Prediction Model of Hemorrhage Transformation in Patient with Acute Ischemic Stroke Based on Multiparametric MRI Radiomics and Machine Learning

Abstract

Intravenous thrombolysis is the most commonly used drug therapy for patients with acute ischemic stroke, which is often accompanied by complications of intracerebral hemorrhage transformation (HT). This study proposed to build a reliable model for pretreatment prediction of HT. Specifically, 5400 radiomics features were extracted from 20 regions of interest (ROIs) of multiparametric MRI images of 71 patients. Furthermore, a minimal set of all-relevant features were selected by LASSO from all ROIs and used to build a radiomics model through the random forest (RF). To explore the significance of normal ROIs, we built a model only based on abnormal ROIs. In addition, a model combining clinical factors and radiomics features was further built. Finally, the models were tested on an independent validation cohort. The radiomics model with 14 All-ROIs features achieved pretreatment prediction of HT (AUC = 0.871, accuracy = 0.848), which significantly outperformed the model with only 14 Abnormal-ROIs features (AUC = 0.831, accuracy = 0.818). Besides, combining clinical factors with radiomics features further benefited the prediction performance (AUC = 0.911, accuracy = 0.894). So, we think that the combined model can greatly assist doctors in diagnosis. Furthermore, we find that even if there were no lesions in the normal ROIs, they also provide characteristic information for the prediction of HT.

Application of radiomics in ischemic stroke

“… 43 To predict hemorrhage transformation, Meng et al. 44 introduced a multifaceted MRI radiomics-based model coupled with machine learning. Four distinct predictive models were cultivated within a primary training cohort and subsequently validated using an independent dataset.…”

Role of Brain Imaging in the Prediction of Intracerebral Hemorrhage Following Endovascular Therapy for Acute Stroke

“…Another approach for predicting ICH after thrombolysis combined MRI radiomics from different regions of interest using different parametric maps with clinical data into a random forest model, which achieved an improved area under the curve of 0.91. 73 The drawback of this approach is the time-consuming aspect of feature extraction from manually selected regions and the use of a single vendor's software for producing the parametric maps. Yu et al 74 investigated different machine learning models in patients treated with thrombolysis or EVT using pretreatment diffusion-weighted imaging and raw MRI perfusion imaging to predict the voxel-wise ICH probability.…”

Predictive Value of CT Perfusion in Hemorrhagic Transformation after Acute Ischemic Stroke: A Systematic Review and Meta-Analysis

“…A recent study exploring MRI and machine learning prediction models for predicting HT in AIS patients has shown that combining clinical variables, such as blood pressure and glucose, with radiomics features improved the prediction performance [91]. A logical next step for future research could be to establish a more individualized, precise, and stable multi-parameter diagnostic model for predicting HT with different etiologies, infarct sites, treatment approaches, treatment outcomes, and other factors.…”