Secure and Collaborative Breast Cancer Detection Using Federated Learning

To attain the urgent need for diagnosis of breast cancer as expected in early periods, this research reviews the application of a federated learning model for breast cancer detection. The model consolidates disconnected data from numerous sources using a rich set of mammography pictures without compromising the patient's confidentiality. Manufactured with an accuracy rating of 95. 3% and this is how the federated learning strategy surpassed the conventional centralized ideal models by 3%. 7%. Some prominent parameters prove that there are substantial improvements in terms of improvement of detection and, therefore, this approach can be considered as a possible instrument in practical applications of medicine. The capabilities of using the federated learning model that referred to heterogeneous data while preserving privacy in the prospective of overturning breast cancer detection.