Lã Hoàng Quý Phạm Toàn Định *

* Tác giả liên hệ (dinh.pt@vlu.edu.vn)

Abstract

This paper presents a Res-UNet-based deep learning architecture that incorporates residual learning, an attention mechanism, and dilated convolutions to improve boundary preservation and feature representation in medical images. The model was trained and evaluated using an MRI dataset containing 3,929 images. Experimental results indicate that the proposed approach provides better segmentation performance compared with several existing methods, including traditional U-Net, Fuzzy C-Means (FCM), Attention U-Net, UNet++, SegNet, Swin-UNet, TransUNet, and nnUNet. The model achieved BF = 98.0%, SSIM = 99.2%, ACC = 99.7%, Dice coefficient = 86.6%, and IoU = 76.6%. Furthermore, tumor presence classification based on the generated segmentation labels reached an accuracy of 99.7% on the test set. These findings suggest that the proposed architecture is a promising approach for medical image analysis.

Keywords: Brain tumour, diagnosis, MRI images, Res-UNet, segmentation

Tóm tắt

Trong nghiên cứu này, một kiến trúc mạng học sâu Res-UNet cải tiến, tích hợp quá trình học phần dư, cơ chế chú ý và tích chập giãn đã được đề xuất nhằm nâng cao khả năng bảo toàn biên và biểu diễn các đặc trưng của ảnh. Mô hình đề xuất được huấn luyện và đánh giá trên bộ dữ liệu MRI gồm 3.929 ảnh. Kết quả thực nghiệm cho thấy phương pháp đề xuất đạt quả phân đoạn tốt hơn so với một số thuật toán khác bao gồm U-Net truyền thống, Fuzzy C-Means (FCM), Attention U-Net, UNet++, SegNet, Swin-UNet, TransUNet và nnUNet, với BF = 98,0%, SSIM = 99,2%, ACC = 99,7%, Dice coefficient = 86,6% và IoU = 76,6%. Ngoài ra, kết quả nghiên cứu cũng đã giúp phân loại sự hiện diện của khối u dựa trên nhãn phân đoạn đạt độ chính xác 99,7% trên tập kiểm tra. Những kết quả này khẳng định hiệu quả của mô hình đề xuất và các ứng dụng trong phân tích ảnh y tế.

Từ khóa: Ảnh MRI, chẩn đoán, phân đoạn, Res-UNet, u não

Article Details

Tài liệu tham khảo

Abdalla, H. E. M., & Esmail, M. (2018). Brain tumor detection by using artificial neural network. In Yousif, A. F., Ahmed, S. Y. M., & Elhag, N. A. A., 2018 International Conference on Computer, Control, Electrical, and Electronics Engineering (pp. 1-6). IEEE. https://doi.org/10.1109/ICCCEEE.2018.8515763

Akkus, Z., Galimzianova, A., Hoogi, A., Rubin, D. L., & Erickson, B. J. (2017). Deep learning for brain MRI segmentation: state of the art and future directions. Journal of Digital Imaging, 30(4), 449-459. https://doi.org/10.1007/s10278-017-9983-4

Amin, J., Sharif, M., Yasmin, M., & Fernandes, S. L. (2018). Big data analysis for brain tumor detection: Deep convolutional neural networks. Future Generation Computer Systems, 87, 290-297. https://doi.org/10.1016/j.future.2018.04.065

Buda, M., Saha, A., & Mazurowski, M. A. (2019). Association of genomic subtypes of lower-grade gliomas with shape features automatically extracted by a deep learning algorithm, Computers in Biology and Medicine, 109, 218-225.
https://doi.org/10.1016/j.compbiomed.2019.05.002

Datta, S., Barua, R., & Das, J. (2020). Application of artificial intelligence in modern healthcare. In Leonel Pereira, Alginates: Recent uses of this natural polymer (pp. 121-134). IntechOpen. https://doi.org/10.5772/intechopen.90454

Farmanfarma, K. K., Mohammadian, M., Shahabinia, Z., Hassanipour, S., & Salehiniya, H. (2019). Brain cancer in the world: an epidemiological review. World Cancer Research Journal, 6(5), 1-5.

Khalighi, S., Reddy, K., Midya, A., Pandav, K. B., Madabhushi, A., & Abedalthagafi, M. (2024). Artificial intelligence in neuro-oncology: advances and challenges in brain tumor diagnosis, prognosis, and precision treatment. NPJ Precision Oncology, 8(1), 80. https://doi.org/10.1038/s41698-024-00575-0

Humphreys, S. C., Block, J. E., Sivaganesan, A., Nel, L. J., Peterman, M., & Hodges, S. D. (2025). Optimizing the clinical adoption of total joint replacement of the lumbar spine through imaging, robotics and artificial intelligence. Expert Review of Medical Devices, 22(5), 405-413.
https://doi.org/10.1080/17434440.2025.2484252

McKinney, P. A. (2004). Brain tumours: incidence, survival, and aetiology. Journal of Neurology, Neurosurgery & Psychiatry, 75(2), 12-17.
https://doi.org/10.1136/jnnp.2004.040741

National Cancer Institute. (n.d.). The Cancer Genome Atlas Program (TCGA).
https://www.cancer.gov/ccg/research/genome-sequencing/tcga

Tamilselvan, K. S., Murugesan, G., & Gnanasekaran, B. (2013). Brain tumor detection from clinical CT and MRI images using WT-FCM algorithm. In P. Rangarajan, 2013 International Conference on Green Computing, Communication and Conservation of Energy (pp. 260-263). IEEE. https://doi.org/10.1109/ICGCE.2013.6823441

Paschen, U. P. C., & Kietzmann, J. (2020). Artificial intelligence: Building blocks and an innovation typology. Business Horizons, 63(2), 147-155.
https://doi.org/10.1016/j.bushor.2019.10.004

Ruan, J., Li, J., & Xiang, S. (2025). VM-UNet: Vision Mamba UNet for medical image segmentation. ACM Transactions on Multimedia Computing, Communications, and Applications, 1-17. https://doi.org/10.1145/3767748

Saeedi, S., Rezayi, S., Keshavarz, H., & R. Niakan Kalhori, S. (2023). MRI-based brain tumor detection using convolutional deep learning methods and chosen machine learning techniques. BMC Medical Informatics and Decision Making, 23(1), 1-17. https://doi.org/10.1186/s12911-023-02114-6

Starmans, M. P. A., van der Voort, S.R., Castillo Tovar, J.M., Veenland, J.F., Klein, S., & Niessen, W.J. (2020). Radiomics: Data mining using quantitative medical image features. In Zhou, K. S., Rueckert, D., & Fichtinger, G., Handbook of Medical Image Computing and Computer Assisted Intervention (pp. 429-456). Academic Press.
https://doi.org/10.1016/B978-0-12-816176-0.00023-5

Velastegui, R., Tatarchenko, M., Karaoglu, S., & Gevers, T. (2024). Image semantic segmentation of indoor scenes: A survey. Computer Vision and Image Understanding, 248, 104102.
https://doi.org/10.1016/j.cviu.2024.104102

Vo, V. T., & Pham, T. D. (2021). Segmentation for the magnetic resonance images of Covid-19 patients using adaptive automatic kernel-based fuzzy clustering algorithm. In Dinh, S. K., Nguyen, X. H., Nursanty, E., Van, T. C., Truong, V. M., Ton, T. L., Rahaman, H., Narvios, W. M. O., Galindo, R., Pepito, J., Pham, T. K., Ngo, T.T.T., Hai, D., Ngo, T. H., Nguyen, H.. T., Phuoc, H., Do, T. H. D., Dang, Q. N., & Lam, T. M. D., 1st van lang international conference on heritage and technology conference proceeding (p. 040005). AIP Conference Proceedings. https://doi.org/10.1063/5.0066463

Wu, J., Ji, W., Fu, H., Xu, M., Jin, Y., & Xu, Y. (2024). Medsegdiff-v2: Diffusion-based medical image segmentation with transformer. In Wooldridge, M., Dy, J., & Natarajan, S., The Thirty-Eighth AAAI Conference on Artificial Intelligence (pp. 6030–6038). AAAI Press. https://doi.org/10.1609/aaai.v38i6.28418