An Accurate Diagnosis and Classification of Breast Mammogram Using Transfer Learning in Deep Convolutional Neural Network

Timely and precise identification of breast cancer (BC) is essential for enhancing patient results. This work investigates the capacity of deep learning to automate the categorization of BC using mammogram images. Our proposal involves using transfer learning with a customized Inception v3 architecture. This method utilizes pre-trained characteristics to tailor the model to the particular mammogram domain. Next, modified the last layers of the network and refined it using a mammogram dataset for classification. We assess the efficacy ofAour adapted Inception v3 model on a standard mammogram dataset, contrasting it with other deep learning structures and conventional machine learning techniques. The suggested technique demonstrates higher accuracy, sensitivity, and specificity in distinguishing normal and malignant breast tissue when compared to other approaches. This modified Inception v3 architecture has the capacity to strengthen the effectiveness in addition to the precision of breast cancer screening, eventually resulting in improved patient care. The proposed work is evaluated in comparison to other existing methodologies, including Inception v3, Inception v2, and VGG 16 and 19. AThe proposed modified Inception v3 model achieves 96.1% which is comparatively higher than the other methods.