Hybrid Guided Filter And Deep Learning Based Brain Tumour Segmentation in MRI Images

The proposed project presents an effective and efficient approach for the detection and analysis of brain tumors from MRI images using image processing techniques. The system successfully performs multiple stages including preprocessing, segmentation, tumor detection, and performance evaluation. The preprocessing step enhances the quality of MRI images by reducing noise and improving contrast, which helps in accurate tumor identification. The segmentation process plays a key role in isolating the tumor region from the normal brain tissues. By applying thresholding and morphological operations, the system is able to clearly distinguish tumor areas. The detected tumor region is further visualized using overlay techniques and bounding boxes, making it easier to understand the tumor location and extent. In addition to detection, the system calculates the tumor area percentage, which helps in classifying the tumor into different categories such as small, moderate, or large. This provides valuable information for medical diagnosis and decision-making. The system also evaluates its performance using metrics such as Accuracy, Precision, Recall, F1 Score, Intersection over Union (IoU), and Dice Coefficient. The obtained results demonstrate high accuracy and strong overlap between predicted tumor regions and ground truth masks, indicating the reliability of the proposed method. Overall, the developed system provides a simple, fast, and cost-effective solution for brain tumor detection. It reduces manual effort and assists medical professionals in diagnosing brain tumors more efficiently. This project serves as a strong foundation for further improvements using advanced techniques.

[1] N. Sharma and L. M. Aggarwal, “Automated medical image segmentation techniques,” J. Med. Phys., vol. 35, no. 1, pp. 3–14, Jan. 2010. [2] D. Thomson et al., “Evaluation of an automatic segmentation algorithm for definition of head and neck organs at risk,” Radiat. Oncol., vol. 9, no. 1, pp. 1–12, Dec. 2014. [3] E. Ermis et al., “Fully automated brain resection cavity delineation for radiation target volume definition in glioblastoma patients using deep learning,” Radiat. Oncol., vol. 15, no. 1, pp. 1–10, Dec. 2020. [4] O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional networks for biomedical image segmentation,” in Proc. Int. Conf. Med. Image Comput. Comput.-Assist. Intervent. Cham, Switzerland: Springer, 2015, pp. 234–241. [5] Ö. Çiçek, A. Abdulkadir, S. S. Lienkamp, T. Brox, and O. Ronneberger, “3D U-Net: Learning dense volumetric segmentation from sparse annotation,” in Proc. Int. Conf. Med. Image Comput. Comput.-Assist. Intervent. Cham, Switzerland: Springer, 2016, pp. 424–432. [6] G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, “Densely connected convolutional networks,” in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), Jul. 2017, pp. 4700–4708. 36 [7] K. He, X. Zhang, S. Ren, and J. Sun, “Identity mappings in deep residual networks,” in Proc. Eur. Conf. Comput. Vis. Cham, Switzerland: Springer, 2016, pp. 630–645. [8] Y.LeCun,L.Bottou,Y.Bengio,andP.Haffner,“Gradient-basedlearning applied to document recognition,” Proc. IEEE, vol. 86, no. 11, pp. 2278–2324, Nov. 1998. [9]  A. Krizhevsky, I. Sutskever, and G. E. Hinton, “ImageNet classification with deep convolutional neural networks,” in Proc. Adv. Neural Inf. Process. Syst., vol. 25, 2012, pp. 84–90. [10] K. Sun, B. Xiao, D. Liu, and J. Wang, “Deep high-resolution representation learning for human pose estimation,” in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), Jun. 2019, pp. 5693–5703.