Early Prediction and Classification of Breast Cancer Survival Based on Machine Learning Models
Himanshu Sinha, Milavkumar Shah · 2025
Breast cancer is the most prevalent illness in women globally and is characterised by the uncontrolled growth of cells in the breast tissues. Breast cancer continues to rank among the leading causes of mortality for women globally, highlighting the need for accurate and prompt patient survival prediction. This study manages this challenge by using CatBoost and XGBoost machine learning algorithms to predict and classify breast cancer survival. In the following study, it presents the latest development in early breast cancer diagnosis with help of distinct AI and ML approaches. Using the SEER Breast Cancer dataset, the study measures the efficacy of the predictive models including CatBoost, Extra Tree Classifier, LightGBM, and XGBoost. Performance is supported by experimental data, making efficient use of the cross-validation technique and comparing the results with conventional diagnostic techniques. It shows the effectiveness of these models with CatBoost validated accuracy of 96.7% and XGBoost validated accuracy of 96.2% are better than traditional methods. Some other measures like recall, precision and F1-score also provide evidence for the above findings regarding the reliability of these models. The work highlights the applicability of these technologies in clinical diagnosis and provides future research suggestions for using and optimising these technologies in breast cancer diagnosis based on multimodal data fusion, explainability, and large-scale clinical trials.