Papillary thyroid carcinoma (PTC) is the most common type of thyroid cancer. As it can spread to nearby lymph nodes, early assessment is important for treatment planning. To detect central lymph nodes with ultrasound can be difficult, combining information from different ultrasound techniques with clinical data improves preoperative prediction.

In a study published in Journal of Imaging Informatics in Medicine, a team led by Prof. LI Hai from the Hefei Institutes of Physical Science of the Chinese Academy of Sciences developed an explainable artificial intelligence (AI) approach to assess the risk of central lymph node metastasis in patients with PTC before surgery.

Researchers analyzed ultrasound images and clinical information from 428 patients with 508 PTC nodules at four hospitals. The resulting machine-learning model showed good performance in an independent external test, achieving an area under curve (AUC) of 0.844. The model also identified imaging features contributing to each prediction, allowing radiologists to review the evidence behind AI assessment.

To assess clinical usability, a controlled reader study was conducted in which six radiologists with varying experience independently evaluated cases under three conditions: without AI assistance, with basic AI assistance, and with explainable AI assistance. With the support of explainable AI, the radiologists achieved higher diagnostic accuracy and showed greater agreement in their assessments.

"The value of medical AI lies not only in predicting risk, but also in making the evidence behind each prediction visible and open to clinical scrutiny. By combining multimodal imaging with radiologists' expertise, explainable AI supports more transparent and collaborative clinical decision-making," said Prof. LI Hai from the Hefei Institutes of Physical Science, one corresponding author of this study.

The approach developed in this study not only estimates the risk of lymph node metastasis, but also shows the imaging features behind its prediction, helping radiologists better understand the results.