Deep learning-based prediction of hematoma expansion using a single brain computed tomographic slice in patients with spontaneous intracerebral hemorrhages

We aimed to predict hematoma expansion in Intracerebral hemorrhage (ICH) patients by using the deep learning technique in this work. We retrospectively collected ICH patients between May 2015 and May 2019. Head CT scans were performed at admission, 6 hours, 24 hours, and 72 hours after admission. CT scans were mandatory when neurological deficits occurred. Uni- and multi-variate analyses were conducted to illustrate the association between clinical variables and hematoma expansion. Convolutional Neural Network (CNN) was adopted to predict hematoma expansion based on brain CT slices. In addition, five machine learning methods, including Support Vector Machine (SVM), Multi-Layer Perceptron (MLP), Naive Bayes (NB), Decision Tree (DT), and Random Forest (RF), are also performed to predict hematoma expansion based on clinical variables for comparisons. A total of 223 patients were included. It was revealed that patients’ older age (odds ratio (OR) [95 Confident Interval, CI]: 1.783 [1.417-1.924]), cerebral hemorrhage and breaking into the ventricle (2.524 [1.291-1.778]), coagulopathy (2.341 [1.677-3.454]), baseline NIHSS (1.545 [1.132-3.203]) and GCS scores (0.782 [0.432-0.918]) independently associated with hematoma expanding. After 4-5 epochs, the CNN framework was well-trained. The average sensitivity, specificity, and accuracy of CNN prediction are 0.9197, 0.8837, and 0.9058, respectively. Compared with five machine learning methods based on clinical variables, CNN can also achieve better performance. More than 90% of hematoma with or without expansion can be precisely classified by deep learning technology within this study, which is better than other methods based on clinical variables only. Deep learning technology could favorably predict hematoma expansion from non-contrast CT scan images.