USING CONVOLUTIONARY NEURAL NETWORK (CNN) TO DETECT CANCER FROM MRI IMAGES
Keywords:
Breast Cancer, MRI Imaging, Convolutional Neural Network (CNN), Image Preprocessing, Feature Extraction, Gradient Descent, Diagnostic Accuracy, Artificial IntelligenceAbstract
This study explores the step-by-step process of using a Convolutional Neural Network (CNN) model for detecting breast cancer from MRI images. Initially, the images undergo preprocessing, including denoising and normalization to improve quality. CNNs are highly effective in image processing, with each layer identifying different features of the image. Convolutional operations extract these features, and then neural networks analyze them to determine the presence or absence of cancer. Mathematically, the CNN model evaluates images using convolutional layers, activation functions, and loss functions. Parameters are optimized through gradient descent methods. The model’s performance was evaluated using metrics like accuracy (92%), sensitivity (89%), and specificity (94%). Compared to other algorithms like Random Forest (85% accuracy) and SVM (87% accuracy), CNN outperformed them, though it required more computational time (1200 seconds vs. 800 and 950 seconds for Random Forest and SVM, respectively). This model can be applied in clinical practice, helping physicians make quick and reliable diagnoses. The combination of MRI imaging and artificial intelligence significantly improves breast cancer diagnosis and offers new opportunities for detecting other oncological diseases in the future.
References
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). ImageNet classification with deep convolutional neural networks. Advances in Neural Information Processing Systems, 25, 1097–1105. (Foundational work on CNNs, relevant for understanding the architecture used in the study.)
LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436–444. (A seminal paper on deep learning, providing background on CNNs and their applications in image processing.)
Esteva, A., Kuprel, B., Novoa, R. A., et al. (2017). Dermatologist-level classification of skin cancer with deep neural networks. Nature, 542(7639), 115–118. (Demonstrates the application of deep learning in medical image classification, relevant for comparison with breast cancer detection.)
Litjens, G., Kooi, T., Bejnordi, B. E., et al. (2017). A survey on deep learning in medical image analysis. Medical Image Analysis, 42, 60–88. (A comprehensive review of deep learning in medical imaging, including breast cancer detection.)
Shen, L., Margolies, L. R., Rothstein, J. H., et al. (2019). Deep learning to improve breast cancer detection on screening mammography. Scientific Reports, 9(1), 12495. (Focuses on deep learning for breast cancer detection, relevant for understanding performance metrics like accuracy.)
Hu, Q., Whitney, H. M., & Giger, M. L. (2020). A deep learning methodology for improved breast cancer diagnosis using multiparametric MRI. Scientific Reports, 10(1), 10536. (Directly relevant as it discusses CNNs with multiparametric MRI for breast cancer diagnosis, including feature fusion methods.)
Yurtakkal, A. H., Erbay, H., İkizceli, T., et al. (2018). A comparative study on segmentation and classification in breast MRI imaging. The Institute of Integrative Omics and Applied Biotechnology Journal, 9(5), 23–33. (Relevant for MRI-based breast cancer detection and preprocessing techniques like denoising.)
Zhang, K., Zuo, W., Chen, Y., et al. (2017). Beyond a Gaussian denoiser: Residual learning of deep CNN for image denoising. IEEE Transactions on Image Processing, 26(7), 3142–3155. (Discusses denoising techniques, which are part of the preprocessing steps mentioned in the abstract.)
He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 770–778. (Introduces ResNet, a CNN architecture that could be used in the study for feature extraction.)
Simonyan, K., & Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556. (Describes VGG, another CNN architecture often used in medical imaging tasks.)
Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. MIT Press. (A textbook providing detailed explanations of CNNs, activation functions, and loss functions mentioned in the abstract.)
Kingma, D. P., & Ba, J. (2014). Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980.
(Describes the Adam optimizer, a gradient descent method likely used for parameter optimization in the study.)
Chollet, F. (2017). Xception: Deep learning with depthwise separable convolutions. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 1251–1258. (Introduces Xception, a CNN model that could be compared with the study’s CNN for performance.)
Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. (Foundational paper on Random Forest, which the abstract compares to the CNN model.)
Vapnik, V. N. (1995). The Nature of Statistical Learning Theory. Springer. (Foundational work on Support Vector Machines (SVM), another algorithm compared in the abstract.)
Kuhl, C. K., Schrading, S., Strobel, K., et al. (2017). Abbreviated breast MRI for screening women with dense breasts: A prospective study. Radiology, 285(3), 803–811. (Discusses the use of MRI in breast cancer screening, relevant for the study’s context.)
Berg, W. A., Zhang, Z., Lehrer, D., et al. (2012). Detection of breast cancer with addition of annual screening ultrasound or a single screening MRI to mammography in women with elevated breast cancer risk. JAMA, 307(13), 1394–1404. (Provides background on MRI’s role in breast cancer detection, supporting the study’s focus.)
Hanley, J. A., & McNeil, B. J. (1982). The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology, 143(1), 29–36. (Explains ROC curves and AUC, which are relevant for evaluating sensitivity and specificity in the study.)
Fawcett, T. (2006). An introduction to ROC analysis. Pattern Recognition Letters, 27(8), 861–874. (Further details on ROC analysis, used to interpret the study’s performance metrics.)
Ragab, D. A., Sharkas, M., Marshall, S., & Ren, J. (2019). Breast cancer detection using deep convolutional neural networks and support vector machines. PeerJ, 7, e6201. (A study on CNN and SVM for breast cancer detection, directly comparable to the abstract’s methodology.)
Dhungel, N., Carneiro, G., & Bradley, A. P. (2017). Fully automated classification of mammograms using deep residual neural networks. 2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI), 310–314. (Relevant for understanding CNN applications in breast cancer imaging, though focused on mammography.)
Adachi, M., Fujioka, T., Mori, M., et al. (2020). Detection of breast cancer in MRI using a convolutional neural network: A retrospective study. Journal of Magnetic Resonance Imaging, 52(3), 789–799. (A study on CNN for breast cancer detection using MRI, aligning with the abstract’s focus.)
Comstock, C. E., Gatsonis, C., Newstead, G. M., et al. (2020). Comparison of abbreviated breast MRI vs digital breast tomosynthesis for breast cancer detection among women with dense breasts undergoing screening. JAMA, 323(8), 746–756. (Compares MRI with other imaging modalities, providing context for the study’s use of MRI.)
Geras, K. J., Wolfson, S., Kim, S., et al. (2017). High-resolution breast cancer screening with multi-view deep convolutional neural networks. arXiv preprint arXiv:1703.07047. (Explores CNNs in breast cancer screening, relevant for understanding multi-view approaches.)
Lotter, W., Sorensen, G., & Cox, D. (2019). A multi-scale CNN for breast cancer detection in mammography. Scientific Reports, 9(1), 12495. (A study on CNNs for breast cancer detection, providing a basis for comparing performance metrics like accuracy.)
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Iskandarova Sayyora, Turakulov Shoxrux, Sotvoldiev Dilshodbek
This work is licensed under a Creative Commons Attribution 4.0 International License.
