Application of Artificial Intelligence to Evaluate Thyroid Nodules

Abstract

The incidence of thyroid cancer has increased steeply worldwide over the past few decades. The diagnosis of thyroid nodule depends on ultrasonography and fine needle aspiration & cytology, relying mainly on the experience and effort of clinicians and pathologists. Besides, these diagnostic tests are prone to subjective interpretation and interobserver variabilities. Artificial intelligence (AI) is a technology used to extract and quantify key image information by simulating complex human cognitive functions. It broadly includes machine & deep learning and represents a potent method that may facilitate and help clinicians to identify and classify human pathologies. While the outcomes of validation studies have been varied, there are abundant researches showing that AI can perform equally as or better than some humans, particularly less experienced doctors. In thyroid field, AI is mostly being adopted to ultrasonographic and cyto/histo-pathologic diagnosis of thyroid nodules. This review aims to provide an overview of the basic concepts of AI technologies and present the current status of application of AI to ultrasonographic and cyto/histo-pathological diagnosis of thyroid nodule.

Ähnliche Arbeiten