A Detailed Investigation and Analysis of Using Machine Learning Techniques for Thyroid Diagnosis

Abstract

A Method of Classification Based on Norms Data mining greatly benefits several subfields within the healthcare industry. Detecting and treating diseases at an early stage is a challenging but essential objective in the healthcare field. If they are discovered early enough, many diseases can be diagnosed and treated while they are still in their early stages. Conditions that affect the thyroid are one example of this type of example. In the past, thyroid disorders were identified through a process that involved observing a patient's symptoms and doing a battery of blood tests. The primary goal is to enhance the accuracy with which diseases are detected in the initial stages of their progression. The healthcare business may gain a significant amount from using data mining techniques for decision-making, disease diagnosis, and the provision of superior treatment to patients at reduced prices. Thyroiditis is ongoing. The act of classifying things into different groups is significant. This study aims to determine the connection between TSH, T3, and T4 and hyperthyroidism and hypothyroidism. It also tries to determine the relationship between TSH, T3, T4, and gender. Additionally, the research will attempt to predict thyroid disease using several classification systems. Our study shows that the Neural network classifier generates the highest classification accuracy of 98.4%.