Model Generation and Prediction of Breast Cancer Malignancy Using Machine Learning Algorithms

Healthcare providers continue to face challenges in identifying breast cancer malignancy, despite using mammography and magnetic resonance imaging, which have limitations. As a result, there is a growing interest in machine learning (ML) for its precision in diagnosis and outcome prediction. This study utilized various ML algorithms to create models for diagnosing breast cancer malignancy, using data from the Wisconsin Diagnostic Breast Cancer database (WDBC). Logistic regression and support vector machines (SVM) models were employed to predict breast cancer malignancy. Logistic regression identified four key parameters: bland chromatin, bare nuclei, marginal adhesion, and clump thickness. It should be mentioned that SVM had higher accuracy and area under the ROC curve (۰.۹۹). Both of ML models effectively predicted breast cancer malignancy based on these attributes, making them valuable tools in clinical settings for predicting breast cancer malignancy.

کلیدواژه ها:

Breast cancer ، Machine Learning ، Malignancy ، Logistic Regression ، Support vector machine