Related Experiment Video
Updated: Jul 28, 2025
Predicting Amputation using Local Circulating Mononuclear Progenitor Cells in Angioplasty-treated Patients with Critical Limb Ischemia
Published on: September 22, 2020
Tissue Outcome Prediction in Patients with Proximal Vessel Occlusion and Mechanical Thrombectomy Using Logistic
Florian Welle1, Kristin Stoll1, Christina Gillmann2
1Neuroimaging Laboratory, Department of Neurology, University of Leipzig Medical Center, Leipzig, Germany.
Insights
A new multiparametric logistic model using perfusion CT data improves prediction of tissue outcome in stroke patients undergoing mechanical thrombectomy, outperforming simple thresholding methods.
Area of Science:
- Neuroimaging
- Radiology
- Stroke Medicine
Background:
- Perfusion CT aids patient selection for mechanical thrombectomy in acute stroke.
- Current selection often uses simple thresholding of perfusion maps, not fully utilizing data.
- Advanced modeling can potentially improve outcome prediction.
Purpose of the Study:
- To implement and evaluate a multiparametric logistic model for predicting tissue outcome in stroke patients.
- To compare the performance of this model against single-parameter thresholding methods.
- To assess the value of multimodal CT imaging data in outcome prediction.
Main Methods:
- A multiparametric mass-univariate logistic model was developed.
- Data from 405 stroke patients with anterior circulation occlusion undergoing thrombectomy were used.
- Model inputs included multimodal CT (perfusion, angiography, non-contrast), demographics, and clinical data, trained with recanalization and infarct data.
Main Results:
- Perfusion parameters (CBF, CBV, Tmax) were sufficient for tissue outcome prediction.
- The logistic model showed comparable volumetric accuracy but superior topographical accuracy (AUC) versus thresholding.
- Higher spatial accuracy (Dice index) was observed in internal cross-validation.
Conclusions:
- Multiparametric logistic prediction using perfusion CT data offers superior topographical accuracy compared to simple thresholding.
- Perfusion data is more valuable than non-contrast CT, CTA, or clinical information for outcome prediction.
- This approach holds potential for personalized biomarker development in mechanical thrombectomy for acute stroke.
Abstract:
Perfusion CT is established to aid selection of patients with proximal intracranial vessel occlusion for thrombectomy in the extended time window. Selection is mostly based on simple thresholding of perfusion parameter maps, which, however, does not exploit the full information hidden in the high-dimensional perfusion data. We implemented a multiparametric mass-univariate logistic model to predict tissue outcome based on data from 405 stroke patients with acute proximal vessel occlusion in the anterior circulation who underwent mechanical thrombectomy. Input parameters were acute multimodal CT imaging (perfusion, angiography, and non-contrast) as well as basic demographic and clinical parameters. The model was trained with the knowledge of recanalization status and final infarct localization. We found that perfusion parameter maps (CBF, CBV, and Tmax) were sufficient for tissue outcome prediction. Compared with single-parameter thresholding-based models, our logistic model had comparable volumetric accuracy, but was superior with respect to topographical accuracy (AUC of receiver operating characteristic). We also found higher spatial accuracy (Dice index) in an independent internal but not external cross-validation. Our results highlight the value of perfusion data compared with non-contrast CT, CT angiography and clinical information for tissue outcome-prediction. Multiparametric logistic prediction has high potential to outperform the single-parameter thresholding-based approach. In the future, the combination of tissue and functional outcome prediction might provide an individual biomarker for the benefit from mechanical thrombectomy in acute stroke care.
