Multimodal ultrasound radiomics containing microflow images for the prediction of central lymph node metastasis in papillary thyroid carcinoma

ObjectivesThis study aimed to construct a model by applying radiomics and machine learning (ML) to multimodal ultrasound images (including grayscale, elastography and microflow images) along with clinical data to predict central lymph node metastasis (CLNM) in patients with papillary thyroid cancer...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiangyuan Ben, Qiying Yv, Pengfei Zhu, Junhao Ren, Pu Zhou, Guifang Chen, Ying He
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-07-01
Series:Frontiers in Oncology
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fonc.2025.1604951/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849318342133809152
author Jiangyuan Ben
Jiangyuan Ben
Qiying Yv
Pengfei Zhu
Junhao Ren
Pu Zhou
Guifang Chen
Ying He
Ying He
author_facet Jiangyuan Ben
Jiangyuan Ben
Qiying Yv
Pengfei Zhu
Junhao Ren
Pu Zhou
Guifang Chen
Ying He
Ying He
author_sort Jiangyuan Ben
collection DOAJ
description ObjectivesThis study aimed to construct a model by applying radiomics and machine learning (ML) to multimodal ultrasound images (including grayscale, elastography and microflow images) along with clinical data to predict central lymph node metastasis (CLNM) in patients with papillary thyroid cancer (PTC).MethodsA cohort of 213 patients who underwent thyroidectomy accompanied by lymph node dissection (LND) and were pathologically diagnosed with PTC postoperatively was enrolled and randomized to the training cohort (n = 170) or testing cohort (n = 43). Radiomics features were extracted from multimodal images and subsequently screened via the least absolute shrinkage and selection operator (LASSO). The same methods were applied to screen clinical features. Nine ML algorithms were used to construct clinical models, radiomics models and fusion models. Model performance was assessed via receiver operating characteristic curves (ROC), decision curve analysis (DCA), and Delong test. Finally, the optimal model was interpreted and visualized via Shapley additive explanation (SHAP).ResultsIn each modality, 1561 features were extracted from the ultrasound images. Sixteen features were ultimately retained, including 6 grayscale features, 6 elastography features, and 4 microflow features. From the clinical features, including gender, age, traditional ultrasound signs and serological indicators, 2 relevant features were selected. Among the prediction models, the fusion model constructed by Multilayer Perceptron (MLP) algorithm showed the best diagnostic performance, outperforming the other models in both the training cohort (AUC = 0.886) and the testing cohort (AUC = 0.873).ConclusionsThe fusion model based on clinical data and multimodal ultrasound radiomics has better predictive ability and net clinical benefit for CLNM in patients with PTC, confirms the diagnostic value of microflow images for CLNM, and can help to evaluate patients’ preoperative lymph node status and make the correct decision on the surgical procedure.
format Article
id doaj-art-630f9262f0e5444ba86aea2061fdbe1f
institution Kabale University
issn 2234-943X
language English
publishDate 2025-07-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Oncology
spelling doaj-art-630f9262f0e5444ba86aea2061fdbe1f2025-08-20T03:50:53ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2025-07-011510.3389/fonc.2025.16049511604951Multimodal ultrasound radiomics containing microflow images for the prediction of central lymph node metastasis in papillary thyroid carcinomaJiangyuan Ben0Jiangyuan Ben1Qiying Yv2Pengfei Zhu3Junhao Ren4Pu Zhou5Guifang Chen6Ying He7Ying He8Cancer Research Center Nantong, Affiliated Tumor Hospital of Nantong University, and Medical School of Nantong University, Nantong, ChinaDepartment of Ultrasound, Affiliated Tumor Hospital of Nantong University, Nantong, Jiangsu, ChinaCancer Research Center Nantong, Affiliated Tumor Hospital of Nantong University, and Medical School of Nantong University, Nantong, ChinaDepartment of Ultrasound, Affiliated Tumor Hospital of Nantong University, Nantong, Jiangsu, ChinaDepartment of Ultrasound, Affiliated Tumor Hospital of Nantong University, Nantong, Jiangsu, ChinaDepartment of Ultrasound, Affiliated Tumor Hospital of Nantong University, Nantong, Jiangsu, ChinaDepartment of Ultrasound, Affiliated Tumor Hospital of Nantong University, Nantong, Jiangsu, ChinaCancer Research Center Nantong, Affiliated Tumor Hospital of Nantong University, and Medical School of Nantong University, Nantong, ChinaDepartment of Ultrasound, Affiliated Tumor Hospital of Nantong University, Nantong, Jiangsu, ChinaObjectivesThis study aimed to construct a model by applying radiomics and machine learning (ML) to multimodal ultrasound images (including grayscale, elastography and microflow images) along with clinical data to predict central lymph node metastasis (CLNM) in patients with papillary thyroid cancer (PTC).MethodsA cohort of 213 patients who underwent thyroidectomy accompanied by lymph node dissection (LND) and were pathologically diagnosed with PTC postoperatively was enrolled and randomized to the training cohort (n = 170) or testing cohort (n = 43). Radiomics features were extracted from multimodal images and subsequently screened via the least absolute shrinkage and selection operator (LASSO). The same methods were applied to screen clinical features. Nine ML algorithms were used to construct clinical models, radiomics models and fusion models. Model performance was assessed via receiver operating characteristic curves (ROC), decision curve analysis (DCA), and Delong test. Finally, the optimal model was interpreted and visualized via Shapley additive explanation (SHAP).ResultsIn each modality, 1561 features were extracted from the ultrasound images. Sixteen features were ultimately retained, including 6 grayscale features, 6 elastography features, and 4 microflow features. From the clinical features, including gender, age, traditional ultrasound signs and serological indicators, 2 relevant features were selected. Among the prediction models, the fusion model constructed by Multilayer Perceptron (MLP) algorithm showed the best diagnostic performance, outperforming the other models in both the training cohort (AUC = 0.886) and the testing cohort (AUC = 0.873).ConclusionsThe fusion model based on clinical data and multimodal ultrasound radiomics has better predictive ability and net clinical benefit for CLNM in patients with PTC, confirms the diagnostic value of microflow images for CLNM, and can help to evaluate patients’ preoperative lymph node status and make the correct decision on the surgical procedure.https://www.frontiersin.org/articles/10.3389/fonc.2025.1604951/fullpapillary thyroid cancermultimodal ultrasoundmicroflowelastographylymph node metastasismachine learning
spellingShingle Jiangyuan Ben
Jiangyuan Ben
Qiying Yv
Pengfei Zhu
Junhao Ren
Pu Zhou
Guifang Chen
Ying He
Ying He
Multimodal ultrasound radiomics containing microflow images for the prediction of central lymph node metastasis in papillary thyroid carcinoma
Frontiers in Oncology
papillary thyroid cancer
multimodal ultrasound
microflow
elastography
lymph node metastasis
machine learning
title Multimodal ultrasound radiomics containing microflow images for the prediction of central lymph node metastasis in papillary thyroid carcinoma
title_full Multimodal ultrasound radiomics containing microflow images for the prediction of central lymph node metastasis in papillary thyroid carcinoma
title_fullStr Multimodal ultrasound radiomics containing microflow images for the prediction of central lymph node metastasis in papillary thyroid carcinoma
title_full_unstemmed Multimodal ultrasound radiomics containing microflow images for the prediction of central lymph node metastasis in papillary thyroid carcinoma
title_short Multimodal ultrasound radiomics containing microflow images for the prediction of central lymph node metastasis in papillary thyroid carcinoma
title_sort multimodal ultrasound radiomics containing microflow images for the prediction of central lymph node metastasis in papillary thyroid carcinoma
topic papillary thyroid cancer
multimodal ultrasound
microflow
elastography
lymph node metastasis
machine learning
url https://www.frontiersin.org/articles/10.3389/fonc.2025.1604951/full
work_keys_str_mv AT jiangyuanben multimodalultrasoundradiomicscontainingmicroflowimagesforthepredictionofcentrallymphnodemetastasisinpapillarythyroidcarcinoma
AT jiangyuanben multimodalultrasoundradiomicscontainingmicroflowimagesforthepredictionofcentrallymphnodemetastasisinpapillarythyroidcarcinoma
AT qiyingyv multimodalultrasoundradiomicscontainingmicroflowimagesforthepredictionofcentrallymphnodemetastasisinpapillarythyroidcarcinoma
AT pengfeizhu multimodalultrasoundradiomicscontainingmicroflowimagesforthepredictionofcentrallymphnodemetastasisinpapillarythyroidcarcinoma
AT junhaoren multimodalultrasoundradiomicscontainingmicroflowimagesforthepredictionofcentrallymphnodemetastasisinpapillarythyroidcarcinoma
AT puzhou multimodalultrasoundradiomicscontainingmicroflowimagesforthepredictionofcentrallymphnodemetastasisinpapillarythyroidcarcinoma
AT guifangchen multimodalultrasoundradiomicscontainingmicroflowimagesforthepredictionofcentrallymphnodemetastasisinpapillarythyroidcarcinoma
AT yinghe multimodalultrasoundradiomicscontainingmicroflowimagesforthepredictionofcentrallymphnodemetastasisinpapillarythyroidcarcinoma
AT yinghe multimodalultrasoundradiomicscontainingmicroflowimagesforthepredictionofcentrallymphnodemetastasisinpapillarythyroidcarcinoma