A Generic Model to Analyse and Predict Brain Tumour from MRI and CT Medical Images Using CNN

Abstract

The human brain is the primary controller of the human system. Abnormal growth and division of cells in the brain leads to a brain tumour, which if left untreated can progress into brain cancer. In the domain of human health, Computer Vision plays a significant role by reducing subjective human judgment and delivering accurate results. CT scans, X-Ray, and MRI scans are the most widely used imaging modalities, with Magnetic Resonance Imaging (MRI) being the most reliable and secure, capable of detecting minute structural abnormalities. This paper proposes a generic Convolutional Neural Network (CNN)-based model for automated detection and prediction of brain tumours from MRI and CT medical images. Pre-processing is performed using the Bilateral Filter (BF) for noise removal, followed by binary thresholding and CNN-based segmentation for reliable tumour region detection. Training, testing, and validation datasets are used. Performance is assessed through accuracy, sensitivity, and specificity metrics. The proposed model achieves 84% accuracy and yields promising results with minimal computational time.

Introduction

Brain tumours are life-threatening conditions that require early and accurate detection, typically using MRI or CT scans. However, manual diagnosis by radiologists is time-consuming, prone to human error, and inconsistent, especially when identifying small or early-stage tumours.

To address these issues, the paper proposes an automated brain tumour detection system using image processing and deep learning, specifically a CNN-based model. The system processes MRI images through a pipeline that includes preprocessing (noise removal and enhancement), segmentation (isolating tumour regions), feature extraction, and CNN-based classification to determine whether a tumour is present.

The study reviews earlier approaches such as clustering, edge detection, morphological methods, SVMs, and traditional neural networks, highlighting that while they improved detection over time, they still struggle with accuracy and robustness compared to deep learning methods.

The proposed system improves performance by combining image enhancement techniques (like bilateral filtering and Sobel edge detection) with CNN classification. It achieves better accuracy (84%) than traditional methods, along with higher sensitivity and specificity, making it more reliable for medical diagnosis.

Conclusion

This paper presented an automated CNN-based model for the detection and prediction of brain tumours from MRI and CT medical images. The proposed pipeline integrates Bilateral Filter preprocessing, binary thresholding, morphological segmentation, and CNN classification into a unified, end-to-end framework. Experimental results confirm that the proposed model achieves 84% accuracy with strong sensitivity and specificity values, outperforming traditional machine learning baselines including KNN, SVM, and standalone ANN. The system demonstrates that principled image preprocessing combined with deep learning classification produces clinically reliable diagnostic support, reducing the manual burden on radiologists and enabling earlier tumour detection.

References

[1] A. Sivaramakrishnan and M. Karnan, \"A Novel Based Approach for Extraction Of Brain Tumor In MRI Images Using Soft Computing Techniques,\" International Journal Of Advanced Research In Computer And Communication Engineering, vol. 2, no. 4, April 2013. [2] A. Aslam, E. Khan, and M. M. Sufyan Beg, \"Improved Edge Detection Algorithm for Brain Tumor Segmentation,\" Procedia Computer Science, vol. 58, pp. 430-437, 2015. [3] B. Sathya and R. Manavalan, \"Image Segmentation by Clustering Methods: Performance Analysis,\" International Journal of Computer Applications, vol. 29, no. 11, September 2011. [4] B. Devkota et al., \"Image Segmentation for Early Stage Brain Tumor Detection using Mathematical Morphological Reconstruction,\" Procedia Computer Science, vol. 125, pp. 115-123, 2018. [5] K. Sudharani, T. C. Sarma, and K. Satya Rasad, \"Intelligent Brain Tumor lesion classification and identification from MRI images using K-NN,\" ICCICCT, pp. 777-780, 2015. [6] J. Kaur, S. Agrawal, and V. Rig, \"A Comparative Analysis of Thresholding and Edge Detection Segmentation Techniques,\" International Journal of Computer Applications, vol. 39, pp. 29-34, 2012. [7] M. Kumar and K. K. Mehta, \"A Texture based Tumor detection and automatic Segmentation using Seeded Region Growing Method,\" International Journal of Computer Technology and Applications, vol. 2, no. 4, pp. 855-859, 2011. [8] D. Mahmoud and M. Eltaher, \"Brain Tumor Detection Using Artificial Neural Networks,\" Journal of Science and Technology, vol. 13, pp. 31-39, 2012. [9] Y. Pan et al., \"Brain tumor grading based on Neural Networks and Convolutional Neural Networks,\" IEEE Engineering in Medicine and Biology Society Annual Conference (EMBC), pp. 699-702, 2015. [10] S. Pereira et al., \"Brain Tumor Segmentation Using Convolutional Neural Networks in MRI Images,\" IEEE Transactions on Medical Imaging, vol. 35, no. 5, pp. 1240-1251, 2016. [11] A. Sankari and S. Vigneshwari, \"Automatic tumor segmentation using convolutional neural networks,\" ICONSTEM, pp. 268-272, 2017. [12] N. Varuna Shree and T. N. R. Kumar, \"Identification and classification of brain tumor MRI images with feature extraction using DWT and probabilistic neural network,\" Brain Informatics, vol. 5, pp. 23-30, 2018. [13] K. Vinotha, \"Brain Tumor Detection and Classification Using Histogram Equalization and Fuzzy Support Vector Machine,\" International Journal of Engineering and Computer Science, vol. 3, no. 5, pp. 5823-5827, 2014. [14] P. S. Mukambika and K. Uma Rani, \"Segmentation and Classification of MRI Brain Tumor,\" IRJET, vol. 4, no. 7, pp. 683-688, 2017. [15] Z. Shi et al., \"Survey on Neural Networks Used for Medical Image Processing,\" International Journal of Computational Science, vol. 3, no. 1, pp. 86-100, 2009.

Copyright

Copyright © 2026 Ms. Reena Pranavi Vejendla, Ms. Simma Thanmayee, Ms. Polisetti Venkata Guna Sree Mayi, Ms. Yenadava Haritha. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.