Deep symmetric three-dimensional convolutional neural networks for identifying acute ischemic stroke via diffusion-weighted images

Article type: Research Article

Authors: Cui, Liyuan^{a; 1} | Han, Shanhua^{b; 1} | Qi, Shouliang^{a; c; d; *} | Duan, Yang^e | Kang, Yan^{f; c} | Luo, Yu^b

Affiliations: [a] College of Medicine and Biological Information Engineering, Northeastern University, Shenyang, China | [b] Radiology Department, Shanghai Fourth People’s Hospital Affiliated to Tongji University School of Medicine, Shanghai, China | [c] Engineering Research Center of Medical Imaging and Intelligent Analysis, Ministry of Education, Northeastern University, Shenyang, China | [d] Key Laboratory of Intelligent Computing in Medical Image, Ministry of Education, Northeastern University, Shenyang, China | [e] Department of Radiology, General Hospital of Northern Theater Command, Shenyang, China | [f] Medical Device Innovation Research Center, Shenzhen Technology University, Shenzhen, China

Correspondence: [*] Corresponding author: Shouliang Qi, College of Medicine and Biological Information Engineering, Northeastern University, No. 195 Chuangxin Avenue, Hunnan District, Shenyang, 110169, China. ORCID: 0000-0003-0977-1939; Tel.: +86 24 8368 0230; Fax: +86 24 8368 1955; E-mail: [email protected].

Note: [1] Equal contribution as joint first author.

Abstract: BACKGROUND:Acute ischemic stroke (AIS) results in high morbidity, disability, and mortality. Early and automatic diagnosis of AIS can help clinicians administer the appropriate interventions. OBJECTIVE:To develop a deep symmetric 3D convolutional neural network (DeepSym-3D-CNN) for automated AIS diagnosis via diffusion-weighted imaging (DWI) images. METHODS:This study includes 190 study subjects (97 AIS and 93 Non-AIS) by collecting both DWI and Apparent Diffusion Coefficient (ADC) images. 3D DWI brain images are split into left and right hemispheres and input into two paths. A map with 125×253×14×12 features is extracted by each path of Inception Modules. After the features computed from two paths are subtracted through L-2 normalization, four multi-scale convolution layers produce the final predation. Three comparative models using DWI images including MedicalNet with transfer learning, Simple DeepSym-3D-CNN (each 3D Inception Module is replaced by a simple 3D-CNN layer), and L-1 DeepSym-3D-CNN (L-2 normalization is replaced by L-1 normalization) are constructed. Moreover, using ADC images and the combination of DWI and ADC images as inputs, the performance of DeepSym-3D-CNN is also investigated. Performance levels of all three models are evaluated by 5-fold cross-validation and the values of area under ROC curve (AUC) are compared by DeLong’s test. RESULTS:DeepSym-3D-CNN achieves an accuracy of 0.850 and an AUC of 0.864. DeLong’s test of AUC values demonstrates that DeepSym-3D-CNN significantly outperforms other comparative models (p < 0.05). The highlighted regions in the feature maps of DeepSym-3D-CNN spatially match with AIS lesions. Meanwhile, DeepSym-3D-CNN using DWI images presents the significant higher AUC than that either using ADC images or using DWI-ADC images based on DeLong’s test (p < 0.05). CONCLUSIONS:DeepSym-3D-CNN is a potential method for automatically identifying AIS via DWI images and can be extended to other diseases with asymmetric lesions.

Keywords: Acute ischemic stroke, diffusion-weighted imaging, deep learning, symmetric convolutional neural networks, automatic diagnosis

DOI: 10.3233/XST-210861

Journal: Journal of X-Ray Science and Technology, vol. 29, no. 4, pp. 551-566, 2021