Objective:

To evaluate the clinicopathological characteristics and recurrence patterns of bilateral disease in classic papillary thyroid carcinoma (PTC) and to develop a multimodal AI model for preoperative identification of bilateral disease, addressing the ongoing debate regarding its clinical significance compared to unilateral multifocality.

Key Findings:

• 26.1% of patients had multifocal disease, with 62.7% showing bilateral involvement, highlighting the prevalence of bilateral disease.
• Bilateral disease was linked to larger tumor size, higher lymph node metastasis rates, and worse recurrence-free survival, indicating its aggressive nature.
• Bilateral disease was identified as an independent risk factor for recurrence (HR = 9.664, P = 0.005), underscoring its clinical significance.
• Integrated Model D showed superior performance with AUC of 0.970 in training, 0.932 in validation, and 0.848 in external validation, demonstrating its robustness.
• Model D significantly improved diagnostic accuracy for junior radiologists by +20.9%, emphasizing its practical application.

Interpretation:

Bilateral disease in classic PTC is a significant risk factor for recurrence, and the AI-driven model can enhance preoperative identification, potentially transforming surgical planning and patient outcomes.

Limitations:

• The study was retrospective and conducted in a single country, which may limit generalizability to broader populations.
• Potential biases in data collection and patient selection could affect results, necessitating cautious interpretation of findings.

Conclusion:

Bilateral disease in classic PTC poses a higher recurrence risk than unilateral multifocality. The developed AI model effectively predicts bilateral disease preoperatively, improving diagnostic accuracy across varying radiologist experience levels.