quantification of Hemorrhagic Transformation After Endovascular Thrombectomy
Stroke: Vascular and Interventional Neurology · 19.11.2026
DOI: 10.1161/svin.126.002506 ↗Kurz gefasst
Fachartikel in der Zeitschrift Stroke: Vascular and Interventional Neurology, online erschienen am 19.11.2026. Verfasst von Wi-Sun Ryu und 39 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
BACKGROUND: Hemorrhagic transformation after endovascular thrombectomy is a principal determinant of clinical outcome. Although artificial intelligence (AI) algorithms for spontaneous hemorrhage detection have demonstrated high diagnostic accuracy, none have been rigorously validated for postprocedural hemorrhagic transformation across diverse imaging modalities. METHODS: We conducted a multicenter diagnostic accuracy study within a nationwide stroke registry (18 centers, June 2022–December 2023). Patients who underwent endovascular thrombectomy for anterior circulation large-vessel occlusion and received follow-up noncontrast computed tomography, gradient-recalled echo, or susceptibility-weighted imaging within 168 hours were included. AI-derived hemorrhage volumes (mL) from a regulatory-cleared algorithm (noncontrast computed tomography) and newly developed algorithms (gradient-recalled echo, susceptibility-weighted imaging) were compared against expert-determined ECASS (European Cooperative Acute Stroke Study) classification. Diagnostic performance was assessed by sensitivity, specificity, and receiver operating characteristic analysis. Three-month modified Rankin Scale scores were evaluated for volume–outcome association. RESULTS: Among 1490 patients (median age 73 years; 57.4% male), hemorrhagic transformation was present in 41.4% and parenchymal hemorrhage in 11.1%. Parenchymal hemorrhage detection sensitivity exceeded 94% across all modalities (noncontrast computed tomography, 95.4%; gradient-recalled echo, 94.4%; and susceptibility-weighted imaging, 98.3%), with area under the curves of 0.900, 0.943, and 0.953, respectively. Youden-optimized volume thresholds for parenchymal hemorrhage ranged from 1.6 to 1.8 mL. AI-derived volume demonstrated an exposure-response association with 3-month modified Rankin Scale (Spearman ρ=0.353; P
Zitieren
APA
Ryu, W., Ha, H., Sunwoo, L., Lee, M., Kang, K., Kim, J. G., Lee, S. J., Cha, J., Park, T. H., Lee, J., Lee, K., Kwon, D. H., Lee, J., Park, H., Cho, Y., Hong, K., Lee, M., Oh, M. S., Yu, K., …, & Bae, H. (2026). Automated Detection and Quantification of Hemorrhagic Transformation After Endovascular Thrombectomy. Stroke: Vascular and Interventional Neurology. https://doi.org/10.1161/svin.126.002506
Vancouver
Ryu W, Ha H, Sunwoo L, Lee M, Kang K, Kim JG, et al. Automated Detection and Quantification of Hemorrhagic Transformation After Endovascular Thrombectomy. Stroke: Vascular and Interventional Neurology. 2026. doi:10.1161/svin.126.002506
BibTeX
@article{Ryu2026,
author = {Ryu, Wi-Sun and Ha, Hee-Jung and Sunwoo, Leonard and Lee, Myungjae and Kang, Kyusik and Kim, Jae Guk and Lee, Soo Joo and Cha, Jae-Kwan and Park, Tai Hwan and Lee, Jeong-Yoon and Lee, Kyungbok and Kwon, Doo Hyuk and Lee, Jun and Park, Hong-Kyun and Cho, Yong-Jin and Hong, Keun-Sik and Lee, Minwoo and Oh, Mi Sun and Yu, Kyung-Ho and Gwak, Dong-Seok and Kim, Dong-Eog and Kim, Hyunsoo and Kim, Joon-Tae and Kim, Joong-Goo and Choi, Jay Chol and Kim, Wook-Joo and Kwon, Jee Hyun and Yum, Kyu Sun and Shin, Dong-Ick and Hong, Jeong-Ho and Sohn, Sung-Il and Lee, Sang-Hwa and Kim, Chulho and Jeong, Hae-Bong and Park, Kwang-Yeol and Lee, Keon-Joo and Kim, Chi Kyung and Kang, Jihoon and Kim, Jun Yup and Bae, Hee-Joon},
title = {Automated Detection and Quantification of Hemorrhagic Transformation After Endovascular Thrombectomy},
journal = {Stroke: Vascular and Interventional Neurology},
year = {2026},
doi = {10.1161/svin.126.002506}
}
EUVIMED