PACS-Integrated AI Predicts Neck Lymph Node Spread in Papillary Thyroid Cancer on CT

A PACS-integrated AI tool for CT-based cervical lymph node assessment in papillary thyroid carcinoma achieved AUCs of 0.948 (internal) and 0.943 (external) for metastasis prediction, and significantly improved diagnostic accuracy among junior radiologists (p<0.05).

• Neck-level localization precision ranged from 0.919–0.990 (internal) and 0.848–1.000 (external) across levels I–VI.
• Lymph node segmentation yielded mean Dice coefficients of 0.674 (internal) and 0.633 (external).
• In a reader study, AI assistance boosted junior radiologists' accuracy (all p < 0.05).

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.