An Intelligent Model for Improved Breast Cancer Prognosis

This research suggests developing a deep learning model using customized CNN to categorize and predict breast cancer in timely period. The utilizes large dataset of images obtained from Kaggle, an online repository. Pre-processing techniques were applied the eliminate noise, such as shadows on images, resize lessen high computation cost. was separated into training set 80% (48, 852) test 20% (16, 284). employed mine meaningful features classify them based predefined criteria, assessing presence severity cancer. Additionally, could provide treatment recommendations depending patient's health account other pertinent aspects. model's performance evaluated confusion matrix, revealing 95% accuracy rate, 100% recall value, 90% precision F1 score. classifier's AUC value 88%, indicating reliability for prognosis. proposed methodology may significantly increase diagnostic speed accuracy, resulting earlier detection better patient outcomes.