Harnessing Inception V3 for Efficient Classification of Breast Cancer in Mammography Images
Ritu Rani, Sheifali Gupta, Jayapal Lande · 2025
Given that breast cancer still ranks highest among female causes of death, early identification becomes even more important. This work investigates the use of the Inception V3 deep learning model for mammography imaging automated breast cancer detection. The model is adjusted to improve its classification ability using a dataset of 3,383 annotated mammograms split into normal and malignant cases. With an incredible accuracy of 89.44%, the Inception V3 model beats conventional models such VGG16, which only attained a meager 53% accuracy. Among the crucial tests showing how successfully the model distinguishes benign from malignant tumours are precision, recall, and F1 score. The outcomes show how effectively deep learning approaches could be able to increase the diagnostic accuracy and efficiency of clinical settings. This work promotes future validation on larger datasets and early breast cancer detection initiatives to raise the therapeutic usefulness of the model.