Thyroid Disease Prediction with Machine Learning Algorithms

Abstract

In recent years, machine learning techniques have been used to predict disease occurrences, limit disease spread, and find appropriate treatment methods in their early stages. Thyroid diseases are common diseases in this era, and women had the largest share in them, as the thyroid gland secretes hormones in the blood and affects the food metabolism in the body as well as the building and growth of the body in children. This study deals with detection of the hormonal activity of the thyroid gland, with its two types: hyperthyroidism (excess secretion of the hormone) and hypothyroidism (decreased secretion of the hormone) using various of machine learning algorithms as a classifier, such as Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), KNearest Neighbor (KNN), Support Vector Machine (SVM) and Naive Bays (NB) classifiers. All algorithms are simulated by a Python in Anaconda environment specific uses Spyder platform, comparing and choosing the most accurate. The DT and RF algorithm showed the best results through five experiments reach to 0.9933 and 0.9973, respectively.