Performance Evaluation of the Software Medical Device to Detect the Thyrotoxic State in Patients With Hyperthyroidism

What this Trial Is About

Researchers are evaluating the performance of Glandy HYPER, a software medical device that uses heart rate data collected from commercially available wearable devices to detect thyrotoxicosis in patients with Graves' disease. This observational study focuses on adults aged 22 and older who are either newly diagnosed with or currently treated for Graves' disease. The goal is to compare the device's analysis with thyroid function test results, specifically free T4 levels, without changing standard clinical care. Participants will wear a smartwatch (Apple or Samsung) to monitor heart rate during sleep over a 12-week period. Blood samples for thyroid function tests will be taken at baseline and at 4, 8, and 12 weeks. The software device will analyze the heart rate data alongside test results to assess its ability to detect thyrotoxicosis. No treatment or intervention will be applied, and the device's output will not be shared with participants or used to guide their care. During the study, heart rate data from the wearable devices and thyroid function test results will be collected at multiple time points to evaluate the device's accuracy. Researchers will measure outcomes such as the F1 score, sensitivity, specificity, and area under the curve (AUC) of the software's detection of thyrotoxicosis. The study is classified as low risk and participants will continue their usual medical care throughout the 12 weeks. The study also aims to compare its results with previous studies conducted in Korea to assess the software's generalizability.

Evaluating an AI Tool for Detecting Thyrotoxic States