To develop and validate prediction models integrating ultrasonographic features and laboratory indicators for malignancy risk assessment in thyroid nodules.
Key Findings:
Logistic regression model achieved an AUC of 0.924 in internal validation and 0.929 in external validation.
No significant differences in AUCs between models in internal validation; logistic regression outperformed others in external validation.
The logistic regression model showed balanced classification performance and good calibration.
Interpretation:
The logistic regression model provides a stable and clinically interpretable tool for assessing malignancy risk in C-TIRADS 3–4 thyroid nodules.
Limitations:
Retrospective design may introduce bias.
Limited generalizability due to specific patient populations.
Conclusion:
A logistic regression model based on routine clinical and ultrasound information can effectively assess malignancy risk in thyroid nodules.
When Alexander Shifrin, MD, reflects on his 20 years as an endocrine surgeon, what stands out most is not the technical complexity of the operations he performs, but the consistency with which he can offer something rare when it comes to cancer care.
