Intracranial Hemorrhage Detection in Head CT Using Double-Branch Convolutional Neural Network, Support Vector Machine, and Random Forest

Abstract

Brain hemorrhage is a severe threat to human life, and its timely and correct diagnosis and treatment are of great importance. Multiple types of brain hemorrhage are distinguished depending on the location and character of bleeding. The main division covers five subtypes: subdural, epidural, intraventricular, intraparenchymal, and subarachnoid hemorrhage. This paper presents an approach to detect these intracranial hemorrhage types in computed tomography images of the head. The model trained for each hemorrhage subtype is based on a double-branch convolutional neural network of ResNet-50 architecture. It extracts features from two chromatic representations of the input data: a concatenation of the image normalized in different intensity windows and a stack of three consecutive slices creating a 3D spatial context. The joint feature vector is passed to the classifier to produce the final decision. We tested two tools: the support vector machine and the random forest. The experiments involved 372,556 images from 11,454 CT series of 9997 patients, with each image annotated with labels related to the hemorrhage subtypes. We validated deep networks from both branches of our framework and the model with either of two classifiers under consideration. The obtained results justify the use of a combination of double-source features with the random forest classifier. The system outperforms state-of-the-art methods in terms of F1 score. The highest detection accuracy was obtained in intraventricular (96.7%) and intraparenchymal hemorrhages (93.3%).

Artificial Intelligence-Enabled Medical Analysis for Intracranial Cerebral Hemorrhage Detection and Classification

“…Finally, a computation time (CT) analysis of the AIMA-ICHDC technique with other ICH detection models is shown in Table 3 and Figure 12 [ 27 – 31 ]. From the figure, it can be observed that the SVM, deep CNN, and WEM-DCNN techniques have required higher CTs of 1.483 min, 1.284 min, and 1.268 min, respectively.…”

Deep Learning Applied to Intracranial Hemorrhage Detection

“…Radiologist usually needs from 15 to 30 minutes to diagnose a patient and inform about it in writing. Diagnosis of ICH has been a subject of several papers and is a field of intensive research using traditional techniques [8] and deep learning [22][23][24][25]. The solution [36] has AUC 0.846 (CI 0.837-0.856).…”

“…Recently, some studies have been published on this topic, and many researchers have started to pay attention to it. Among many others, we can cite [22], where the authors use a fully convolutional neural network for classification and segmentation with examination of ICH with computed tomographies; [23] where the InceptionV3 and DenseNet Deep Learning models for dealing with CT; or [24], where the authors combine convolutional neural networks with other machine learning techniques to deal with ICH detection. In [25], the authors use a Dense U-net architecture for the detection of ICH.…”

Classification of Intracranial Hemorrhage CT images for Stroke Analysis with Transformed and Image-based GLCM Features

“…Accurate and fast diagnosis is crucial as 50% of patients' death happen within 24 hours 35 to 52 % of patients enters in critical zone and die within a month as mortality reaches up to 60% after 30 days [2].To perform accurate diagnosis, first stage examination is performed by a physician using non-contrast computed tomography (CT) analysis. CT image analysis helps to identify and locate the bleeding location inside the brain [3].…”