Towards Early Breast Cancer Detection: A Deep Learning Approach
Amina Bekkouche, Mohammed Amine Merzoug, Mourad Hadjila, Wafaa Ferhi · Engineering Technology & Applied Science Research · 2024
Early detection of breast cancer is crucial for patients' recovery chances to be improved. Artificial intelligence techniques, and more particularly Deep Learning (DL), may contribute to enhancing the accuracy of this detection. The main objective of this paper is to propose a DL model in an attempt to detect and classify breast cancer, and thus help people suffering from this disease. The Breast Cancer Wisconsin dataset was implemented to train neural networks, and their performance was subsequently evaluated on certain test datasets. The findings revealed that this approach provides promising results in terms of detection accuracy, with high sensitivity and specificity. The study also compares the performance of this approach with other breast cancer detection works, demonstrating that DL can provide significantly better results.