A new deep-learning AI model, ThyNet, has been developed to help differentiate between malignant tumours and benign thyroid nodules.

Patients with a thyroid nodule who need to undergo a fine needle aspiration (FNA)

feel anxious before the procedure and worry about the evaluation of thyroid nodules.

ThyNet was created to help radiologists improve diagnostic performance. It was trained on 18 049 images of 8339 patients and tested on 4305 images of 2775 patients.

With the help of ThyNet, there will be:

🔹 a decrease in the number of missed malignancy cases

🔹 a significant improvement in the diagnostic performance of radiologists

For patients, these are important results as they show that AI-assisted models:

🔹 will enable early and precise detection

🔹 will offer more accurate diagnosis and timely interventions

🔹 a more precise diagnosis will reduce patients’ worry and fear

Peng S, Liu Y, Lv W, et al. Deep learning-based artificial intelligence model to assist thyroid nodule diagnosis and management: a multicentre diagnostic study. Lancet Digit Health. 2021.