Artificial Intelligence (AI) and Parathyroid Gland Preservation During Thyroidectomy

Preserving the parathyroid glands (PGs) is crucial during thyroidectomy to avoid hypoparathyroidism and its related complications. Recently, artificial intelligence (AI) based techniques have been introduced intraoperatively, allowing surgeons to reduce the level of subjectivity and assist with the identification of PGs accurately and quickly. The aim of this narrative review is to assess the current evidence on the application of AI in parathyroid gland (PG) preservation. PubMed, Google Scholar, and Embase were used for the search for various types of studies and reviews dating from 2015 up to 2025, relating PG detection and preservation during thyroidectomy with AI. A total of 368 articles were identified through the database searching. After duplicates were removed and articles were screened, 358 records not fulfilling the criteria were excluded. In total, 11 studies are included, which focus on artificial intelligence or deep learning identification of parathyroid glands during thyroidectomy. The criteria of inclusion were the use or the development and training of an AI model, deep learning, near-infrared autofluorescence (NIRAF), and intraoperative PG identification and preservation. Although available data remain limited, this narrative review suggests that AI-assisted techniques are feasible and have the potential to improve surgical accuracy and efficiency. To confirm this clinical benefit, further high-quality prospective studies are required.