Beyond the Biopsy: A Comprehensive Machine Learning Based Approach to Thyroid Cancer Staging
Shivam Kumar Bharti, Debolina Ghosh, Mahendra Kumar Gourisaria, Junali Jasmine Jena, Parthasarathi Pattnayak, Sudhansu Shekhar Patra
Abstract
Thyroid cancer is characterized by pernicious growth in thyroid gland which is a vital endocrine gland crucial for metabolism and health rhythm regulation. Pathological alterations result in uncontrolled cell multiplication and the formation of tumors. Precise staging is crucial for optimizing the treatment procedure and mitigating the mortality risk. This research study analyzes the application of machine learning (ML) models to predict cancer stages using clinicopathologic data. Various models like CatBoost, Logistic regression, XG-Boost, SGD, KNN, GaussianNB, Random Forest, Decision tree, Gradient Boosting and AdaBoost were deployed in this work. Notably, XG-Boost, CatBoost and Decision tree achieved outstanding results with an accuracy of 0.9452. This research contributes valuable insights to the field of healthcare by highlighting ML role in revolutionizing thyroid cancer staging, which could lead to more precise therapy. With reduced risk, it holds the possibility of improved results, more precise diagnosis, and customized treatment plan.