Role of Radiomics in Parotid Malignant Disease: A Scoping Review

TL;DR

This review explores how radiomics, using MRI and other imaging, can help distinguish between benign and malignant parotid gland tumors, though more research is needed for clinical use.

Contribution

The paper provides a scoping review of recent radiomics applications in parotid gland malignancies, highlighting diagnostic potential and methodological challenges.

Findings

Radiomics models using MRI, CT, and PET/CT showed good diagnostic performance with AUC values between 0.769 and 0.952.

Methodological differences across studies prevented data pooling and limited result comparison.

Standardized protocols are needed to validate radiomics for clinical use in parotid gland tumor management.

Abstract

This review elucidates how radiomics is transforming the management of malignant neoplasms of the parotid gland: it explores the potential application of radiomics in distinguishing between benign and malignant parotid gland pathology. Radiomics is primarily based on magnetic resonance imaging. Whilst the results are encouraging, further validation is required before its adoption in clinical practice.

Malignant tumors of the salivary glands are rare, accounting for approximately 1–7% of all head and neck tumors. The parotid gland is the most commonly affected gland. An accurate preoperative diagnosis distinguishing between malignant and benign tumors is necessary for the appropriate management of patients. The aim of this review is to analyze the results of the most recent literature studying the use of radiomics in malignant tumors of the parotid gland. A comprehensive literature review was performed using the PubMed/MEDLINE, EMBASE and Cochrane Library databases, in accordance with the PRISMA review criteria (from 2020 to July 2025). The final analysis comprised a total of six articles and 560 patients. Four studies evaluated the role of Magnetic Resonance Imaging (MRI), one of Computed Tomography (CT) and one of Positron Emission Tomography/Computed Tomography (PET/CT). Radiomics models achieved good overall diagnostic performance, with AUC values ranging from 0.769 to 0.952 across studies, although methodological heterogeneity prevented data pooling. The results of this review indicate that radiomics has the potential to play an important role in the management of malignant tumors of the parotid gland. Nevertheless, the absence of clear and standardized protocols does not allow the comparison of results. Further studies are necessary to confirm these findings.

Figures