Rapid prediction of hemorrhagic transformation after endovascular thrombectomy: a multimodal model in patients with post-thrombectomy cerebral hyperdensities - Report - MDSpire

Rapid prediction of hemorrhagic transformation after endovascular thrombectomy: a multimodal model in patients with post-thrombectomy cerebral hyperdensities

  • By

  • Ziwen Wang

  • Guolan Song

  • Ying Tang

  • Jiahong Xu

  • Junli Wang

  • Qingdian Cong

  • Jingjing Fu

  • Yue Wang

  • Jibo Hu

  • Leling Tu

  • Song Cheng

  • Jian Ding

  • Sheng Hu

  • July 7, 2026

  • 0 min

Share

Clinical Report: Swift Assessment of Hemorrhagic Transformation Risk Following EVT

Overview

This study presents a novel multimodal deep learning framework for predicting hemorrhagic transformation (HT) in patients post-endovascular thrombectomy (EVT). The model demonstrated superior predictive performance compared to human neuroradiologists.

Background

Hemorrhagic transformation is a significant complication following EVT for acute ischemic stroke, often leading to poorer functional outcomes. Early identification of patients at high risk for HT is crucial for effective neuromonitoring and management. Current imaging techniques can be time-consuming and may not provide timely risk assessments, necessitating the development of rapid predictive tools.

Data Highlights

MetricValue
Overall incidence of HT70.5% (277/393)
Best AUC of Fusion_Transformer model0.803 (95% CI: 0.708–0.898)
Sensitivity0.672
Specificity0.955
AUC of senior neuroradiologist0.707 (95% CI: 0.596–0.818)

Key Findings

  • The incidence of hemorrhagic transformation after EVT was found to be 70.5% in the study cohort.
  • The Fusion_Transformer model achieved an AUC of 0.803, indicating strong predictive capability.
  • Sensitivity and specificity of the model were reported as 0.672 and 0.955, respectively.
  • The automated model outperformed human neuroradiologists in predicting HT.
  • The study utilized a 2.5D multimodal approach, integrating imaging features with clinical variables.

Clinical Implications

The developed multimodal deep learning framework provides a rapid and accurate method for predicting hemorrhagic transformation.

Conclusion

This study demonstrates the efficacy of a deep learning model in predicting hemorrhagic transformation post-EVT.

Related Resources & Content

  1. Author(s)/Org, Source, Year -- Title
  2. Author(s)/Org, Source, Year -- Title
  3. Author(s)/Org, Source, Year -- Title
  4. Author(s)/Org, Source, Year -- Title
  5. Author(s)/Org, Source, Year -- Title
  6. Author(s)/Org, Source, Year -- Title
  7. Author(s)/Org, Source, Year -- Title
  8. 2026 Guideline for the Early Management of Patients With Acute Ischemic Stroke: A Guideline From the American Heart Association/American Stroke Association | Stroke
  9. Frontiers | Diagnostic performance of dual-energy CT for differentiating acute intracranial hemorrhage from contrast staining: a systematic review and meta-analysis
  10. Large-Core Ischemic Stroke Endovascular Treatment: A Science Advisory From the American Heart Association

Original Source(s)

Related Content