Machine Learning and Artificial Intelligence in Thyroid Cancer Screening and Diagnosis: A Comprehensive Systematic Review
Rushin Patel *
Department of Internal Medicine, Community Hospital of San Bernardino, CA, USA.
Akash Jain
Department of Internal Medicine, Ascension Via Christi Hospital, KS, USA.
Zalak Patel
Department of Internal Medicine, University of California Riverside, CA, USA.
Chieh Yang
Department of Internal Medicine, University of California Riverside, CA, USA.
Darshil Patel
Clinical Research Program, Rush University, IL, USA.
Mrunal Patel
Department of Internal Medicine, Trumbull Regional Medical Center, OH, USA.
*Author to whom correspondence should be addressed.
Abstract
This systematic review explores the role of artificial intelligence (AI) and machine learning (ML) technologies in the diagnosis and treatment of thyroid cancers (TC), focusing on enhancing precision, risk assessment, and tailored care. By analyzing ten studies, the review highlights how AI and ML technologies, such as deep learning (DL) and computer-aided diagnostics (CAD), improve the accuracy of ultrasound imaging, risk stratification, and the detection of high-risk nodules. Despite advancements, challenges persist in transitioning to personalized care, including uneven prognostication and diagnostic uncertainty. The review evaluates the effectiveness of AI and ML compared to conventional methods, their ability to address diverse tumor characteristics, and their strengths and limitations in prognosis prediction. Findings suggest AI's potential in improving precision and risk assessment, but limitations such as inconsistent approaches and biases highlight the need for larger datasets and standardized procedures. Moreover, the review underscores the importance of interpretability and transparency in AI models and calls for further research to validate findings in clinical settings. Despite limitations and challenges, AI's transformative potential in TC management is evident, underscoring the need for ongoing investigation and integration into clinical practice.
Keywords: Thyroid cancer, artificial intelligence, machine learning, thyroid nodule, diagnosis, AI, ML