Predicting Thyroid Dysfunction Using Machine Learning Techniques
Thirumala Akash K, F Mohammed Usman, T. Nitesh Kumar, Mohammed Riyaz Ahmed, Raveendra Gudodagi
Abstract
The history of the thyroid gland dates back to ancient times when it was recognized as a structure in the neck that was important for maintaining good health. Early detection of thyroid disorders is important because it can allow for earlier treatment and potentially improve outcomes. Artificial intelligence (AI) can improve early thyroid disorder detection accuracy and efficiency. Machine learning (ML) is a type of AI that involves training a computer to recognize patterns in data; it can be used to enhance the accuracy and efficiency of early thyroid disorder detection in several ways. Convolutional Neural Networks (CNNs) can improve early thyroid disorder detection accuracy and efficiency by providing a more objective and comprehensive image data analysis. Therefore, CNNs are primarily used for image analysis, and other types of neural networks may be more effective for analyzing other data types. Feature selection is used to identify the most relevant or important features in a dataset for a particular task. In the context of early thyroid disorder detection, feature selection could be used to identify the most important symptoms, risk factors, or other characteristics that are associated with thyroid disorders. This work exploits the various machine-learning approaches for thyroid prediction using algorithms such as Logistic regression (LR), Support Vector Machine (SVM), Decision tree (DT), etc. As a result, the best algorithm has been chosen based on accuracy.