Machine learning (ML) is an interdisciplinary sector in the subset of artificial intelligence (AI) that creates systems to set up logical connections using algorithms, and thus offers predictions for complex data analysis. In the present review, an up-to-date summary of the current state of the art regarding ML and AI implementation for thyroid nodule ultrasound characterization and cancer is provided, highlighting controversies over AI application as well as possible benefits of ML, such as, for example, training purposes. There is evidence that AI increases diagnostic accuracy and significantly limits inter-observer variability by using standardized mathematical algorithms. It could also be of aid in practice settings with limited sub-specialty expertise, offering a second opinion by means of radiomics and computer-assisted diagnosis. The introduction of AI represents a revolutionary event in thyroid nodule evaluation, but key issues for further implementation include integration with radiologist expertise, impact on workflow and efficiency, and performance monitoring.

Background: The aim of this study was to evaluate the ability of intraoperative neuromonitoring in reducing the postoperative recurrent laryngeal nerve palsy rate by a comparison between patients submitted to thyroidectomy with intraoperative neuromonitoring and with routine identification alone. Methods: Between June 2007 and December 2012, 2034 consecutive patients underwent thyroidectomy by a single surgical team. We compared patients who have had neuromonitoring and patients who have undergone surgery with nerve visualization alone. Patients in which neuromonitoring was not utilized (Group A) were 993, patients in which was utilized (group B) were 1041. Results: In group A 28 recurrent laryngeal nerve injuries were observed (2.82%), 21 (2.11%) transient and 7 (0.7%) permanent. In group B 23 recurrent laryngeal nerve injuries were observed (2.21%), in 17 cases (1.63%) transient and in 6 (0.58%) permanent. Differences were not statistically significative. Conclusions: Visual nerve identification remains the gold standard of recurrent laryngeal nerve management in thyroid surgery. Neuromonitoring helps to identify the nerve, in particular in difficult cases, but it did not decrease nerve injuries compared with visualization alone. Future studies are warranted to evaluate the benefit of intraoperative neuromonitoring in thyroidectomy, especially in conditions in which the recurrent nerve is at high risk of injury.

Cervical lymph node metastases in papillary thyroid cancer are common. Although central neck dissection is indicated in clinically nodal-positive disease, it remains controversial in patients with no clinical evidence of nodal metastasis. The aim of this retrospective study was to determine the outcomes of clinically lymph node-negative patients with papillary thyroid cancer who underwent total thyroidectomy without a central neck dissection, in order to determine the rates of recurrence and reoperation in these patients compared with a group of patients submitted to total thyroidectomy with central neck dissection.