Histopathological Image Analysis for Breast Cancer Identification Using Gradient Boosting Classifier

When it comes to cancers, breast cancer is by far the most prevalent among females and the second largest cause of cancer death globally. Radiologists frequently make incorrect diagnoses of breast cancer due to the subtlety of Microcalcification and Lumps. Past research has led to the creation of CAD (Computer Aided Diagnosis) tools that enable the radiologist in quickly and accurately identifying problems. This research study proposes a new method for predicting breast cancer, the Gradient Boosting Classifier (GBC), a supervised machine learning algorithm for identifying breast cancer by training its features. Breast Cancer Kaggle dataset repository provides the histopathological images utilized in this study. The proposed system's performance is rigorously assessed, considering accuracy, sensitivity, and specificity. Notably, it consistently delivers outstanding results during both training and testing phases. In the validation phase, the methods excel, demonstrating an impressive accuracy rate of 96.15%, a sensitivity level of 94%, an F1 score of 95%, and an impressive area under the curve (AUC) of 0.96.