Value of T2-weighted-based radiomics model in distinguishing Warthin tumor from pleomorphic adenoma of the parotid

Differentiation Between Parotid Adenolymphoma and Malignant Tumor Based on Multimodal Functional MRI of Radiomics and Intratumoral Vascular ITSS Classification

BACKGROUND

Differentiating between parotid adenolymphoma and malignant tumors remains challenging.

PURPOSE

This study aims to improve preoperative diagnosis accuracy by evaluating the role of multimodal functional magnetic resonance imaging (MRI) and advanced radiomics analysis.

METHODS

We retrospectively analyzed 124 patients with adenolymphoma and malignant parotid tumors, divided into primary (n = 84) and test (n = 40) cohorts. Tumor regions were manually labeled on susceptibility-weighted imaging (SWI), diffusion-weighted imaging (DWI), and contrast-enhanced T1-weighted imaging (CE-T1WI). Seven radiomics models were constructed using logistic regression. We also incorporated intratumoral susceptibility signal (ITSS) grading and performed histogram analysis of apparent diffusion coefficient (ADC) maps.

RESULTS

The united radiomics model combining SWI, DWI, and CE-T1WI showed the highest diagnostic performance (area under the curve (AUC) = 0.95, accuracy = 0.93, specificity = 0.93) in the primary cohort, outperforming single-sequence and double-sequence models. The test set validated the model's good diagnostic performance (AUC = 0.9). ITSS grading significantly differed between adenolymphomas and malignant tumors (p < 0.001). ADC histogram analysis revealed significant differences in mean, 10th percentile, and kurtosis values between the two groups.

CONCLUSIONS

The multisequence radiomics model combining DWI, SWI, and CE-T1WI provides a comprehensive and accurate noninvasive approach for differentiating parotid adenolymphoma from malignant tumors. This method helps avoid the risks associated with invasive procedures, such as tumor cell implantation and metastasis, while guiding personalized surgical decision-making. By offering a novel diagnostic tool, this study enhances the precision of preoperative tumor characterization and supports more effective treatment planning and prognosis assessment for patients with parotid gland tumors.

Ultrasound-Based Deep Learning Radiomics Models for Predicting Primary and Secondary Salivary Gland Malignancies: A Multicenter Retrospective Study

BACKGROUND

Primary and secondary salivary gland malignancies differ significantly in treatment and prognosis. However, conventional ultrasonography often struggles to differentiate between these malignancies due to overlapping imaging features. We aimed to develop and evaluate noninvasive diagnostic models based on traditional ultrasound features, radiomics, and deep learning-independently or in combination-for distinguishing between primary and secondary salivary gland malignancies.

METHODS

This retrospective study included a total of 140 patients, comprising 68 with primary and 72 with secondary salivary gland malignancies, all pathologically confirmed, from four medical centers. Ultrasound features of salivary gland tumors were analyzed, and a radiomics model was established. Transfer learning with multiple pre-trained models was used to create deep learning (DL) models from which features were extracted and combined with radiomics features to construct a radiomics-deep learning (RadiomicsDL) model. A combined model was further developed by integrating ultrasound features. Least absolute shrinkage and selection operator (LASSO) regression and various machine learning algorithms were employed for feature selection and modeling. The optimal model was determined based on the area under the receiver operating characteristic curve (AUC), and interpretability was assessed using SHapley Additive exPlanations (SHAP).

RESULTS

The RadiomicsDL model, which combines radiomics and deep learning features using the Multi-Layer Perceptron (MLP), demonstrated the best performance on the test set with an AUC of 0.807. This surpassed the performances of the ultrasound (US), radiomics, DL, and combined models, which achieved AUCs of 0.421, 0.636, 0.763, and 0.711, respectively. SHAP analysis revealed that the radiomic feature Wavelet_LHH_glcm_SumEntropy contributed most significantly to the mode.

CONCLUSIONS

The RadiomicsDL model based on ultrasound images provides an efficient and non-invasive method to differentiate between primary and secondary salivary gland malignancies.

Nomogram combining dual-energy computed tomography features and radiomics for differentiating parotid warthin tumor from pleomorphic adenoma: a retrospective study

Introduction

Accurate differentiation between pleomorphic adenomas (PA) and Warthin tumors (WT) in the parotid gland is challenging owing to overlapping imaging features. This study aimed to evaluate a nomogram combining dual-energy computed tomography (DECT) quantitative parameters and radiomics to enhance diagnostic precision.

Methods

This retrospective study included 120 patients with pathologically confirmed PA or WT, randomly divided into training and test sets (7:3). DECT features, including tumor CT values from 70 keV virtual monochromatic images (VMIs), iodine concentration (IC), and normalized IC (NIC), were analyzed. Independent predictors were identified via logistic regression. Radiomic features were extracted from segmented regions of interest and filtered using the K-best and least absolute shrinkage and selection operator. Radiomic models based on 70 keV VMIs and material decomposition images were developed using logistic regression (LR), support vector machine (SVM), and random forest (RF). The best-performing radiomics model was combined with independent DECT predictors to construct a model and nomogram. Model performance was assessed using ROC curves, calibration curves, and decision curve analysis (DCA).

Results

IC (venous phase), NIC (arterial phase), and NIC (venous phase) were independent DECT predictors. The DECT feature model achieved AUCs of 0.842 and 0.853 in the training and test sets, respectively, outperforming the traditional radiomics model (AUCs 0.836 and 0.834, respectively). The DECT radiomics model using arterial phase water-based images with LR showed improved performance (AUCs 0.883 and 0.925). The combined model demonstrated the highest discrimination power, with AUCs of 0.910 and 0.947. The combined model outperformed the DECT features and conventional radiomics models, with AUCs of 0.910 and 0.947, respectively (P<0.05). While the difference in AUC between the combined model and the DECT radiomics model was not statistically significant (P>0.05), it showed higher specificity, accuracy, and precision. DCA found that the nomogram gave the greatest net therapeutic effect across a broad range of threshold probabilities.

Discussion

The nomogram combining DECT features and radiomics offers a promising non-invasive tool for differentiating PA and WT in clinical practice.

The Role of MRI Radiomics Using T2-Weighted Images and the Apparent Diffusion Coefficient Map for Discriminating Between Warthin's Tumors and Malignant Parotid Gland Tumors

BACKGROUND/OBJECTIVES

Differentiating between benign and malignant parotid gland tumors (PGT) is essential for establishing the treatment strategy, which is greatly influenced by the tumor's histology. The objective of this study was to evaluate the role of MRI-based radiomics in the differentiation between Warthin's tumors (WT) and malignant tumors (MT), two entities that proved to present overlapping imaging features on conventional and functional MRI sequences.

METHODS

In this retrospective study, a total of 106 PGT (66 WT, 40 MT) with confirmed histology were eligible for radiomic analysis, which were randomly split into a training group (79 PGT; 49 WT; 30 MT) and a testing group (27 PGT; 17 WT, 10 MT). The radiomic features were extracted from 3D segmentations of PGT performed on the following sequences: PROPELLER T2-weighted images and the ADC map, using a dedicated software. First- and second-order features were derived for each lesion, using original and filtered images.

RESULTS

After employing several feature reduction techniques, including LASSO regression, three final radiomic parameters were identified to be the most significant in distinguishing between the two studied groups, with fair AUC values that ranged between 0.703 and 0.767. All three radiomic features were used to construct a Radiomic Score that presented the highest diagnostic performance in distinguishing between WT and MT, achieving an AUC of 0.785 in the training set, and 0.741 in the testing set.

CONCLUSIONS

MRI-based radiomic features have the potential to serve as promising novel imaging biomarkers for discriminating between Warthin's tumors and malignant tumors in the parotid gland. Nevertheless, it is still to prove how radiomic features can consistently achieve higher diagnostic performance, and if they can outperform alternative imaging methods, ideally in larger, multicentric studies.

Radiomic and Clinical Model in the Prognostic Evaluation of Adenoid Cystic Carcinoma of the Head and Neck

BACKGROUND/OBJECTIVES

Adenoid Cystic Carcinoma (AdCC) is a rare malignant salivary gland tumor, with high rates of recurrence and distant metastasis. This study aims to stratify patients Relapse-Free Survival (RFS) using a combined model of clinical and radiomic features from preoperative MRI.

METHODS

This retrospective study included patients with primary AdCC who underwent surgery and adjuvant radiotherapy. Segmentations were manually performed by two head and neck radiologists. Radiomic features were extracted using the 3D Slicer software. Descriptive statistics was performed. A Survival Random Forest model was employed to select which radiological feature predict RFS. Cox proportional hazards models were constructed using clinical, radiological variables or both. Synthetic data augmentation was applied to address the small sample size and improve model robustness. Models were validated on real data and compared using the C-index and Prediction Error Curves (PEC).

RESULTS

Three Cox models were developed: one with clinical features (C-index = 0.67), one with radiomic features (C-index = 0.68), and one combining both (C-index = 0.77). The combined clinical-radiomic model had the highest predictive accuracy and outperformed models based on clinical or radiomic features. The combined model also exhibited the lowest mean Brier score in PEC analysis, indicating better predictive performance.

CONCLUSIONS

This study demonstrate that a combined radiomic-clinical model can predict RFS in AdCC patients. This model may provide clinicians a valuable tool in patient's management and may aid in personalized treatment planning.

The value of T1- and FST2-Weighted-based radiomics nomogram in differentiating pleomorphic adenoma and Warthin tumor

PURPOSE

To establish a radiomics nomogram based on MRI radiomics features combined with clinical characteristics for distinguishing pleomorphic adenoma (PA) from warthin tumor (WT).

METHODS

294 patients with PA (n = 159) and WT (n = 135) confirmed by histopathology were included in this study between July 2017 and June 2023. Clinical factors including clinical data and MRI features were analyzed to establish clinical model. 10 MRI radiomics features were extracted and selected from T1WI and FS-T2WI, used to establish radiomics model and calculate radiomics scores (Rad-scores). Clinical factors and Rad-scores were combined to serve as crucial parameters for combined model. Through Receiver operator characteristics (ROC) curve and decision curve analysis (DCA), the discriminative values of the three models were qualified and compared, the best-performing combined model was visualized in the form of a radiomics nomogram.

RESULTS

The combined model demonstrated excellent discriminative performance for PA and WT in the training set (AUC=0.998) and testing set (AUC=0.993) and performed better compared with the clinical model and radiomics model in the training set (AUC=0.996, 0.952) and testing model (AUC=0.954, 0.849). The DCA showed that the combined model provided more overall clinical usefulness in distinguishing parotid PA from WT than another two models.

CONCLUSION

An analytical radiomics nomogram based on MRI radiomics features, incorporating clinical factors, can effectively distinguish between PA and WT.

An ultrasound-based ensemble machine learning model for the preoperative classification of pleomorphic adenoma and Warthin tumor in the parotid gland

OBJECTIVES

The preoperative classification of pleomorphic adenomas (PMA) and Warthin tumors (WT) in the parotid gland plays an essential role in determining therapeutic strategies. This study aims to develop and validate an ultrasound-based ensemble machine learning (USEML) model, employing nonradiative and noninvasive features to differentiate PMA from WT.

METHODS

A total of 203 patients with histologically confirmed PMA or WT who underwent parotidectomy from two centers were enrolled. Clinical factors, ultrasound (US) features, and radiomic features were extracted to develop three types of machine learning model: clinical models, US models, and USEML models. The diagnostic performance of the USEML model, as well as that of physicians based on experience, was evaluated and validated using receiver operating characteristic (ROC) curves in internal and external validation cohorts. DeLong's test was used for comparisons of AUCs. SHAP values were also utilized to explain the classification model.

RESULTS

The USEML model achieved the highest AUC of 0.891 (95% CI, 0.774-0.961), surpassing the AUCs of both the US (0.847; 95% CI, 0.720-0.932) and clinical (0.814; 95% CI, 0.682-0.908) models. The USEML model also outperformed physicians in both internal and external validation datasets (both p < 0.05). The sensitivity, specificity, negative predictive value, and positive predictive value of the USEML model and physician experience were 89.3%/75.0%, 87.5%/54.2%, 87.5%/65.6%, and 89.3%/65.0%, respectively.

CONCLUSIONS

The USEML model, incorporating clinical factors, ultrasound factors, and radiomic features, demonstrated efficient performance in distinguishing PMA from WT in the parotid gland.

CLINICAL RELEVANCE STATEMENT

This study developed a machine learning model for preoperative diagnosis of pleomorphic adenoma and Warthin tumor in the parotid gland based on clinical, ultrasound, and radiomic features. Furthermore, it outperformed physicians in an external validation dataset, indicating its potential for clinical application.

KEY POINTS

• Differentiating pleomorphic adenoma (PMA) and Warthin tumor (WT) affects management decisions and is currently done by invasive biopsy. • Integration of US-radiomic, clinical, and ultrasound findings in a machine learning model results in improved diagnostic accuracy. • The ultrasound-based ensemble machine learning (USEML) model consistently outperforms physicians, suggesting its potential applicability in clinical settings.

Performance of radiomics in the differential diagnosis of parotid tumors: a systematic review

Purpose

A systematic review and meta-analysis were conducted to evaluate the diagnostic precision of radiomics in the differential diagnosis of parotid tumors, considering the increasing utilization of radiomics in tumor diagnosis. Although some researchers have attempted to apply radiomics in this context, there is ongoing debate regarding its accuracy.

Methods

Databases of PubMed, Cochrane, EMBASE, and Web of Science up to May 29, 2024 were systematically searched. The quality of included primary studies was assessed using the Radiomics Quality Score (RQS) checklist. The meta-analysis was performed utilizing a bivariate mixed-effects model.

Results

A total of 39 primary studies were incorporated. The machine learning model relying on MRI radiomics for diagnosis malignant tumors of the parotid gland, demonstrated a sensitivity of 0.80 [95% CI: 0.74, 0.86], SROC of 0.89 [95% CI: 0.27-0.99] in the validation set. The machine learning model based on MRI radiomics for diagnosis malignant tumors of the parotid gland, exhibited a sensitivity of 0.83[95% CI: 0.76, 0.88], SROC of 0.89 [95% CI: 0.17-1.00] in the validation set. The models also demonstrated high predictive accuracy for benign lesions.

Conclusion

There is great potential for radiomics-based models to improve the accuracy of diagnosing benign and malignant tumors of the parotid gland. To further enhance this potential, future studies should consider implementing standardized radiomics-based features, adopting more robust feature selection methods, and utilizing advanced model development tools. These measures can significantly improve the diagnostic accuracy of artificial intelligence algorithms in distinguishing between benign and malignant tumors of the parotid gland.

Systematic review registration

https://www.crd.york.ac.uk/prospero/, identifier CRD42023434931.

Differentiation of salivary gland tumours using diffusion-weighted image-based virtual MR elastography: a pilot study

OBJECTIVES

Differentiation among benign salivary gland tumours, Warthin tumours (WTs), and malignant salivary gland tumours is crucial to treatment planning and predicting patient prognosis. However, differentiation of those tumours using imaging findings remains difficult. This study evaluated the usefulness of elasticity determined from diffusion-weighted image (DWI)-based virtual MR elastography (MRE) compared with conventional magnetic resonance imaging (MRI) findings in differentiating the tumours.

METHODS

This study included 17 benign salivary gland tumours, 6 WTs, and 11 malignant salivary gland tumours scanned on neck MRI. The long and short diameters, T1 and T2 signal intensities, tumour margins, apparent diffusion coefficient (ADC) values, and elasticity from DWI-based virtual MRE of the tumours were evaluated. The interobserver agreement in measuring tumour elasticity and the receiver operating characteristic (ROC) curves were also assessed.

RESULTS

The long and short diameters and the T1 and T2 signal intensities showed no significant difference among the 3 tumour groups. Tumour margins and the mean ADC values showed significant differences among some tumour groups. The elasticity from virtual MRE showed significant differences among all 3 tumour groups and the interobserver agreement was excellent. The area under the ROC curves of the elasticity were higher than those of tumour margins and mean ADC values.

CONCLUSION

Elasticity values based on DWI-based virtual MRE of benign salivary gland tumours, WTs, and malignant salivary gland tumours were significantly different. The elasticity of WTs was the highest and that of benign tumours was the lowest. The elasticity from DWI-based virtual MRE may aid in the differential diagnosis of salivary gland tumours.

Differentiation of parotid pleomorphic adenoma from Warthin tumor using signal intensity ratios on fat-suppressed T2-weighted magnetic resonance imaging

OBJECTIVE

To investigate the value of magnetic resonance imaging (MRI) signal intensity ratios (SIRs) based on fat-suppressed T2-weighted imaging (FS-T2WI), together with demographic features, MRI anatomical characteristics, and SIRs of histopathological patterns of the tumors, in the differentiation of parotid pleomorphic adenoma (PA) from Warthin tumor (WT).

STUDY DESIGN

In total, 90 patients with PA and 56 patients with WT were enrolled in the study. SIRs of tumor to normal parotid gland (SIR-T/P), spinal cord (SIR-T/S), and muscle (SIR-T/M) were calculated. Demographic and radiological features of the 2-patient groups were compared with univariate analysis and multivariate logistic regression analysis. The area under the receiver operating characteristic curve (AUC), sensitivity, and specificity were analyzed to evaluate the utility of SIRs in distinguishing between PA and WT.

RESULTS

SIR-T/P exhibited outstanding discriminating ability (AUC = 0.934), SIR-T/S had excellent discrimination (AUC = 0.839), and SIR-T/M showed acceptable discrimination (AUC = 0.728). When SIR-T/P of 1.96 was selected as the cutoff value, sensitivity and specificity were 0.756 and 0.982, respectively. SIR-T/P, age, sex, and number of lesions were identified as independent predictors by multivariate logistic regression analysis. Differences in SIRs between histopathological patterns were significant.

CONCLUSION

SIR-T/P based on FS-T2WI is an effective discriminator in the differential diagnosis between PA and WT. Age, sex, and number of lesions provided additional value in differentiation.

Radiomics Analysis in Characterization of Salivary Gland Tumors on MRI: A Systematic Review

Radiomics analysis can potentially characterize salivary gland tumors (SGTs) on magnetic resonance imaging (MRI). The procedures for radiomics analysis were various, and no consistent performances were reported. This review evaluated the methodologies and performances of studies using radiomics analysis to characterize SGTs on MRI. We systematically reviewed studies published until July 2023, which employed radiomics analysis to characterize SGTs on MRI. In total, 14 of 98 studies were eligible. Each study examined 23-334 benign and 8-56 malignant SGTs. Least absolute shrinkage and selection operator (LASSO) was the most common feature selection method (in eight studies). Eleven studies confirmed the stability of selected features using cross-validation or bootstrap. Nine classifiers were used to build models that achieved area under the curves (AUCs) of 0.74 to 1.00 for characterizing benign and malignant SGTs and 0.80 to 0.96 for characterizing pleomorphic adenomas and Warthin's tumors. Performances were validated using cross-validation, internal, and external datasets in four, six, and two studies, respectively. No single feature consistently appeared in the final models across the studies. No standardized procedure was used for radiomics analysis in characterizing SGTs on MRIs, and various models were proposed. The need for a standard procedure for radiomics analysis is emphasized.