This research aims to develop an optimized deep-learning model capable of detecting breast cancer from medical images, which could be mammograms or histopathological slides. Breast cancer is one of the leading causes of cancer deaths in the world, making breast cancer detection extremely important for enhancing survival rates, when detected early. The traditional breast cancer detection process relies on a medical professional putting their eyes on a medical image, which is typically an inefficient process and disposed to human error. As deep learning and machine learning become more ubiquitous, particularly Convolutional Neural Networks (CNNs), they have opened ways for automation and improved accuracy in breast cancer detection. This project will use the Inception V3 model, which is an established CNN architecture, to develop a reliable breast cancer detection system that classifies images of breast images as benign or malignant. Karri Swathi Dept. of Computer Science and Engineering Institute of Aeronautical Engineering Dundigal, Hyderabad, India [email protected] Godishala Sreenidhi Dept. of Computer Science and Engineering Institute of Aeronautical Engineering Dundigal, Hyderabad, India [email protected] likelihood of positive patient outcomes in response to early diagnosis. Further machine or deep learning implementation appears to be a favorable alternative to traditional and time-antique diagnostic and medical behavioral methods. Opportunities for further project improvements can continue evolving, thus incorporating deploying the target theology into increased images per class, utilizing ensemble methods, or deploying into clinical behavioral context and evidentiary articulation after image literature review Keywords— Breast cancer classification, Inception v3 Convolutional Neural Network (CNN). I. INTRODUCTION The theme of this project is to cultivate a deep learning model using the Inception V3 architecture to reliably detect breast cancer from medical images. This involved objectives such as optimizing the input image pre-processing, training the Inception V3 model using a labeled breast cancer image dataset, and then evaluating the performance of the model using standards from accuracy, precision, recall, F1 Score, and AUC, comparison of model results with existing methods