Differentiation of Benign From Malignant Parotid Gland Tumors Using Conventional MRI Based on Radiomics Nomogram

Abstract

ObjectivesWe aimed to develop and validate radiomic nomograms to allow preoperative differentiation between benign- and malignant parotid gland tumors (BPGT and MPGT, respectively), as well as between pleomorphic adenomas (PAs) and Warthin tumors (WTs).Materials and MethodsThis retrospective study enrolled 183 parotid gland tumors (68 PAs, 62 WTs, and 53 MPGTs) and divided them into training (n = 128) and testing (n = 55) cohorts. In total, 2553 radiomics features were extracted from fat-saturated T2-weighted images, apparent diffusion coefficient maps, and contrast-enhanced T1-weighted images to construct single-, double-, and multi-sequence combined radiomics models, respectively. The radiomics score (Rad-score) was calculated using the best radiomics model and clinical features to develop the radiomics nomogram. The receiver operating characteristic curve and area under the curve (AUC) were used to assess these models, and their performances were compared using DeLong’s test. Calibration curves and decision curve analysis were used to assess the clinical usefulness of these models.ResultsThe multi-sequence combined radiomics model exhibited better differentiation performance (BPGT vs. MPGT, AUC=0.863; PA vs. MPGT, AUC=0.929; WT vs. MPGT, AUC=0.825; PA vs. WT, AUC=0.927) than the single- and double sequence radiomics models. The nomogram based on the multi-sequence combined radiomics model and clinical features attained an improved classification performance (BPGT vs. MPGT, AUC=0.907; PA vs. MPGT, AUC=0.961; WT vs. MPGT, AUC=0.879; PA vs. WT, AUC=0.967).ConclusionsRadiomics nomogram yielded excellent diagnostic performance in differentiating BPGT from MPGT, PA from MPGT, and PA from WT.

Histogram Analysis Based on Neurite Orientation Dispersion and Density MR Imaging for Differentiation Between Glioblastoma Multiforme and Solitary Brain Metastasis and Comparison of the Diagnostic Performance of Two ROI Placements

“…In both ROIs, the diagnostic performance of multivariate logistic regression model was superior to optimal single parameter. Multivariate logistic regression model, combined with multiple parameters, can provide comprehensive information related to tumor heterogeneity, which has better application value for the differential diagnosis of GBM and SBM 40 . Furthermore, this study showed that ROI placement can affect diagnostic performance.…”

“…Multivariate logistic regression model, combined with multiple parameters, can provide comprehensive information related to tumor heterogeneity, which has better application value for the differential diagnosis of GBM and SBM. 40 Furthermore, this study showed that ROI placement can affect diagnostic performance. For the whole tumor area, the performance of multivariate logistic regression model had better than that for the peritumoral edema area.…”

Construction and Application of a Nomogram for Predicting Benign and Malignant Parotid Tumors

“…The judgment criteria for MR image features 6,7,9 were as follows: (1) tumor morphology (a round or quasi-round tumor was considered regular, whereas a lobulated and bulging tumor was considered irregular); (2) tumor boundaries (the tumor boundaries were clear if the tumor had smooth edges and an envelope, whereas the tumor boundaries were not clear if the tumor had rough and blurry edges); (3) signal uniformity of the tumor (the tumor signals were considered uniform if the tumor signals were homogeneous on T1WI and T2WI, and the uniformity was not affected by cystic changes, bleeding, or calcifications; otherwise, the tumor signals were heterogeneous); (4) tumor location (for a tumor diameter <2 cm, the facial nerve line was used to separate the parotid gland into deep and superficial lobes; for a tumor diameter >2 cm, the U line was used, and if the U line passing through the tumor divided the tumor into 2 approximately equal parts, the tumor was considered to be located in both the deep and superficial lobes); (5) whether or not the tumor protruded outside the parotid gland (the tumor was considered to protrude outside the parotid gland if the maximum tumor diameter exceeded the parotid gland in the transverse view; otherwise, the tumor was considered not to protrude); and (6) ADC value of the tumor (the ADC value of the tumor was measured on the ADC graph, using the T2-weighted coronal image as a reference). Specifically, 3 regions of interest were chosen away from the sites of bleeding, cystic changes, and necrosis, each measuring approximately 10 mm 2 .…”

“…Recently reported nomograms that have been built based on CT and ultrasound radiomics not only clearly diagnose the types of parotid tumors, but also effectively differentiate malignant from benign parotid tumors. [6][7][8] There are few reports on MRI radiomics-based nomograms. [6][7][8] Moreover, each nomogram is built based solely on patient information from the researcher's institution.…”

“…[6][7][8] There are few reports on MRI radiomics-based nomograms. [6][7][8] Moreover, each nomogram is built based solely on patient information from the researcher's institution. To deepen the clinical application of MRI-based nomogram in differentiating malignant from benign parotid tumors, the present study retrospectively determined the preoperative prediction value of MRI-based nomogram and recruited patients to validate the prediction value.…”