Stroke classification from computed tomography scans using 3D convolutional neural network. (July 2022)

• Record Type: Journal Article
• Authors: Neethi, A.S.; Niyas, S.; Kannath, Santhosh Kumar; Mathew, Jimson; Anzar, Ajimi Mol; Rajan, Jeny
• Is Part Of: Biomedical signal processing and control. Volume 76(2022)
• Publication Date: 2022-07
• Subjects: 3D convolutional neural networks -- Deep learning -- Non contrast computed tomography -- Stroke classification; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2022.103720
• Languages: English
• ISSNs: 1746-8094

Abstract

Graphical abstract: Highlights: Stroke is a cerebrovascular condition which is mainly due to the disturbances in blood circulation to the brain regions. The CT scan is the most commonly preferred imaging modality to identify the type of stroke. But the classification task using CT scan is challenging due to the poor signal-to-noise ratio, low contrast to distinguish soft brain regions, interference from similar intensity regions, and motion artifacts. We propose a 3D-based fully convolutional classification model to identify stroke cases from non-contrast computed tomography (NCCT) images. The network architecture was decoupled into modules or blocks based on the functionality, but side by side unanimously contributing to the solution. We present a custom build pre-processing module for improving the efficacy of the CNN model by enhancing the feature space. To empirically determine the optimal sample depth for the 3D model, the data distribution of positive slices per case was plotted, and four different depths were selected for determining the final sample depth. The strided slicing augmentation technique was used to increase the 3D sample volume for training and overcoming the class imbalance problem. The proposed method shows superior classification performance over the state-of- the-art baseline 3D CNN model as well as the 3D variants of benchmark 2D CNN models. Abstract: Stroke is a cerebrovascular condition with a significant morbidity and mortality rate and causes physical disabilities for survivors. Once the symptoms are identified, it requires a time-critical diagnosis with the help of the most commonly available imaging techniques. Computed tomography (CT) scans are used worldwide for preliminary stroke diagnosis. It demands the expertise and experience of a radiologist to identify the stroke type, which is critical for initiating the treatment. This work attempts to gather those domain skills and build a model from CT scans to diagnose stroke. The non-contrast computed tomography (NCCT) scan of the brain comprises volumetric images or a 3D stack of image slices. So, a model that aims to solve the problem by targeting a 2D slice may fail to address the volumetric nature. We propose a 3D-based fully convolutional classification model to identify stroke cases from CT images that take into account the contextual longitudinal composition of volumetric data. We formulate a custom pre-processing module to enhance the scans and aid in improving the classification performance. Some of the significant challenges faced by 3D CNN are the less number of training samples, and the number of scans is mostly biased in favor of normal patients. In this work, the limitation of insufficient training volume and class imbalanced data have been rectified with the help of a strided slicing approach. A block-wise design was used to formulate the proposed network, with the initial part focusing on adjusting the dimensionality, at the same time retaining the features. Later on, the accumulated feature maps were effectively learned utilizing bundled convolutions and skip connections. The results of the proposed method were compared against 3D CNN stroke classification models on NCCT, various 3D CNN architectures on other brain imaging modalities, and 3D extensions of some of the classical CNN architectures. The proposed method achieved an improvement of 14.28% in the F1-score over the state-of-the-art 3D CNN stroke classification model.