Hossein Mohammadian Foroushani1, Ali Hamzehloo2, Atul Kumar2, Yasheng Chen2, Laura Heitsch3, Agnieszka Slowik4, Daniel Strbian5, Jin-Moo Lee2, Daniel S Marcus6, Rajat Dhar7. 1. Department of Electrical and Systems Engineering, Washington University in St. Louis McKelvey School of Engineering, 1 Brookings Drive, St. Louis, MO, 63130-4899, USA. 2. Department of Neurology, Washington University in St. Louis School of Medicine, 660 S Euclid Avenue, Campus, Box 8111, St. Louis, MO, 63110, USA. 3. Department of Emergency Medicine, Washington University in St. Louis School of Medicine, 660 S. Euclid Ave, Campus, Box 8072, St. Louis, MO, 63110, USA. 4. Department of Neurology, Jagiellonian University Medical College, Kraków, Poland. 5. Department of Neurology, Helsinki University Hospital, Helsinki, Finland. 6. Department of Radiology, Washington University in St. Louis School of Medicine, 525 Scott Ave, Campus, Box 8225, St. Louis, MO, 63110, USA. 7. Department of Neurology, Washington University in St. Louis School of Medicine, 660 S Euclid Avenue, Campus, Box 8111, St. Louis, MO, 63110, USA. dharr@wustl.edu.
Abstract
BACKGROUND: Malignant cerebral edema is a devastating complication of stroke, resulting in deterioration and death if hemicraniectomy is not performed prior to herniation. Current approaches for predicting this relatively rare complication often require advanced imaging and still suffer from suboptimal performance. We performed a pilot study to evaluate whether neural networks incorporating data extracted from routine computed tomography (CT) imaging could enhance prediction of edema in a large diverse stroke cohort. METHODS: An automated imaging pipeline retrospectively extracted volumetric data, including cerebrospinal fluid (CSF) volumes and the hemispheric CSF volume ratio, from baseline and 24 h CT scans performed in participants of an international stroke cohort study. Fully connected and long short-term memory (LSTM) neural networks were trained using serial clinical and imaging data to predict those who would require hemicraniectomy or die with midline shift. The performance of these models was tested, in comparison with regression models and the Enhanced Detection of Edema in Malignant Anterior Circulation Stroke (EDEMA) score, using cross-validation to construct precision-recall curves. RESULTS: Twenty of 598 patients developed malignant edema (12 required surgery, 8 died). The regression model provided 95% recall but only 32% precision (area under the precision-recall curve [AUPRC] 0.74), similar to the EDEMA score (precision 28%, AUPRC 0.66). The fully connected network did not perform better (precision 33%, AUPRC 0.71), but the LSTM model provided 100% recall and 87% precision (AUPRC 0.97) in the overall cohort and the subgroup with a National Institutes of Health Stroke Scale (NIHSS) score ≥ 8 (p = 0.0001 vs. regression and fully connected models). Features providing the most predictive importance were the hemispheric CSF ratio and NIHSS score measured at 24 h. CONCLUSIONS: An LSTM neural network incorporating volumetric data extracted from routine CT scans identified all cases of malignant cerebral edema by 24 h after stroke, with significantly fewer false positives than a fully connected neural network, regression model, and the validated EDEMA score. This preliminary work requires prospective validation but provides proof of principle that a deep learning framework could assist in selecting patients for surgery prior to deterioration.
BACKGROUND: Malignant cerebral edema is a devastating complication of stroke, resulting in deterioration and death if hemicraniectomy is not performed prior to herniation. Current approaches for predicting this relatively rare complication often require advanced imaging and still suffer from suboptimal performance. We performed a pilot study to evaluate whether neural networks incorporating data extracted from routine computed tomography (CT) imaging could enhance prediction of edema in a large diverse stroke cohort. METHODS: An automated imaging pipeline retrospectively extracted volumetric data, including cerebrospinal fluid (CSF) volumes and the hemispheric CSF volume ratio, from baseline and 24 h CT scans performed in participants of an international stroke cohort study. Fully connected and long short-term memory (LSTM) neural networks were trained using serial clinical and imaging data to predict those who would require hemicraniectomy or die with midline shift. The performance of these models was tested, in comparison with regression models and the Enhanced Detection of Edema in Malignant Anterior Circulation Stroke (EDEMA) score, using cross-validation to construct precision-recall curves. RESULTS: Twenty of 598 patients developed malignant edema (12 required surgery, 8 died). The regression model provided 95% recall but only 32% precision (area under the precision-recall curve [AUPRC] 0.74), similar to the EDEMA score (precision 28%, AUPRC 0.66). The fully connected network did not perform better (precision 33%, AUPRC 0.71), but the LSTM model provided 100% recall and 87% precision (AUPRC 0.97) in the overall cohort and the subgroup with a National Institutes of Health Stroke Scale (NIHSS) score ≥ 8 (p = 0.0001 vs. regression and fully connected models). Features providing the most predictive importance were the hemispheric CSF ratio and NIHSS score measured at 24 h. CONCLUSIONS: An LSTM neural network incorporating volumetric data extracted from routine CT scans identified all cases of malignant cerebral edema by 24 h after stroke, with significantly fewer false positives than a fully connected neural network, regression model, and the validated EDEMA score. This preliminary work requires prospective validation but provides proof of principle that a deep learning framework could assist in selecting patients for surgery prior to deterioration.
Authors: Rajat Dhar; Ali Hamzehloo; Atul Kumar; Yasheng Chen; June He; Laura Heitsch; Agnieszka Slowik; Daniel Strbian; Jin-Moo Lee Journal: J Cereb Blood Flow Metab Date: 2021-05-20 Impact factor: 6.200
Authors: Hossein Mohammadian Foroushani; Ali Hamzehloo; Atul Kumar; Yasheng Chen; Laura Heitsch; Agnieszka Slowik; Daniel Strbian; Jin-Moo Lee; Daniel S Marcus; Rajat Dhar Journal: Neurocrit Care Date: 2020-07-29 Impact factor: 3.210
Authors: Matthew I Miller; Agni Orfanoudaki; Michael Cronin; Hanife Saglam; Ivy So Yeon Kim; Oluwafemi Balogun; Maria Tzalidi; Kyriakos Vasilopoulos; Georgia Fanaropoulou; Nina M Fanaropoulou; Jack Kalin; Meghan Hutch; Brenton R Prescott; Benjamin Brush; Emelia J Benjamin; Min Shin; Asim Mian; David M Greer; Stelios M Smirnakis; Charlene J Ong Journal: Neurocrit Care Date: 2022-05-09 Impact factor: 3.532