Texture Analysis of Magnetic Resonance Imaging: Differentiating Multicystic Ameloblastomas From Central Vascular Malformations

ABSTRACT Purpose The aim of this study was to distinguish multicystic ameloblastomas (MABs) from central vascular malformations (CVMs) using magnetic resonance imaging (MRI) texture analysis. Materials and Methods We analyzed data from participants who underwent MRI between April 2006 and November 2021 at our university hospital's outpatient department. Participants were divided into two groups based on the lesion type. MRI texture features were the predictor variables. The outcome variable was the lesion type: MAB or CVM. Covariates included the demographic variables of age and sex. The Mann–Whitney U test was used to analyze bivariate statistics. Receiver operating characteristic curve analysis was used to assess the ability of MRI texture features to distinguish between lesions. Statistical significance was set at p < 0.05. Results The MRI data of 27 participants (mean age ± standard deviation, 43.44 ± 17.49 years [range, 9–71], 18 men) were analyzed (22 had MAB, and 5 had CVM). The six MRI texture features differed significantly between MAB and CVM ( p < 0.05). Areas under the curve values of MRI texture features ranged from 0.84 to 0.96. Conclusion These results suggest that MABs and CVMs may be distinguished by analyzing MRI texture features, thus helping clinicians in decision‐making while selecting appropriate treatment.