Prediction of Prognosis in Acute Ischemic Stroke Based on Multimodal MRI Radiomics and Deep Learning.

Motivation: Acute ischemic stroke(AIS) is associated with high rates of disability and mortality, there are currently no reliable methods for early prediction of poor prognosis in AIS. Goal(s): Exploring the value of radiomics and deep learning based on multimodal MRI in predicting poor prognosis in Acute ischemic stroke. Approach: This study combines the Clinic Model, Radiomics Model, and Deep Learning Model to develop the CRD Model (Clinic-Radiomics-Deep Learning). The predictive efficacy of each model for poor prognosis is evaluated using receiver operating characteristic curves. Results: The CRD model, based on multimodal MRI, demonstrates high diagnostic efficacy and reliability in predicting poor prognosis in AIS Impact: These findings suggest that CRD model holds considerable potential for aiding clinicians in risk assessment and decision-making for AIS patients.