Abstract:Background: Parotid gland tumors (PGTs) are the most common benign tumors of salivary gland tumors. However, the diagnostic value of relative values of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and diffusion kurtosis imaging (DKI) parameters for PGTs has not been extensively studied. Therefore, this study aimed to evaluate the diagnostic performance of combined DKI and DCE-MRI for differentiating PGTs by introducing the concept of relative value. Methods: The DCE-MRI and DKI imaging data of 142 patients with PGTs between June 2018 and August 2022 were collected. Patients were divided into four groups by histopathology: malignant tumors (MTs), pleomorphic adenomas (PAs), Warthin tumors (WTs), and basal cell adenomas (BCAs). All MRI examinations were conducted using a 3 T MRI scanner with a 20-channel head and neck coil. Quantitative parameters of DCE-MRI and DKI and their relative values were determined. Kruskal-Wallis H test, post-hoc test with Bonferroni correction, one-way analysis of variance (ANOVA) and post-hoc test with least significant difference (LSD) method, and the receiver operating characteristic (ROC) curve were used for statistical analysis. Statistical significance was set at P<0.05. Results: Only the combination of DKI and DCE-MRI parameters could reliably distinguish BCAs from other PGTs. PAs demonstrated the lowest transfer constant from plasma to extravascular extracellular space (Ktrans) value [0.09 (0.06, 0.20) min-1], relative Ktrans (rKtrans) [-0.24 (-0.64, 1.00)], rate constant from extravascular extracellular space to plasma (Kep) value [0.32 (0.22, 0.53) min-1], relative Kep (rKep) [0.32 (0.22, 0.53) min-1], and initial area under curve (iAUC) value [0.15 (0.09, 0.26) mmol·s/kg] compared with WTs, BCAs, and MTs (all P<0.05). The Ktrans values for MTs were substantially lower [0.17 (0.10, 0.31) min-1] than those for WTs (P=0.01). The Kep values for MTs [0.71 (0.52, 1.28) min-1] were substantially lower (all P<0.05) than those for WTs and BCAs. PAs and BCAs had higher diffusion coefficient (D) values and lower diffusion kurtosis (K) values and relative K (rK) values than MTs and WTs. However, the D and K values did not differ significantly even in their relative values of PAs and BCAs (all P>0.05). By using logistic regression, the combination of K value and rKep value further enhanced their discriminatory power between PAs and WTs [area under the ROC curve (AUC), 0.986], the combination of K and rKep value further enhanced their discriminatory power between PAs and MTs (AUC, 0.915), and the combination of D and Kep value further enhanced their discriminatory power between BCAs and MTs (AUC, 0.909). Conclusions: DKI and DCE-MRI can be used to differentiate PGTs quantitatively and can complement each other. The combined use of DKI and DCE-MRI parameters can improve the diagnostic accuracy of distinguishing PGTs.

Imaging features contributing to the diagnosis of ameloblastomas and keratocystic odontogenic tumours: logistic regression analysis.

Texture analysis of diffusion kurtosis imaging for differentiating malignant from benign sinonasal lesions: added value to conventional imaging features

Application of Deep Learning in Differential Diagnosis of Ameloblastoma and Odontogenic Keratocyst Based on Panoramic Radiographs

Differential Diagnosis of OKC and SBC on Panoramic Radiographs: Leveraging Deep Learning Algorithms

[Imaging Analysis of Ameloblastoma, Odontogenic Keratocyst and Dentigerous Cyst in the Maxilla Using Spiral CT and Cone Beam CT].

Recent methods for the diagnosis and differentiation of ameloblastoma: a narrative review

Comparison of Radiographical Characteristics and Diagnostic Accuracy of Intraosseous Jaw Lesions on Panoramic Radiographs and CBCT

[Computed tomographic features of desmoplastic ameloblastoma of the jaw].

CT-based Radiomic Signatures for Predicting Histopathologic Features in Head and Neck Squamous Cell Carcinoma.

New perspectives in the differential diagnosis of jaw lesions: Machine learning and inflammatory biomarkers

Digital pathology-based artificial intelligence models for differential diagnosis and prognosis of sporadic odontogenic keratocysts

The impact of day care on socially disadvantaged families: an example of the use of process evaluation within a randomized controlled trial.

Structural requirements for thioester bond formation in human complement component C3. Reassessment of the role of thioester bond integrity on the conformation of C3.

CT-based radiomics analysis of different machine learning models for differentiating benign and malignant parotid tumors

The value of diffusion kurtosis imaging and dynamic contrast-enhanced magnetic resonance imaging in the differential diagnosis of parotid gland tumors

Automatic segmentation of ameloblastoma on ct images using deep learning with limited data

Premature Rupture of the Membranes: An Evidence-Based Approach to Clinical Care

Machine learning-based radiomics for predicting BRAF-V600E mutations in ameloblastoma

Endodontic Treatment Outcomes in Cone Beam Computed Tomography Images—Assessment of the Diagnostic Accuracy of AI

Radiomic biomarkers of locoregional recurrence: prognostic insights from oral cavity squamous cell carcinoma preoperative CT scans

Automated oral cancer identification using histopathological images: a hybrid feature extraction paradigm