Deepnoid Publishes Research Paper in Official Journal of the American Thyroid Association

The 'AI-based Thyroid Fine Needle Aspiration Cytology (FNAC) Results' jointly researched by Deepnoid, a first-generation domestic medical artificial intelligence (AI) specialized company, St. Vincent's Hospital, and Uijeongbu St. Mary's Hospital has been published in the official journal of the American Thyroid Association (ATA), Thyroid.

Thyroid fine needle aspiration cytology (hereafter thyroid FNAC) is a test used to determine the malignancy of thyroid nodules. It is used in the diagnosis of thyroid diseases because it provides results quickly and simply.

The joint research team proved through the study that using AI in thyroid FNAC increases diagnostic accuracy. The sensitivity, specificity, and accuracy without using AI were 87.3%, 90.6%, and 88.9%, respectively. When AI was applied, the diagnostic results improved to 95.2%, 96.3%, and 95.8%, enhancing all performance metrics. This is the highest level among domestic and international studies published to date.

Researcher Hwang Gi-su of Deepnoid, who conducted the study, said, "AI is increasingly being applied in the fields of pathology and cytology, showing meaningful results," and added, "We plan to continue research in this field to help pathologists in actual medical settings and improve diagnostic efficiency."

The AI model was trained using data from 200 pathology laboratories nationwide collected through the cytopathology quality control project.

This content was produced with the assistance of AI translation services.