Dual-phase contrast-enhanced computed tomography radiomics combined with clinical factors for predicting minimal extrathyroidal extension in papillary thyroid microcarcinoma

Introduction: Minimal extrathyroidal extension (mETE) is an adverse feature of papillary thyroid microcarcinoma (PTMC), but its preoperative identification remains challenging. Objective: To develop and validate a dual-phase contrast-enhanced computed tomography (CT) radiomics model and evaluate the complementary value of clinical factors for predicting mETE in PTMC. Methods: This retrospective study included 311 patients with pathologically confirmed PTMC, randomly assigned to training (n = 217) and testing (n = 94) cohorts. Radiomics features were extracted from arterial- and venous-phase contrast-enhanced CT images and reduced using correlation analysis and least absolute shrinkage and selection operator logistic regression. Independent clinical and conventional imaging predictors were identified by multivariable logistic regression. Clinical, radiomics, and combined models were evaluated by receiver operating characteristic curve analysis, calibration, decision curve analysis, net reclassification improvement, and integrated discrimination improvement. Results: Thirty-eight radiomics features were retained. Systemic immune-inflammation index, neutrophil-to-lymphocyte ratio, and ultrasonographic capsular contact status were independent predictors of mETE. In the training cohort, the clinical, radiomics, and combined models achieved areas under the curve (AUCs) of 0.765, 0.916, and 0.923, respectively; corresponding AUCs in the testing cohort were 0.615, 0.822, and 0.825. Both the radiomics and combined models demonstrated significantly better discrimination than the clinical model, whereas their difference was not statistically significant (p = 0.894). Conclusion: The dual-phase CT radiomics model showed good discrimination for mETE in PTMC. Clinical factors provided complementary information but did not significantly improve discrimination. The radiomics-based model may serve as an auxiliary tool for preoperative risk assessment. However, prospective multicenter studies and independent external validation are warranted to establish its clinical utility.

Authors

Institutions

Publication Details

Journal
Eurasian Journal of Medicine and Oncology
Published
2026-11-16
DOI
https://doi.org/10.36922/ejmo026300370
Primary Topic
Radiomics and Machine Learning in Medical Imaging
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Dual-phase contrast-enhanced computed tomography radiomics combined with clinical factors for predicting minimal extrathyroidal extension in papillary thyroid microcarcinoma

Jun Ma, Bo Zhou, Ya Zou, Xuehong He et al.
2 links
Eurasian Journal of Medicine and Oncology
Radiomics and Machine Learning in Medical Imaging
article

Dual-phase contrast-enhanced computed tomography radiomics combined with clinical factors for predicting minimal extrathyroidal extension in papillary thyroid microcarcinoma

Jun Ma, Bo Zhou, Ya Zou, Xuehong He, Zuhua Song, Liang Lv, Jie Huang
article en

Abstract

Introduction: Minimal extrathyroidal extension (mETE) is an adverse feature of papillary thyroid microcarcinoma (PTMC), but its preoperative identification remains challenging. Objective: To develop and validate a dual-phase contrast-enhanced computed tomography (CT) radiomics model and evaluate the complementary value of clinical factors for predicting mETE in PTMC. Methods: This retrospective study included 311 patients with pathologically confirmed PTMC, randomly assigned to training (n = 217) and testing (n = 94) cohorts. Radiomics features were extracted from arterial- and venous-phase contrast-enhanced CT images and reduced using correlation analysis and least absolute shrinkage and selection operator logistic regression. Independent clinical and conventional imaging predictors were identified by multivariable logistic regression. Clinical, radiomics, and combined models were evaluated by receiver operating characteristic curve analysis, calibration, decision curve analysis, net reclassification improvement, and integrated discrimination improvement. Results: Thirty-eight radiomics features were retained. Systemic immune-inflammation index, neutrophil-to-lymphocyte ratio, and ultrasonographic capsular contact status were independent predictors of mETE. In the training cohort, the clinical, radiomics, and combined models achieved areas under the curve (AUCs) of 0.765, 0.916, and 0.923, respectively; corresponding AUCs in the testing cohort were 0.615, 0.822, and 0.825. Both the radiomics and combined models demonstrated significantly better discrimination than the clinical model, whereas their difference was not statistically significant (p = 0.894). Conclusion: The dual-phase CT radiomics model showed good discrimination for mETE in PTMC. Clinical factors provided complementary information but did not significantly improve discrimination. The radiomics-based model may serve as an auxiliary tool for preoperative risk assessment. However, prospective multicenter studies and independent external validation are warranted to establish its clinical utility.

Eurasian Journal of Medicine and OncologyVol. 0(0)
Chongqing University (CN)
Good health and well-being
Radiomics and Machine Learning in Medical Imaging
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon