Analysis of the effectiveness of using U-net architecture for classification and segmentation of glioma in MRI images

The purpose of the research  is to analyze the efficiency of the U-net neural network architecture in decision support systems for glioma diagnostics and segmentation of brain areas affected by it on MRI images.

Methods. To conduct experimental studies, a training dataset was generated and the data was normalized. A software implementation of the U-Net neural network architecture was performed using the Keras framework in the Python programming language. The neural network model was trained.

Results. A series of experiments were conducted, during which error and classification matrices were obtained, the efficiency of classification of the trained neural network model for the "Tumor" and "No tumor" classes was assessed using metrics such as Recall, Precision and F1-measure, and the quality of segmentation of glioma-affected areas on the test data set was assessed. The quality of segmentation was assessed using the IoU metric, which reflects the ratio of the areas of the bounding boxes and is used to assess the accuracy of the spatial correspondence of the predicted segmented areas highlighted on the masks. Based on the results of testing the neural network model in solving the problem of segmenting brain areas affected by glioma, the average value of the IoU metric was 0.812, which is an acceptable result.

Conclusion. The testing results showed that the neural network model based on the U-net architecture is able to effectively diagnose the presence of glioma with acceptable values of the classification and segmentation quality metrics, which indicates the possibility of using this neural network model in medical decision support systems for glioma diagnostics, as well as its segmentation on MRI images. However, it is advisable to refine this neural network model to reduce the number of false negative classification results, which is critically important in medical diagnostics.

1. Pellerino A., et al. Epidemiology, risk factors, and prognostic factors of gliomas. Clinical and Translational Imaging. 2022;10:467–475. https://doi.org/10.1007/s40336-02200489-6

2. Wang L., et al. Malignant Glioma. Advances in experimental medicine and biology. 2023;1405:1–30. https://doi.org/10.1007/978-3-031-23705-8_1

3. Patil P., Giridhar P. Epidemiology and Demography of Brain Tumors. In: Evidence based practice in Neuro-oncology. Springer; 2021. P. 3–7. https://doi.org/10.1007/978-98116-2659-3_1

4. Bennett J., et al. Pediatric-type low-grade gliomas in adolescents and young adultschallenges and emerging paradigms. Child’s Nervous System. 2024. P. 1–11. https://doi.org/10.1007/s00381-024-06449-x

5. Gutta S., Acharya J., Shiroishi M.S., Hwang D., Nayak K.S. Improved Glioma Grading Using Deep Convolutional Neural Networks. AJNR. Am. J. Neuroradiol. 20214;(42):233–239. https://doi.org/10.3174/ajnr.A6882

6. Shevchenko T.A. Modern problems of treatment of gliomas of the brain of a high degree of malignancy. Vestnik Rossiiskogo nauchnogo tsentra rentgenoradiologii Minzdrava Rossii = Bulletin of the Russian Scientific Center for X-ray Radiology of the Ministry of Health of the Russian Federation. 2021;(3):10–12. (In Russ.)

7. Price M., et. al. CBTRUS Statistical Report: American Brain Tumor Association & NCI Neuro-Oncology Branch Adolescent and Young Adult Primary Brain and Other Central Nerv-ous System Tumors Diagnosed in the United States in 2016–2020. Neuro-Oncology. 2024:(26):1–53. https://doi.org/10.1093/neuonc/noae047

8. Jin Y., Peng H., Peng J. Brain glioma localization diagnosis based on MRI. World Neurosurgery. 2021;(149):325–332. https://doi.org/10.1016/j.wneu.2020.09.113

9. Butusov A.V., Kiselev A.V., Petrunina E.V., et al. Algorithms for monitoring the effectiveness of therapeutic and rehabilitation procedures based on indicators of clinical blood analysis in the medical decision support system. Izvestiya Yugo-Zapadnogo gosudarstvennogo universiteta. Seriya: Upravlenie, vychislitel’naya tekhnika, informatika. Meditsinskoe priborostroenie = Proceedings of the Southwest State University. Series: Control, Computer Engineering, Information Science. Medical Instruments Engineering. 2023;13(1):170–190. (In Russ.) https://doi.org/10.21869/2223-1536-2023-13-1-170-190

10. net… 10. Tomakova R.A., Dzyubin I.A., Brezhnev A.V. Method and algorithm of training a convolutional neural network designed for an intelligent system of melanoma cognition. Izvestiya Yugo-Zapadnogo gosudarstvennogo universiteta. Seriya: Upravlenie, vychislitel’naya tekhnika, informatika. Meditsinskoe priborostroenie = Proceedings of the Southwest State University. Series: Control, Computer Engineering, Information Science. Medical Instruments Engineering. 2022;12(1):65–83 (In Russ.). https://doi.org/10.21869/2223-15362022-12-1-65-83

11. Zhuge Y., et. al. Automated glioma grading on conventional MRI images using deep convolu-tional neural networks. Medical Physics. 2020; (47):3044–3053. https://doi.org/10.1002/mp.14168

12. Anaraki A.K., Ayati M., Kazemi F. Magnetic resonance imaging-based brain tumor grades classification and grading via convolutional neural networks and genetic algorithms // Biocybernetics and Biomedical Engineering. 2019;39:63–74.

13. Lo C. M., et al. Intelligent Glioma Grading Based on Deep Transfer Learning of MRI Radiomic Features. Applied Sciences. 2019; (9):4926. https://doi.org/10.3390/app9224926

14. Azad R., et al. Medical Image Segmentation Review: The Success of U-Net. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2024. P. 1–20. https://doi.org/10.1109/TPAMI.2024.3435571

15. Pan J. Image Segmentation Based On U-Net and Adjusted U-Nets. Highlights in Science, Engineering and Technology. 2024;85:316–327. https://doi.org/10.54097/accm2w81

16. Bangul K., Hajira F., Ayatullah Q. Drawbacks of Artificial Intelligence and Their Potential Solutions in the Healthcare Sector. Biomed Mater & Devices. 2023;1:731–738. 10.1007/s44174-023-00063-2

17. Azad R., et al. TransDeepLab: Convolution-Free Transformer-Based DeepLab v3+ for Medical Image Segmentation. In: Predictive Intelligence in Medicine: 5th International Workshop, PRIME 2022, Held in Conjunction with MICCAI 2022. Singapore: Springer, 2022. P. 91–102. https://doi.org/10.1007/978-3-031-16919-9_9

18. Bona J., et al. The PRISM semantic cohort builder: a novel tool to search and access clinical data in TCIA imaging collections. Physics in Medicine & Biology. 2022; (68):014003. https://doi.org/10.1088/1361-6560/ac9d1d

19. Bedi P., et al. Increasing the Versatility of Leaky ReLU Using a Nonlinear Function. In: Advanced Machine Intelligence and Signal Processing. Springer; 2022. P. 433–442. https://doi.org/10.1007/978-981-19-0840-8_32

20. Degala S., et. al. Segmentation and Estimation of Fetal Biometric Parameters using an Attention Gate Double U-Net with Guided Decoder Architecture. Computers in Biology and Medicine. 2024;(180). https://doi.org/10.1016/j.compbiomed.2024.109000

21. Kiselev A.V., Brusentsev N.S., Kuleshova E.A. Analysis of the effectiveness of using two-stage neural network models for early detection of forest fires. Izvestiya Yugo-Zapadnogo gosudarstvennogo universiteta. Seriya: Upravlenie, vychislitel’naya tekhnika, informatika. Meditsinskoe priborostroenie = Proceedings of the Southwest State University. Series: Control, Computer Engineering, Information Science. Medical Instruments Engineering. 2024;14(1):8–23. (In Russ.) https://doi.org/10.21869/2223-1536-2024-14-1-8-23

22. Stodt J., Reich C., Clarke N. Unified Intersection Over Union for Explainable Artificial Intelligence. In: Intelligent Systems and Applications: Proceedings of the 2023 Intelligent Systems Conference (IntelliSys). Springer; 2024. P. 758–770. https://doi.org/10.1007/978-3-031-47724-9_50