Attention-based MobileNet for Effective Recognition of Brain Tumors

Tianyi Du

doi:10.54097/w7h5x374

Authors

Tianyi Du

DOI:

https://doi.org/10.54097/w7h5x374

Keywords:

component; brain tumor; attention; SENet; MobileNet.

Abstract

Brain tumor is a very dangerous disease, which often causes human death. When the consulting doctor is inexperienced, it is easy to misdiagnose and cause serious consequences. Therefore, accurate and timely identification of brain tumors from Magnetic Resonance Imaging (MRI) images is crucial due to the life-threatening nature of this disease and the potential consequences of misdiagnosis. To address this challenge, this research employs a neural network model for brain tumor identification, incorporating model fusion techniques to enhance accuracy. Specifically, the study compares the performance of two models, namely MobileNet and SE+MobileNet Block, in recognizing MRI images. The SE+MobileNet Block model integrates a section of SENet into MobileNet, introducing a 3 x 3 convolutional layer, two Fully Connected (FC) layers, and the last two activation functions (Rule and Sigmoid), effectively incorporating an attention mechanism. Comparative analysis using accuracy and loss values demonstrates that the inclusion of the attention mechanism in the MobileNet model (i.e., the SE+MobileNet Block model) improves the recognition accuracy of brain tumor MRI.

Downloads

Download data is not yet available.

References

[1] CBTRUS, “Central Brain Tumor Registry”

https://www.cbtrus.org/www.cbtrus.org/index.html, 2019.

[2] Y. Qiu, C. S. Chang, J. L. Yan, L. Ko and T. S. Chang, "Semantic Segmentation of Intracranial Hemorrhages in Head CT Scans," 2019 IEEE 10th International Conference on Software Engineering and Service Science (ICSESS), Beijing, China, 2019, pp. 112-115, doi: 10.1109/ICSESS47205.2019.9040733.

[3] D. V. Gore and V. Deshpande, "Comparative Study of various techniques using Deep Learning for Brain Tumor Detection," 2020 International Conference for Emerging Technology (INCET), Belgaum, India, 2020, pp. 1-4, doi: 10.1109/INCET49848.2020.9154030.

[4] H. Hu, X. Li, W. Yao and Z. Yao, “Brain Tumor Diagnose Applying CNN through MRI,” 2021 2nd International Conference on Artificial Intelligence and Computer Engineering (ICAICE), Hangzhou, China, 2021, pp. 430-434, doi: 10.1109/ICAICE54393.2021.00090.

[5] S. Montaha, S. Azam, A. K. M. R. H. Rafid, M. Z. Hasan, A. Karim and A. Islam, “TimeDistributed-CNN-LSTM: A Hybrid Approach Combining CNN and LSTM to Classify Brain Tumor on 3D MRI Scans Performing Ablation Study,” in IEEE Access, vol. 10, pp. 60039-60059, 2022.

[6] Kaggle, “Br35H::Brain Tumor Detection 2020”, 2021, https://www.kaggle.com/datasets/ahmedhamada0/brain-tumor-detection?select=Br35H-Mask-RCNN.

[7] Y. Qiu, et al. "Pose-guided matching based on deep learning for assessing quality of action on rehabilitation training," Biomedical Signal Processing and Control, 2022, 72: 103323.

[8] Y. Li, J. Zeng, S. Shan and X. Chen, "Occlusion Aware Facial Expression Recognition Using CNN With Attention Mechanism," in IEEE Transactions on Image Processing, vol. 28, no. 5, pp. 2439-2450, May 2019, doi: 10.1109/TIP.2018.2886767.

[9] M. Abid, P. Compagnon and G. Lefebvre, "Improved CNN-based Magnetic Indoor Positioning System using Attention Mechanism," 2021 International Conference on Indoor Positioning and Indoor Navigation (IPIN), Lloret de Mar, Spain, 2021, pp. 1-8, doi: 10.1109/IPIN51156.2021.9662602.

[10] A. Howard, et al. "Mobilenets: Efficient convolutional neural networks for mobile vision applications," arXiv preprint arXiv:1704.04861, 2017.