Identification of Breast Cancer Using an Ensemble of Deep Learning Techniques

Breast cancer is the second most common cancer in women and the fifth most common cancer overall; case finding is important for enhancing survival. In this research, the use of multiple DL models using an ensemble learning technique to detect BC is carried out to determine the effectiveness of deep learning techniques. Due to the integration of four different models, namely VGG-16, is-MobileNetV2, DenseNet-169, as well as InceptionV4, the classification accuracy is enhanced and the false-positive rates are minimized. For each model, the accuracy in classifying mammography pictures into benign, malignant, or normal images is evaluated. The ensemble model, which combined all four DL approaches, outperformed these models with 99.13% accuracy. Precision, recall, and F1-score showed that this model was superior in BC detection. Advanced DL methods and ensemble framework optimization could boost performance. Early BC identification and treatment will require integration into real-world diagnostic systems and model refinement for several imaging modalities.