Abstract

Accurate and efficient image classification is a critical requirement for intelligent medical systems, especially under performance and resource constraints. Traditional manual interpretation and shallow learning models are often inefficient and lack sufficient classification accuracy. To address these challenges, this paper proposes a multi-scale deep learning framework based on an improved ResNet architecture for medical image classification, achieving high performance. In the proposed framework, By introducing a multi-scale feature extraction module at the network input and combining it with a residual downsampling structure based on Inception-v2, the ResNet34 backbone network's ability to represent and extract multi-scale features is effectively improved. Furthermore, a channel attention mechanism is introduced to adaptively reweight feature channels, enabling the model to emphasize more informative feature representations while suppressing redundant features. Experimental evaluation employs five-fold cross-validation using a brain tumor image dataset. Experimental results show that the framework achieves an average accuracy of approximately 98.8% while maintaining high classification performance, which is about 1% higher than the original ResNet34, and the number of model parameters is reduced to about 80% of the baseline model. These results demonstrate that the proposed method effectively improves classification performance while enhancing computational efficiency, making it suitable for performance-constrained intelligent image analysis systems.

References

• XUllah, N., Khan, J. A., Khan, M. S., Khan, W., Hassan, I., Obayya, M., ... & Salama, A. S. (2022). An effective approach to detect and identify brain tumors using transfer learning. Applied Sciences, 12(11), 5645.
• Asif, S., Zhao, M., Chen, X., & Zhu, Y. (2023). BMRI-NET: A deep stacked ensemble model for multi-class brain tumor classification from MRI images. Interdisciplinary Sciences: Computational Life Sciences, 15(3), 499-514.
• Abd El Kader, I., Xu, G., Shuai, Z., Saminu, S., Javaid, I., & Salim Ahmad, I. (2021). Differential deep convolutional neural network model for brain tumor classification. Brain Sciences, 11(3), 352.
• Xie, Y., Zaccagna, F., Rundo, L., Testa, C., Agati, R., Lodi, R., ... & Tonon, C. (2022). Convolutional neural network techniques for brain tumor classification (from 2015 to 2022): Review, challenges, and future perspectives. Diagnostics, 12(8), 1850.
• Li, Z., Liu, F., Yang, W., Peng, S., & Zhou, J. (2021). A survey of convolutional neural networks: analysis, applications, and prospects. IEEE transactions on neural networks and learning systems, 33(12), 6999-7019.
• Hossain, T., Shishir, F. S., Ashraf, M., Al Nasim, M. A., & Shah, F. M. (2019, May). Brain tumor detection using convolutional neural network. In 2019 1st international conference on advances in science, engineering and robotics technology (ICASERT) (pp. 1-6). IEEE.
• Abiwinanda, N., Hanif, M., Hesaputra, S. T., Handayani, A., & Mengko, T. R. (2018, May). Brain tumor classification using convolutional neural network. In World Congress on Medical Physics and Biomedical Engineering 2018: June 3–8, 2018, Prague, Czech Republic (Vol. 1) (pp. 183-189). Singapore: Springer Nature Singapore.
• Vankdothu, R., Hameed, M. A., & Fatima, H. (2022). A brain tumor identification and classification using deep learning based on CNN-LSTM method. Computers and Electrical Engineering, 101, 107960.
• Ayadi, W., Elhamzi, W., Charfi, I., & Atri, M. (2021). Deep CNN for brain tumor classification. Neural processing letters, 53(1), 671-700.
• Khan, H. A., Jue, W., Mushtaq, M., & Mushtaq, M. U. (2021). Brain tumor classification in MRI image using convolutional neural network. Mathematical Biosciences and Engineering.
• Kumar, R. L., Kakarla, J., Isunuri, B. V., & Singh, M. (2021). Multi-class brain tumor classification using residual network and global average pooling. Multimedia Tools and Applications, 80(9), 13429-13438.
• He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 770-778).
• Ding, X., Zhang, X., Han, J., & Ding, G. (2022). Scaling up your kernels to 31x31: Revisiting large kernel design in cnns. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (pp. 11963-11975).
• Díaz-Pernas, F. J., Martínez-Zarzuela, M., Antón-Rodríguez, M., & González-Ortega, D. (2021, February). A deep learning approach for brain tumor classification and segmentation using a multiscale convolutional neural network. In Healthcare (Vol. 9, No. 2, p. 153). MDPI.
• Gao, S. H., Cheng, M. M., Zhao, K., Zhang, X. Y., Yang, M. H., & Torr, P. (2019). Res2net: A new multi-scale backbone architecture. IEEE transactions on pattern analysis and machine intelligence, 43(2), 652-662.
• Hu, J., Shen, L., & Sun, G. (2018). Squeeze-and-excitation networks. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 7132-7141).
• Deepak, S., & Ameer, P. M. (2019). Brain tumor classification using deep CNN features via transfer learning. Computers in biology and medicine, 111, 103345.
• Singh, P., Sizikova, E., & Cirrone, J. (2022). CASS: cross architectural self-supervision for medical image analysis. arXiv preprint arXiv:2206.04170.
• Li, Y., Yan, X., Xiao, M., Wang, W., & Zhang, F. (2023, December). Investigation of creating accessibility linked data based on publicly available accessibility datasets. In Proceedings of the 2023 13th International Conference on Communication and Network Security (pp. 77-81).
• Shorten, C., & Khoshgoftaar, T. M. (2019). A survey on image data augmentation for deep learning. Journal of big data, 6(1), 1-48.
• Swati, Z. N. K., Zhao, Q., Kabir, M., Ali, F., Ali, Z., Ahmed, S., & Lu, J. (2019). Brain tumor classification for MR images using transfer learning and fine-tuning. Computerized Medical Imaging and Graphics, 75, 34-46.
• Kaplan, K., Kaya, Y., Kuncan, M., & Ertunç, H. M. (2020). Brain tumor classification using modified local binary patterns (LBP) feature extraction methods. Medical hypotheses, 139, 109696.
• Huang, M., Xiong, Z. Y., & Zhu, J. L. (2023). MRI brain tumor classification based on multiscale residual network. Chinese Journal of Magnetic Resonance Imaging, 14(1), 124-129.