Evaluating the PTFinder Dual-Channel Near-Infrared and White-Light Imaging System with AI to Quickly Identify Parathyroid Glands in Surgical Specimens

What this Trial Is About

Researchers are evaluating the PTFinder system, a device that uses dual-camera near-infrared autofluorescence imaging to make parathyroid glands glow on screen during thyroid or parathyroid surgery. The study aims to measure how quickly and accurately PTFinder identifies real parathyroid glands compared to white-light examination. This observational study involves about 180 adult patients undergoing surgery at three Chinese hospitals and explores whether improved gland detection reduces the risk of low blood calcium after surgery. Additionally, collected images will help develop and test an AI model for automated gland recognition. During the study, excised thyroid or parathyroid tissue specimens will be scanned ex vivo with PTFinder, timing how fast the first gland is found and marking suspected glands. Then, a blinded investigator will examine the same tissue under normal white light. Marked tissue and random negative samples will undergo frozen section or rapid PTH testing, with any differences resolved by permanent histology. The procedure adds only a few minutes to surgery and does not alter the surgical process. The collected data will also be used to train a machine-learning algorithm for parathyroid recognition as a secondary goal. Participants will have blood tests for calcium and parathyroid hormone at 1 hour, 24 hours, and 7 days after surgery to monitor gland recovery and hypocalcemia risk. Data accuracy and device performance will be analyzed statistically, and all information is securely stored. The study involves close monitoring during surgery and post-operative follow-up visits, lasting up to seven days after surgery to evaluate safety and gland function.

Dual-Channel Near-Infrared Autofluorescence Imaging and AI Analysis to Locate Parathyroid Glands (PTFinder)