Radiomics Nomogram for Identifying Sub-1 cm Benign and Malignant Thyroid Lesions

CT-Based Radiomics Models for Differentiation of Benign and Malignant Thyroid Nodules: A Multicenter Development and Validation Study

BACKGROUND. CT is increasingly detecting thyroid nodules. Prior studies indicated a potential role of CT-based radiomics models in characterizing thyroid nodules, although these studies lacked external validation. OBJECTIVE. The purpose of this study was to develop and validate a CT-based radiomics model for the differentiation of benign and malignant thyroid nodules. METHODS. This retrospective study included 378 patients (mean age, 46.3 ± 13.9 [SD] years; 86 men, 292 women) with 408 resected thyroid nodules (145 benign, 263 malignant) from two centers (center 1: 293 nodules, January 2018 to December 2022; center 2: 115 nodules, January 2020 to December 2022) who underwent preoperative multiphase neck CT (noncontrast, arterial, and venous phases). Nodules from center 1 were divided into training (n = 206) and internal validation (n = 87) sets; all nodules from center 2 formed an external validation set. Radiologists assessed nodules for morphologic CT features. Nodules were manually segmented on all phases, and radiomic features were extracted. Conventional (clinical and morphologic CT), noncontrast CT radiomics, arterial phase CT radiomics, venous phase CT radiomics, multiphase CT radiomics, and combined (clinical, morphologic CT, and multiphase CT radiomics) models were established using feature selection methods and evaluated by ROC curve analysis, calibration-curve analysis, and decision-curve analysis. RESULTS. The combined model included patient age, three morphologic features (cystic change, "edge interruption" sign, abnormal cervical lymph nodes), and 28 radiomic features (from all three phases). In the external validation set, the combined model had an AUC of 0.923, and, at an optimal threshold derived in the training set, sensitivity of 84.0%, specificity of 94.1%, and accuracy of 87.0%. In the external validation set, the AUC was significantly higher for the combined model than for the conventional model (0.827), noncontrast CT radiomics model (0.847), arterial phase CT radiomics model (0.826), venous phase CT radiomics model (0.773), and multiphase CT radiomics model (0.824) (all p < .05). In the external validation set, the calibration curves indicated the lowest (i.e., best) Brier score for the combined model; in the decision-curve analysis, the combined model had the highest net benefit for most of the range of threshold probabilities. CONCLUSION. A combined model incorporating clinical, morphologic CT, and multiphase CT radiomics features exhibited robust performance in differentiating benign and malignant thyroid nodules. CLINICAL IMPACT. The combined radiomics model may help guide further management for thyroid nodules detected on CT.

Predicting papillary thyroid microcarcinoma in American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS) 3 nodules: radiomics analysis based on intratumoral and peritumoral ultrasound images

Background

A subset of patients undergoing thyroid surgery for presumed benign thyroid disease presented with papillary thyroid microcarcinoma (PTMC). A non-invasive and precise method for early recognition of PTMC are urgently needed. The aim of this study was to construct and validate a nomogram that combines intratumoral and peritumoral radiomics features as well as clinical features for predicting PTMC in the American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS) 3 nodules using ultrasonography.

Methods

A retrospective review was conducted on a cohort of 221 patients who presented with ACR TI-RADS 3 nodules. These patients were subsequently pathologically diagnosed with either PTMC or benign thyroid nodules. These patients were randomly divided into a training and test cohort with an 8:2 ratio for developing the clinical model, intratumor-region model, peritumor-region model and the combined-region model respectively. The radiomics features were extracted from ultrasound (US) images of each patient. We employed K-nearest neighbor (KNN) model as the base model for building the radiomics signature and clinical signature. Finally, a radiomics-clinical nomogram that combined intratumoral and peritumoral radiomics features as well as clinical features was developed. The prediction performance of each model was assessed by the area under the curve (AUC), sensitivity, specificity and calibration curve.

Results

A total of 23 radiomics features were selected to develop radiomics models. The combined-region radiomics model showed favorable prediction efficiency in both the training dataset (AUC: 0.955) and the test dataset (AUC: 0.923). A radiomics-clinical nomogram was constructed and achieved excellent calibration and discrimination, which yielded an AUC value of 0.950, a sensitivity of 0.950 and a specificity of 0.920.

Conclusions

This study proposed the nomogram that contributes to the accurate and intuitive identification of PTMC in ACR TI-RADS 3 nodules.

A Novel Interpretable Radiomics Model to Distinguish Nodular Goiter From Malignant Thyroid Nodules

OBJECTIVES

The purpose of this study is to inquire about the potential association between radiomics features and the pathological nature of thyroid nodules (TNs), and to propose an interpretable radiomics-based model for predicting the risk of malignant TN.

METHODS

In this retrospective study, computed tomography (CT) imaging and pathological data from 141 patients with TN were collected. The data were randomly stratified into a training group (n = 112) and a validation group (n = 29) at a ratio of 4:1. A total of 1316 radiomics features were extracted by using the pyradiomics tool. The redundant features were removed through correlation testing, and the least absolute shrinkage and selection operator (LASSO) or the minimum redundancy maximum relevance standard was used to select features. Finally, 4 different machine learning models (RF Hybrid Feature, SVM Hybrid Feature, RF, and LASSO) were constructed. The performance of the 4 models was evaluated using the receiver operating characteristic curve. The calibration curve, decision curve analysis, and SHapley Additive exPlanations method were used to evaluate or explain the best radiomics machine learning model.

RESULTS

The optimal radiomics model (RF Hybrid Feature model) demonstrated a relatively high degree of discrimination with an area under the receiver operating characteristic curve (AUC) of 0.87 (95% CI, 0.70-0.97; P < 0.001) for the validation cohort. Compared with the commonly used LASSO model (AUC, 0.78; 95% CI, 0.60-0.91; P < 0.01), there is a significant improvement in AUC in the validation set, net reclassification improvement, 0.79 (95% CI, 0.13-1.46; P < 0.05), and integrated discrimination improvement, 0. 20 (95% CI, 0.10-0.30; P < 0.001).

CONCLUSION

The interpretable radiomics model based on CT performs well in predicting benign and malignant TNs by using quantitative radiomics features of the unilateral total thyroid. In addition, the data preprocessing method incorporating different layers of features has achieved excellent experimental results.

CLINICAL RELEVANCE STATEMENT

As the detection rate of TNs continues to increase, so does the diagnostic burden on radiologists. This study establishes a noninvasive, interpretable and accurate machine learning model to rapidly identify the nature of TN found in CT.

Radiomics nomogram for predicting axillary lymph node metastasis in breast cancer based on DCE-MRI: A multicenter study

OBJECTIVES

This study aims to develop and validate a radiomics nomogram based on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) to noninvasively predict axillary lymph node (ALN) metastasis in breast cancer.

METHODS

This retrospective study included 263 patients with histologically proven invasive breast cancer and who underwent DCE-MRI examination before surgery in two hospitals. All patients had a defined ALN status based on pathological examination results. Regions of interest (ROIs) of the primary tumor and ipsilateral ALN were manually drawn. A total of 1,409 radiomics features were initially computed from each ROI. Next, the low variance threshold, SelectKBest, and least absolute shrinkage and selection operator (LASSO) algorithms were used to extract the radiomics features. The selected radiomics features were used to establish the radiomics signature of the primary tumor and ALN. A radiomics nomogram model, including the radiomics signature and the independent clinical risk factors, was then constructed. The predictive performance was evaluated by the receiver operating characteristic (ROC) curves, calibration curve, and decision curve analysis (DCA) by using the training and testing sets.

RESULTS

ALNM rates of the training, internal testing, and external testing sets were 43.6%, 44.3% and 32.3%, respectively. The nomogram, including clinical risk factors (tumor diameter) and radiomics signature of the primary tumor and ALN, showed good calibration and discrimination with areas under the ROC curves of 0.884, 0.822, and 0.813 in the training, internal and external testing sets, respectively. DCA also showed that radiomics nomogram displayed better clinical predictive usefulness than the clinical or radiomics signature alone.

CONCLUSIONS

The radiomics nomogram combined with clinical risk factors and DCE-MRI-based radiomics signature may be used to predict ALN metastasis in a noninvasive manner.

A review of radiomics and genomics applications in cancers: the way towards precision medicine

The application of radiogenomics in oncology has great prospects in precision medicine. Radiogenomics combines large volumes of radiomic features from medical digital images, genetic data from high-throughput sequencing, and clinical-epidemiological data into mathematical modelling. The amalgamation of radiomics and genomics provides an approach to better study the molecular mechanism of tumour pathogenesis, as well as new evidence-supporting strategies to identify the characteristics of cancer patients, make clinical decisions by predicting prognosis, and improve the development of individualized treatment guidance. In this review, we summarized recent research on radiogenomics applications in solid cancers and presented the challenges impeding the adoption of radiomics in clinical practice. More standard guidelines are required to normalize radiomics into reproducible and convincible analyses and develop it as a mature field.

A systematic review and meta-analysis comparing tumor progression and complications between radiofrequency ablation and thyroidectomy for papillary thyroid carcinoma

BACKGROUND

Papillary thyroid cancer (PTC) is the most frequent thyroid cancers worldwide. The efficacy and acceptability of radiofrequency ablation (RFA) in the treatment of PTC have been intensively studied. The aim of this study is to focus on extra detailed that may influent for PTC or papillary thyroid microcarcinoma (PTMC).

MATERIALS AND METHODS

We identified a total of 1,987 records of a primary literature searched in PubMed, Embase, Cochrane Library, and Google Scholar by key words, from 2000 to 2022. The outcome of studies included complication, costs, and local tumor progression. After scrutiny screening and full-text assessment, six studies were included in the systematic review. Heterogeneity was estimated using I2, and the quality of evidence was assessed for each outcome using the GRADE guidelines.

RESULTS

Our review enrolled 1,708 patients reported in six articles in the final analysis. There were 397 men and 1,311 women in the analysis. Two of these studies involved PTC and four focused on PTMC. There were 859 patients in the RFA group and 849 patients in the thyroidectomy group. By contrast, the tumor progression of RFA group was as same as that surgical groups [odds ratio, 1.31; 95% CI, 0.52-3.29; heterogeneity (I2 statistic), 0%, p = 0.85]. The risk of complication rates was significantly lower in the RFA group than that in the surgical group [odds ratio, 0.18; 95% CI, 0.09-0.35; heterogeneity (I2 statistic), 40%, p = 0.14].

CONCLUSIONS

RFA is a safe procedure with a certain outcome for PTC. RFA can achieve a good efficacy and has a lower risk of major complications.

The diagnostic value of a nomogram based on multimodal ultrasonography for thyroid-nodule differentiation: A multicenter study

Objective

To establish and verify a nomogram based on multimodal ultrasonography (US) for the assessment of the malignancy risk of thyroid nodules and to explore its value in distinguishing benign from malignant thyroid nodules.

Methods

From September 2020 to December 2021, the data of 447 individuals with thyroid nodules were retrieved from the multicenter database of medical images of the National Health Commission’s Capacity Building and Continuing Education Center, which includes data from more than 20 hospitals. All patients underwent contrast-enhanced US (CEUS) and elastography before surgery or fine needle aspiration. The training set consisted of three hundred datasets from the multicenter database (excluding Zhejiang Cancer Hospital), and the external validation set consisted of 147 datasets from Zhejiang Cancer Hospital. As per the pathological results, the training set was separated into benign and malignant groups. The characteristics of the lesions in the two groups were analyzed and compared using conventional US, CEUS, and elastography score. Using multivariate logistic regression to screen independent predictive risk indicators, then a nomogram for risk assessment of malignant thyroid nodules was created. The diagnostic performance of the nomogram was assessed utilizing calibration curves and receiver operating characteristic (ROC) from the training and validation cohorts. The nomogram and The American College of Radiology Thyroid Imaging, Reporting and Data System were assessed clinically using decision curve analysis (DCA).

Results

Multivariate regression showed that irregular shape, elastography score (≥ 3), lack of ring enhancement, and unclear margin after enhancement were independent predictors of malignancy. During the training (area under the ROC [AUC]: 0.936; 95% confidence interval [CI]: 0.902–0.961) and validation (AUC: 0.902; 95% CI: 0.842–0.945) sets, the multimodal US nomogram with these four variables demonstrated good calibration and discrimination. The DCA results confirmed the good clinical applicability of the multimodal US nomogram for predicting thyroid cancer.

Conclusions

As a preoperative prediction tool, our multimodal US-based nomogram showed good ability to distinguish benign from malignant thyroid nodules.

Radiomics Analysis of Computed Tomography for Prediction of Thyroid Capsule Invasion in Papillary Thyroid Carcinoma: A Multi-Classifier and Two-Center Study

OBJECTIVE

To investigate the application of computed tomography (CT)-based radiomics model for prediction of thyroid capsule invasion (TCI) in papillary thyroid carcinoma (PTC).

METHODS

This retrospective study recruited 412 consecutive PTC patients from two independent institutions and randomly assigned to training (n=265), internal test (n=114) and external test (n=33) cohorts. Radiomics features were extracted from non-contrast (NC) and artery phase (AP) CT scans. We also calculated delta radiomics features, which are defined as the absolute differences between the extracted radiomics features. One-way analysis of variance and least absolute shrinkage and selection operator were used to select optimal radiomics features. Then, six supervised machine learning radiomics models (k-nearest neighbor, logistic regression, decision tree, linear support vector machine [L-SVM], Gaussian-SVM, and polynomial-SVM) were constructed. Univariate was used to select clinicoradiological risk factors. Combined models including optimal radiomics features and clinicoradiological risk factors were constructed by these six classifiers. The prediction performance was evaluated using the receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA).

RESULTS

In the internal test cohort, the best combined model (L-SVM, AUC=0.820 [95% CI 0.758-0.888]) performed better than the best radiomics model (L-SVM, AUC = 0.733 [95% CI 0.654-0.812]) and the clinical model (AUC = 0.709 [95% CI 0.649-0.783]). Combined-L-SVM model combines 23 radiomics features and 1 clinicoradiological risk factor (CT-reported TCI). In the external test cohort, the AUC was 0.776 (0.625-0.904) in the combined-L-SVM model, showing that the model is stable. DCA demonstrated that the combined model was clinically useful.

CONCLUSIONS

Our combined model based on machine learning incorporated with CT radiomics features and the clinicoradiological risk factor shows good predictive ability for TCI in PTC.