Share：
Share this content in WeChat
X
[Chinese] [PDF] 200 7
• Clinical Article •
Value of integrated MRI radiomics and clinical factors for post-thrombolytic hemorrhagic transformation in acute ischemic stroke
YIMURAN·Subi   PAHATI·Tuxunjiang   AINIKAERJIANG·Aihemaiti   HANJIAERBIEKE·Kukun   XU Rui  HAN Bingyan  XIE Chao  WANG Yunling 

DOI:10.12015/issn.1674-8034.2025.08.006.


[Abstract] Objective To investigate the predictive value of MRI-based radiomics models and clinical factor models for hemorrhagic transformation (HT) risk after thrombolysis in acute ischemic stroke (AIS).Materials and Methods Clinical and imaging data were retrospectively collected from 730 AIS patients at first presentation across two Centers. Data from Center 1 were randomly divided into a training set (436 cases) and an internal validation set (188 cases) in a 7:3 ratio. Univariate and multivariate logistic regression analyses identified independent clinical predictors of HT. Three models were constructed: (1) a clinical factor model, (2) a MRI radiomics model, and (3) a combined model integrating both features. External validation was performed using data from 106 patients from Center 2. Receiver operating characteristic (ROC) curves and area under the curve (AUC) values evaluated the predictive performance of the models, while DeLong's test was applied to compare statistical differences between AUCs.Results In the training set, the AUCs for the clinical factor model, radiomics model, and combined model were 0.810 (95% CI: 0.756 to 0.864), 0.896 (95% CI: 0.865 to 0.928), and 0.928 (95% CI: 0.899 to 0.958), respectively. In the internal validation set, the corresponding AUCs were 0.757 (95% CI: 0.671 to 0.843), 0.852 (95% CI: 0.791 to 0.913), and 0.872 (95% CI: 0.809 to 0.935). In the external validation set, the AUCs were 0.720 (95% CI: 0.602 to 0.839), 0.804 (95% CI: 0.711 to 0.897), and 0.828 (95% CI: 0.751 to 0.905), respectively. Decision curve analysis indicated that the combined model provided the highest net benefit.Conclusions Both MRI-based radiomics models and clinical factor models demonstrated predictive value for HT risk after thrombolysis in AIS. The integration of these two approaches achieved the best performance, offering potential clinical utility for individualized risk stratification.
[Keywords] ischemic stroke；hemorrhagic transformation；magnetic resonance imaging；machine learning；radiomics

YIMURAN·Subi 1   PAHATI·Tuxunjiang 1   AINIKAERJIANG·Aihemaiti 1   HANJIAERBIEKE·Kukun 1   XU Rui1   HAN Bingyan2   XIE Chao2   WANG Yunling1*  

1 Department of Radiology, Affiliated First Hospital of Xinjiang Medical University, Urumqi 830054, China

2 Department of Medical Imaging, Affiliated Seventh Hospital of Xinjiang Medical University, Urumqi 830028, China

Corresponding author: WANG Y L, E-mail: 1079806994@qq.com

Conflicts of interest   None.

Received  2025-02-17
Accepted  2025-07-07
DOI: 10.12015/issn.1674-8034.2025.08.006
DOI:10.12015/issn.1674-8034.2025.08.006.

[1]
GBD 2019 Stroke Collaborators. Global, regional, and national burden of stroke and its risk factors, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019[J]. Lancet Neurol, 2021, 20(10): 795-820. DOI: 10.1016/S1474-4422(21)00252-0.
[2]
ANDE S R, GRYNSPAN J, AVIV R I, et al. Imaging for Predicting Hemorrhagic Transformation of Acute Ischemic Stroke-A Narrative Review[J]. Can Assoc Radiol J, 2022, 73(1): 194-202. DOI: 10.1177/08465371211018369.
[3]
VON KUMMER R, BRODERICK J P, CAMPBELL B C, et al. The Heidelberg Bleeding Classification: Classification of Bleeding Events After Ischemic Stroke and Reperfusion Therapy[J]. Stroke, 2015, 46(10): 2981-2986. DOI: 10.1161/STROKEAHA.115.010049.
[4]
HACKE W, KASTE M, FIESCHI C, et al. Randomised double-blind placebo-controlled trial of thrombolytic therapy with intravenous alteplase in acute ischaemic stroke (ECASS II)[J]. Lancet, 1998, 352(9136): 1245-1251. DOI: 10.1016/s0140-6736(98)08020-9.
[5]
VAN DER STEEN W, VAN DER ENDE N A M, LUIJTEN S P R, et al. Type of intracranial hemorrhage after endovascular stroke treatment: association with functional outcome[J]. J Neurointerv Surg, 2023, 15(10): 971-976. DOI: 10.1136/jnis-2022-019474.
[6]
HE J, FU F, ZHANG W, et al. Prognostic significance of the clinical and radiological hemorrhagic transformation subtypes in acute ischaemic stroke: A systematic review and meta-analysis[J]. Eur J Neurol, 2022, 29(11): 3449-3459. DOI: 10.1111/ene.15482.
[7]
WU F, WU Q, ZHOU Q, et al. Predicting haemorrhagic transformation through serum biochemical indices for patients after endovascular treatment: a retrospective study.[J]. Neurol Neurochir Pol, 2024, 58(3): 300-315. DOI: 10.5603/PJNNS.97759.
[8]
LIU C, ZHANG Y, NIU L, et al. High Level of the Fibrin Degradation Products at Admission Predicts Parenchymal Hematoma and Unfavorable Outcome of Ischemic Stroke After Intravenous Thrombolysis[J/OL]. Front Neurol, 2022, 12: 797394 [2025-02-17]. https://www.frontiersin.org/articles/10.3389/fneur.2021.797394. DOI: 10.3389/fneur.2021.797394.
[9]
HUDÁK L, KOVÁCS B K, BAGOLY Z, et al. Clinicopathological Observations in Acute Stroke Patients Treated with Intravenous Thrombolysis[J]. J Clin Med, 2024, 13(19): 6012-6012. DOI: 10.3390/JCM13196012.
[10]
ISSAIY M, ZAREI D, KOLAHI S, et al. Machine learning and deep learning algorithms in stroke medicine: a systematic review of hemorrhagic transformation prediction models.[J/OL]. J Neurol, 2024, 272(1): 37 [2025-02-17]. https://link.springer.com/article/10.1007/s00415-024-12810-6. DOI: 10.1007/S00415-024-12810-6.
[11]
XU Y, LI X, WU D, et al. Machine Learning-Based Model for Prediction of Hemorrhage Transformation in Acute Ischemic Stroke After Alteplase[J/OL]. Front Neurol, 2022, 13: 897903 [2025-02-17]. https://www.frontiersin.org/articles/10.3389/fneur.2022.897903. DOI: 10.3389/fneur.2022.897903.
[12]
XU J J, WANG Z N, XU X, et al. The study of machine learning based on DWI infarction and cerebrospinal fluid in predicting malignant edema after thrombectomy for acute stroke[J]. Journal of Medical Imaging, 2024, 34(1): 7-10.
[13]
XUEWEN L, CHANGYAN X, CHENGMING S, et al. Machine learning predicts the risk of hemorrhagic transformation of acutecerebral infarction and in-hospital death[J/OL]. Comput Methods Programs Biomed, 2023, 237: 107582 [2025-02-17]. https://pubmed.ncbi.nlm.nih.gov/37156021/. DOI: 10.1016/J.CMPB2023.107580.
[14]
LIU J M, LUO H W, YU P F, et al. Construction and evaluation of a machine-learning-based model for predic-ting healthcare-associated infection in patients with acute ischemic stroke[J]. Chin J Infec Contl, 2023, 22(2): 129-135. DOI: 10.12138/j.iSSN.1671-9638.20233300.
[15]
LI X, XU C, SHANG C, et al. Machine learning predicts the risk of hemorrhagic transformation of acute cerebral infarction and in-hospital death[J/OL]. Comput Methods Programs Biomed, 2023, 237: 107582 [2025-02-17]. https://www.sciencedirect.com/science/article/pii/S0169260723002136. DOI: 10.1016/j.cmpb.2023.107582.
[16]
KASSNER A, LIU F, THORNHILL R E, et al. Prediction of hemorrhagic transformation in acute ischemic stroke using texture analysis of postcontrast T1-weighted MR images[J]. J Magn Reson Imaging, 2009, 30(5): 933-941. DOI: 10.1002/jmri.21940.
[17]
BOUTS M J, TIEBOSCH I A, RUDRAPATNA U S, et al. Prediction of hemorrhagic transformation after experimental ischemic stroke using MRI-based algorithms[J]. J Cereb Blood Flow Metab, 2017, 37(8): 3065-3076. DOI: 10.1177/0271678X16683692.
[18]
Neurology Branch of Chinese Medical Association, Cerebrovascular Disease Group of Neurology Branch of Chinese Medical Association. Chinese guidelines for diagnosis and treatment of acute ischemic stroke 2018[J]. Chin J Neurol, 2018, 51(9): 666-682. DOI: 10.3760/cma.j.issn.1006-7876.2018.09.004.
[19]
VAN KRANENDONK K R, TREURNIET K M, BOERS A M M, et al. Hemorrhagic transformation is associated with poor functional outcome in patients with acute ischemic stroke due to a large vessel occlusion[J]. J Neurointerv Surg, 2019, 11(5): 464-468. DOI: 10.1136/neurintsurg-2018-014141.
[20]
LIU Z, HUANG X B, PENG M Y, et al. Construction of a nomogram model for hemorrhagic transformation risk after mechanical thrombectomy in patients with acute stroke[J]. Chin J Magn Reson Imaging, 2022, 13(4): 15-19. DOI: 10.12015/issn.1674-8034.2022.04.003.
[21]
PFEILSCHIFTER W, SPITZER D, CZECH-ZECHMEISTER B, et al. Increased Risk of Hemorrhagic Transformation in Ischemic Stroke Occurring During Warfarin Anticoagulation[J]. Stroke, 2011, 42(4): 1116-1121. DOI: 10.1161/STROKEAHA.110.604652.
[22]
KOELLHOFFER E C, MCCULLOUGH L D. The effects of estrogen in ischemic stroke[J]. Transl Stroke Res, 2013, 4(4): 390-401. DOI: 10.1007/s12975-012-0230-5.
[23]
SUN J C, LAM C, CHRISTIE L, et al. Risk factors of hemorrhagic transformation in acute ischaemic stroke: A systematic review and meta-analysis[J/OL]. Front Neurol, 2023, 14: 1079205 [2025-02-17]. https://www.frontiersin.org/articles/10.3389/fneur.2023.1079205. DOI: 10.3389/fneur.2023.1079205.
[24]
MARINI S, MERINO J, MONTGOMERY B E, et al. Mendelian Randomization Study of Obesity and Cerebrovascular Disease[J]. Ann Neurol, 2020, 87(4): 516-524. DOI: 10.1002/ana.25686.
[25]
HORN J W, FENG T, MØRKEDAL B, et al. Obesity and Risk for First Ischemic Stroke Depends on Metabolic Syndrome: The HUNT Study[J]. Stroke, 2021, 52(11): 3555-3561. DOI: 10.1161/STROKEAHA.120.033016.
[26]
ELABI O F, KARAMPATSI D, VERCALSTEREN E, et al. DPP-4 Inhibitor and Sulfonylurea Differentially Reverse Type 2 Diabetes-Induced Blood-Brain Barrier Leakage and Normalize Capillary Pericyte Coverage[J]. Diabetes, 2023, 72(3): 405-414. DOI: 10.2337/db22-0674.
[27]
HOU L, XU M, YU Y, et al. Exploring the causal pathway from ischemic stroke to atrial fibrillation: a network Mendelian randomization study[J/OL]. Mol Med, 2020, 26(1): 7 [2025-02-17]. https://molmed.biomedcentral.com/articles/10.1186/s10020-019-0133-y. DOI: 10.1186/s10020-019-0133-y.
[28]
SIDDIQI T J, USMAN M S, RASHID A M, et al. Clinical Outcomes in Hypertensive Emergency: A Systematic Review and Meta-Analysis[J/OL]. J Am Heart Assoc, 2023, 12(14): e029355 [2025-02-17]. https://www.ahajournals.org/doi/10.1161/JAHA.122.029355. DOI: 10.1161/JAHA.122.029355.
[29]
XU J, LIU Y, WANG A, et al. Blood pressure fluctuation pattern and stroke outcomes in acute ischemic stroke[J]. Hypertens Res, 2019, 42(11): 1776-1782. DOI: 10.1038/s41440-019-0292-9.
[30]
CHEN X, LIU H, YE H, et al. Systolic blood pressure trajectories after acute ischemic strokes and clinical outcomes: A systematic review[J]. J Clin Hypertens (Greenwich), 2022, 24(8): 963-970. DOI: 10.1111/jch.14537.
[31]
LI C, ZHANG Y, XU T, et al. Systolic Blood Pressure Trajectories in the Acute Phase and Clinical Outcomes in 2-Year Follow-up Among Patients With Ischemic Stroke[J]. Am J Hypertens, 2019, 32(3): 317-325. DOI: 10.1111/jch.14537.
[32]
LIU S, GAO J, ZHAO H, et al. Impact of blood pressure variability on hemorrhagic transformation post-rt-PA thrombolysisin patients with acute ischemic stroke[J/OL]. SAGE Open Med, 2024, 12: 20503121241283881 [2025-02-17]. https://journals.sagepub.com/doi/10.1177/20503121241283881. DOI: 10.1177/202503121241283881.
[33]
LUO X, CHENG Y, WU C, et al. An interpretable machine learning-based prediction model for risk of death for patients with ischemic stroke in intensive care unit[J]. Journal of Southern Medical University, 2023(7): 1241-1247. DOI: 10.12122/j.issn.1673-4254.2023.07.21.
[34]
XU C F, HUANG Y, ZHANG K Z. Acute ischemic stroke patients with onset time with in 4.5 h identified by machine learning combined with magnetic resonance diffusion weighted imaging[J]. Journal of Jilin University (Medical Science Edition), 2023, 49(6): 1635-1641. DOI: 10.13481/j.1671-587X.20230631.
[35]
DAGHER R, OZKARA B B, KARABACAK M, et al. Artificial intelligence/machine learning for neuroimaging to predict hemorrhagic transformation: Systematic review/meta-analysis.[J]. J Neuroimaging, 2024, 34(5): 505-514. DOI: 10.1111/jon.13223.

PREV Correlation between Willis ring integrity and plaque characteristics and multiple infarctions in patients with acute ischemic stroke
NEXT Construction and validation of a non-invasive differentiation model for glioblastoma and primary central nervous system lymphoma based on clinical-multimodal magnetic resonance imaging radiomics
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn