Related Experiment Video
Updated: Aug 11, 2025
03:55
Computer-Aided Three-Dimensional Visualization in the Treatment of Locally Advanced Thyroid Cancer
Published on: June 9, 2023
581
Deep learning-based multifeature integration robustly predicts central lymph node metastasis in papillary thyroid
Zhongzhi Wang1, Limeng Qu2, Qitong Chen2
1Department of General Surgery, the Affiliated Zhuzhou Hospital Xiangya Medical College, Central South University, Zhuzhou, Hunan, China.
BMC Cancer
|February 8, 2023
Summary
Accurate diagnosis of central lymph node metastasis in papillary thyroid cancer (PTC) is challenging. A deep learning model integrating multiple features shows high efficacy in predicting PTC lymph node metastasis, aiding clinical decisions.
Area of Science:
- Oncology
- Medical Imaging
- Artificial Intelligence
Background:
- Accurate diagnosis of central lymph node metastasis (CLNM) in papillary thyroid cancer (PTC) remains a clinical challenge.
- Genetic sequencing advances personalized cancer therapy development by targeting specific genetic variants.
Purpose of the Study:
- To develop and validate accurate prediction models for CLNM in PTC.
- To explore the utility of artificial intelligence (AI) and deep learning in predicting PTC metastasis.
Main Methods:
- Analysis of clinicopathological data from 488 PTC patients.
- Construction of a nomogram prediction model using logistic regression.
- Development of a convolutional neural network (CNN) deep learning model.
Main Results:
- Independent risk factors for CLNM identified: age, nodule diameter, capsular invasion, and BRAF V600E mutation.
- The CNN model achieved an AUC of 0.89 (training) and 0.78 (test) for CLNM prediction.
- Validated models demonstrated high sensitivity and specificity across different metastasis groups, including small nodules (<1 cm).
Conclusions:
- A deep learning-based multifeature integration model offers a valuable tool for clinical diagnosis and treatment planning in PTC.
- AI-driven prediction models show high efficacy and broad applicability for CLNM detection.
