Accelerating Prediction of Malignant Cerebral Edema After Ischemic Stroke with Automated Image Analysis and Explainable Neural Networks

Keyphrases

• Ischemic Stroke 100%
• Automated Image Analysis 100%
• Malignant Cerebral Edema 100%
• Area Under the Precision-Recall Curve 100%
• Explainable Neural Network 100%
• Regression Model 50%
• Computed Tomography 50%
• Edema 50%
• National Institutes of Health Stroke Scale (NIHSS) 50%
• Cerebrospinal Fluid Volume 50%
• Hemicraniectomy 50%
• Long Short-term Memory Neural Network 50%
• Volumetric Data 50%
• Cerebrospinal Fluid 25%
• Clinical Data 25%
• Stroke Complications 25%
• Neural Network 25%
• Computational Imaging 25%
• Imaging Data 25%
• Patient Selection 25%
• Advanced Imaging 25%
• Enhanced Detection 25%
• Rare Complication 25%
• Volume Ratio 25%
• Midline Shift 25%
• Fully Connected 25%
• Herniation 25%
• Fully Convolutional Neural Network 25%
• Anterior Circulation Stroke 25%
• Deep Learning Framework 25%
• Automated Imaging 25%
• Imaging pipeline 25%
• Malignant Edema 25%
• Precision-recall Curve 25%
• Long Short-term Memory Model 25%
• Suboptimal Performance 25%
• Fully Connected Model 25%
• Fully Connected Network 25%
• Neural Network Regression Model 25%

Neuroscience

• Brain Ischemia 100%
• Neural Network 100%
• Cerebral Edema 100%
• Cerebrovascular Accident 100%
• Computed Tomography 42%
• Short-Term Memory 42%
• Midline Shift 14%