Intelligent Methods for Hypothyroidism Associations and Characterisation for Personalised Diagnosis and Treatment

Abstract

Hypothyroidism results from low levels of thyroid hormone produced and has varied aetiology and manifestations. It is potentially fatal in severe cases if it remains undetected and timely untreated. Accordingly, this research deploys intelligent machine learning-based methods to uncover its associations and facilitates accurate and efficient detection through case-based predictions. Specifically, data mining algorithms will be adopted to identify risk factors associated with hypothyroidism, including demographics and medical history. These identified associations can be applied to understand, screen, and predict hypothyroidism in individual patients. This intelligent approach enables proactive management of the disease in its early stages to mitigate the potential of developing subsequent diseases, such as heart disease, infertility, or thyroid tumours. The proposed intelligent methods not only advance our understanding of the disease pathogenesis but also achieve precise diagnosis, providing clinicians with valuable insights to support follow-up treatment decision-making.