### Abstract: Breakthrough in Thyroid Cancer AI Precision Diagnosis by Professor Xiaohai Peng's Team On March 22, 2021, a significant study by Professor Xiaohai Peng's team from the First Affiliated Hospital of Sun Yat-sen University was published in *The Lancet Digital Health*, a leading journal in the field of artificial intelligence (AI) in healthcare. The study, titled "Deep learning-based artificial intelligence model assists in thyroid nodule management: a multi-center, diagnostic study," presents a groundbreaking AI diagnostic model named ThyNet, which aims to improve the accuracy and efficiency of thyroid nodule management. **Key Events and Findings:** - **Prevalence and Challenges:** Thyroid nodules are detected in 40-66% of adults, and thyroid ultrasound is the primary non-invasive method for distinguishing between benign and malignant nodules. However, this method has a misdiagnosis rate of 15-20%, posing a significant challenge in clinical practice. - **Development of ThyNet:** Professor Peng's team utilized deep learning techniques to develop ThyNet, an AI diagnostic model, based on nearly 20,000 thyroid nodule ultrasound images. The model was rigorously tested and validated across seven medical centers, demonstrating an external multi-center accuracy that surpassed the level of experienced ultrasonographers with over 10 years of thyroid ultrasound experience. - **AI-Assisted Decision Model:** By integrating ThyNet with the ACR TIRADS (American College of Radiology Thyroid Imaging, Reporting, and Data System) guidelines, the team created an AI-assisted fine needle aspiration (FNA) decision model. This model significantly reduced the proportion of patients requiring invasive FNA biopsies from 87.7% to 53.4%, with only a minimal increase in the false-negative rate of 0.4%. **Clinical Impact and Ethical Considerations:** - **Clinical Decision Influence:** The study also explored the impact of AI on clinical decision-making. When the AI's diagnostic suggestions differed from those of the ultrasonographers, approximately half of the clinicians modified their initial diagnoses. However, a quarter of these modifications were later confirmed by pathology to be incorrect, highlighting the potential ethical risks associated with AI-assisted diagnosis. - **Ethical Data Support:** This research is the first to provide empirical data on how AI influences clinical decisions and medical behavior, contributing to the ongoing debate about the integration of AI into clinical practice and the ethical considerations that come with it. **Contributors and Collaborations:** - **First Authors:** The lead authors of the paper are Professor Sui Peng and Assistant Researcher Yihao Liu from the Clinical Research Center of the First Affiliated Hospital of Sun Yat-sen University, Professor Weiming Lü from the Department of Thyroid and Breast Surgery, and Professor Longzhong Liu from the Cancer Center of Sun Yat-sen University. - **Corresponding Authors:** The final corresponding author is Professor Xiaohai Peng from the Department of Endocrinology at the First Affiliated Hospital of Sun Yat-sen University. Professor Wei Wang from the Department of Ultrasound Medicine and Professor Erik K Alexander from Harvard University are also listed as co-corresponding authors. - **Collaborative Effort:** The project was a collaborative effort involving the Clinical Research Center, Department of Endocrinology, Department of Ultrasound Medicine, Department of Thyroid and Breast Surgery, and the Medical Big Data Center of the First Affiliated Hospital of Sun Yat-sen University. Additional partners included six large tertiary hospitals in South China (First Affiliated Hospital of Guangxi Medical University, Third Affiliated Hospital of Sun Yat-sen University, Sixth Affiliated Hospital of Sun Yat-sen University, First Affiliated Hospital of Guangzhou University of Traditional Chinese Medicine, and the General Hospital of the Southern Theater Command of the Chinese People's Liberation Army) and the Brigham and Women's Hospital at Harvard University. Technical support was provided by Tsinghua University and the Xiaobai Century team. **Significance and Recognition:** - **Innovative Approach:** The study exemplifies the power of interdisciplinary collaboration and the potential of AI to revolutionize clinical diagnostics. By combining expertise from endocrinology, ultrasound medicine, surgery, and big data, the team produced a robust and reliable AI model that can assist in thyroid nodule management. - **Official Recognition:** The journal's official website featured the study on its homepage, underscoring its importance and the innovative nature of the research. **Conclusion:** Professor Xiaohai Peng's team has made a significant contribution to the field of thyroid cancer diagnostics by developing ThyNet, an AI model that enhances the accuracy of thyroid nodule management. The integrated AI-assisted FNA decision model not only reduces the need for invasive procedures but also provides valuable insights into the ethical implications of AI in clinical settings. This research highlights the potential of AI to improve patient care while emphasizing the need for careful integration and ongoing evaluation of AI tools in healthcare. For more details, the full paper can be accessed at: [https://www.thelancet.com/journals/landig/article/PIIS2589-7500(21)00041-8/fulltext](https://www.thelancet.com/journals/landig/article/PIIS2589-7500(21)00041-8/fulltext).