An efficient deep neural network for automatic classification of acute intracranial hemorrhages in brain CT scans

• Published in: Computers in biology and medicine Vol. 176; p. 108587
• Main Authors: Chen, Yu-Ruei, Chen, Chih-Chieh, Kuo, Chang-Fu, Lin, Ching-Heng
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.06.2024; Elsevier Limited
• Subjects: Artificial intelligence; Artificial neural networks; Brain; Brain - diagnostic imaging; Brain research; Classification; Computational neuroscience; Computed tomography; Databases, Factual; Datasets; Deep Learning; Efficient neural network; Hemorrhage; Humans; Intracranial hemorrhage; Intracranial Hemorrhages - classification; Intracranial Hemorrhages - diagnostic imaging; Machine learning; Medical imaging; Multi-receptive field; Neural networks; Neural Networks, Computer; Neuroimaging; Receptive field; Tomography, X-Ray Computed - methods; Deep learning; Efficient neural network; Multi-receptive field; Intracranial hemorrhage
• ISSN: 0010-4825; 1879-0534
• DOI: 10.1016/j.compbiomed.2024.108587

Recent advancements in deep learning models have demonstrated their potential in the field of medical imaging, achieving remarkable performance surpassing human capabilities in tasks such as classification and segmentation. However, these modern state-of-the-art network architectures often demand substantial computational resources, which limits their practical application in resource-constrained settings. This study aims to propose an efficient diagnostic deep learning model specifically designed for the classification of intracranial hemorrhage in brain CT scans. Our proposed model utilizes a combination of depthwise separable convolutions and a multi-receptive field mechanism to achieve a trade-off between performance and computational efficiency. The model was trained on RSNA datasets and validated on CQ500 dataset and PhysioNet dataset. Through a comprehensive comparison with state-of-the-art models, our model achieves an average AUROC score of 0.952 on RSNA datasets and exhibits robust generalization capabilities, comparable to SE-ResNeXt, across other open datasets. Furthermore, the parameter count of our model is just 3 % of that of MobileNet V3. This study presents a diagnostic deep-learning model that is optimized for classifying intracranial hemorrhages in brain CT scans. The efficient characteristics make our proposed model highly promising for broader applications in medical settings. [Display omitted] •An efficient deep learning diagnostic model for the classification of Intracerebral hemorrhage.•Using only 3 % of MobileNet V3's parameters to achieve an impressive classification performance.•The proposed model shows robustness across various datasets.