JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

Artificial intelligence-assisted precise preoperative prediction of lateral cervical lymph nodes metastasis in papillary thyroid carcinoma via a clinical-CT radiomic combined model

  • 0Department of Breast and Thyroid Surgery, Southwest Hospital, Army Medical University, Chongqing, China.
International Journal of Surgery (london, England) +

|

February 4, 2025

|

February 4, 2025

View abstract on PubMed

Summary

This summary is machine-generated.

This study developed an AI model using CT radiomics to predict lateral cervical lymph node metastasis in papillary thyroid cancer. The combined model accurately identifies patients needing lymph node dissection, improving surgical decisions.

Area Of Science

  • Oncology
  • Radiology
  • Artificial Intelligence

Background

  • Papillary thyroid carcinoma (PTC) is the most common thyroid malignancy.
  • Accurate preoperative prediction of lateral cervical lymph node metastasis (LCLNM) is crucial for surgical planning.
  • Current diagnostic methods for LCLNM have limitations.

Purpose Of The Study

  • To develop and validate an artificial intelligence-assisted model for predicting LCLNM in PTC using computed tomography (CT) radiomics.
  • To provide a noninvasive and accurate tool for identifying PTC patients with LCLNM.
  • To enhance surgical decision-making for lateral cervical lymph node dissection.

Main Methods

  • Retrospective study of 389 PTC patients (training, internal, and external validation sets).
  • Extraction of radiomic features from contrast-enhanced CT images.
  • Development of clinical, radiomic, and combined clinical-radiomic models using machine learning algorithms (including Random Forest).
  • Performance evaluation using ROC curves, calibration curves, and decision curve analysis.

Main Results

  • 13 radiomic features and 4 clinical factors (age, tumor location, capsule invasion, central lymph node metastasis) were associated with LCLNM.
  • The Random Forest-based combined clinical-radiomic model achieved high predictive performance.
  • Areas under the ROC curves were 0.910 (training), 0.876 (internal validation), and 0.821 (external validation).
  • Decision curve analysis confirmed the clinical utility of the combined model.

Conclusions

  • A robust clinical-CT radiomic combined model for predicting LCLNM in PTC was successfully established.
  • This AI-assisted model shows significant potential to improve preoperative risk stratification.
  • The model can aid in optimizing surgical strategies, particularly for lateral cervical lymph node dissection.