Breast Cancer Survival Rate Prediction Using Machine Learning Algorithm

Abstract

Breast cancer is a formidable foe impacting numerous lives, emphasizing the critical need for accurate survival predictions to guide personalized treatment decisions. This research employed a machine learning algorithms, specifically decision tree models, for breast cancer survival prediction. The study seeks a model that deftly handles the intricacies of missing data and resonates meaningfully with the healthcare community. The study used a decision tree model, envisioning a precision virtuoso adeptly hitting the right notes, achieving a score of 0.73 for those overcoming breast cancer, a complex disease, and an admirable 0.87 for those navigating a more arduous journey. With a recall score of 0.90 for survivors, assuring that the majority are acknowledged. Simultaneously, with a score of 0.68 for those facing a more challenging path, it reveals a nuanced understanding of individuals who may not traverse it successfully. Delving into the nuanced realm of these metrics, the F1 score emerges as a meticulously choreographed dance, resonating at 0.81 for survivors and 0.76 for those facing formidable odds. This narrative isn't confined to numerical precision; it's a symphony of a model predicting with nuance. Stepping back, the overall accuracy of 0.79 was not merely a numerical outcome but the model's outstanding performance on the healthcare stage. Transcending precision, crafting a tool that converses in the language of healthcare professionals, facilitating nuanced decision-making. This research journey extended beyond numerical precision; it's an exploration into unraveling complexities and crafting a tool that transcends sterile lab origins. Envision this decision tree model not as an isolated entity but as a collaborative partner in diverse clinical environments. It's an evolving creation, refined based on real-world feedback, a symphony in harmony with the experiences of healthcare practitioners navigating the delicate terrain of breast cancer treatment. The model not only interprets data but comprehends the human stories behind it and aspires not merely to numerical accuracy but to a profound, positive impact on patient outcomes in the complex, real-world arena.