mohamad, A., farouk, R., Abd Elaziz, M. (2025). Examine Breast Cancer images using convolutional networks models for analysis. Bulletin of Faculty of Science, Zagazig University, 2025(2), 50-60. doi: 10.21608/bfszu.2024.307330.1417
abdelmounem ali mohamad; R M farouk; Mohamed Abd Elaziz. "Examine Breast Cancer images using convolutional networks models for analysis". Bulletin of Faculty of Science, Zagazig University, 2025, 2, 2025, 50-60. doi: 10.21608/bfszu.2024.307330.1417
mohamad, A., farouk, R., Abd Elaziz, M. (2025). 'Examine Breast Cancer images using convolutional networks models for analysis', Bulletin of Faculty of Science, Zagazig University, 2025(2), pp. 50-60. doi: 10.21608/bfszu.2024.307330.1417
mohamad, A., farouk, R., Abd Elaziz, M. Examine Breast Cancer images using convolutional networks models for analysis. Bulletin of Faculty of Science, Zagazig University, 2025; 2025(2): 50-60. doi: 10.21608/bfszu.2024.307330.1417
Examine Breast Cancer images using convolutional networks models for analysis
11Mathematics Department, Faculty of Science, Zagazig University, sharqia, Egypt
2Department of Mathematics, Faculty of science, Zagazig university, Egypt
3Faculty of Computer Science and Engineering, Galala University, Suez 435611, Egypt Department of Mathematics, Faculty of Science, Zagazig University, Zagazig 44519, Egypt
Abstract
Examination of Breast Cancer (BC) is critical for patient outcomes, and bio-imaging methods such as Ultrasound Breast Image scans are essential for clinical assessment. However, manually analyzing these images is time-consuming and requires expertise. To tackle this issue, we suggest the use of an image classification model. This paper investigates the use of deep learning models to classify mammographic images for breast cancer. The AlexNet and VGG 19 models were tested using the ultrasound dataset, with the goal of accurately categorizing images into benign and malignant classes. The results showed high performance for all models. This suggests that deep learning models, especially advanced ones like the proposed updated model, are effective in identifying mammographic images and could improve breast cancer accuracy. Our proposed method relies on using a Convolutional Neural Network (CNN) model for feature extraction. After extracting the features, we perform fine-tuning to enhance the model's performance. Finally, the extracted features are passed to a Support Vector Machine (SVM) for classification or examination tasks. Our method not only decreases the workload and analysis time for doctors but also improves the accuracy of tumor detection. Automating image analysis and identifying tumors early and accurately can result in improved patient care. Further research is required to improve these models and investigate their application in clinical environments.
View on SCiNiTO
Top