Related Experiment Video
Updated: Sep 19, 2025
04:23
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
Published on: April 21, 2023
2.0K
Construction of a Multimodal Machine Learning Model for Papillary Thyroid Carcinoma Based on Pathomics and Ultrasound
Yu-Yan Pang1, Zhong-Qing Tang2, Chang Song1
1Department of Pathology, The First Affiliated Hospital of Guangxi Medical University, No. 6 Shuangyong Road, Nanning, Guangxi Zhuang Autonomous Region, 530021, PR China.
Data in Brief
|June 19, 2025
Summary
This study developed an artificial intelligence model integrating pathomics and ultrasound radiomics for improved papillary thyroid carcinoma (PTC) diagnosis. The multimodal approach enhances diagnostic efficiency and accuracy for pathologists.
Area of Science:
- Oncology
- Medical Imaging
- Artificial Intelligence
Background:
- Accurate tumor diagnosis is crucial for effective cancer treatment.
- Integrating diverse data sources can enhance diagnostic precision.
- Papillary thyroid carcinoma (PTC) diagnosis can benefit from advanced analytical methods.
Purpose of the Study:
- To develop a multimodal diagnostic model for papillary thyroid carcinoma (PTC).
- To integrate pathomics and ultrasound radiomics features using machine learning.
- To improve the diagnostic efficiency and accuracy for pathologists.
Main Methods:
- Retrospective analysis of 222 PTC cases and 163 benign thyroid nodule cases.
- Extraction of pathomics from cytopathology images and ultrasound radiomics from lesion outlines.
- Application of eXtreme gradient boosting (XGBoost), support vector machine (SVM), and random forest (RF) algorithms.
Main Results:
- Construction of multimodal diagnostic models for PTC using AI machine learning.
- Evaluation of model efficacy using area under the receiver operating characteristic curve (AUC).
- Comparison of single-modal, multimodal, and artificial diagnostic model performance.
Conclusions:
- Multimodal integration of pathomics and ultrasound radiomics shows promise for PTC diagnosis.
- AI-driven models can enhance the efficiency of pathologists in diagnosing thyroid nodules.
- Further validation is needed to confirm the clinical utility of the developed models.
Keywords:
Machine learningMultimodal modelPapillary thyroid carcinomaRandom forestSupport vector machine
