Related Experiment Video
Updated: Jan 11, 2026
05:41
Author Spotlight: Integrating Ultrasound Imaging with Biochemical Markers for Thyroid Disease Diagnosis
Published on: February 9, 2024
1.0K
Nonenhanced CT-Based radiomics model enhances PTC detection in Hashimoto's thyroiditis
Yun Peng1,2, Kaiyao Huang1,2, Zijian Gong1,2
1Department of Radiology, The Second Affiliated Hospital, Jiangxi Medical College, Nanchang University,, Nanchang, 330006, China.
BMC Cancer
|November 12, 2025
Summary
This study developed a radiomics model using nonenhanced CT (NECT) to detect papillary thyroid carcinoma (PTC) in patients with Hashimoto
Area of Science:
- Radiology
- Oncology
- Medical Imaging
Background:
- Hashimoto's thyroiditis (HT) complicates the detection of papillary thyroid carcinoma (PTC) on conventional imaging.
- Diffuse thyroid changes in HT pose diagnostic challenges for ultrasound and CT scans.
- Early PTC diagnosis in HT patients is crucial for effective treatment.
Purpose of the Study:
- To develop a radiomics model using nonenhanced CT (NECT) for predicting PTC in patients with HT.
- To improve the accuracy of early PTC diagnosis in the context of HT.
Main Methods:
- Retrospective analysis of NECT scans from 130 HT patients (with or without PTC).
- Radiomic features extracted using PyRadiomics, with dimensionality reduction via LASSO analysis.
- Four machine learning models (LR, NB, SVM, MLP) were trained and validated.
Main Results:
- Six radiomic features were selected for model development.
- The Multilayer Perceptron (MLP) model demonstrated the best performance in the external validation cohort.
- MLP achieved an AUC of 0.783, with a sensitivity of 0.643 and specificity of 0.923.
Conclusions:
- A radiomics model based on NECT can effectively identify PTC in patients with HT.
- This approach shows potential for enhancing early diagnosis and intervention strategies for PTC in HT patients.
