Meningioma grading using conventional MRI histogram analysis based on 3D tumor measurement

Area of Science:

• Neuroradiology
• Oncology
• Medical Imaging Analysis

Background:

• Meningiomas are the most common primary intracranial tumors.
• Accurate grading of meningiomas is crucial for treatment planning and prognosis.
• Conventional MRI histogram analysis offers a quantitative approach to tumor characterization.

Purpose of the Study:

• To evaluate the efficacy of conventional MRI histogram analysis using whole tumor measurements for assessing meningioma grading.
• To determine the predictive value of histogram parameters in differentiating high-grade meningioma (HGM) from low-grade meningioma (LGM).

Main Methods:

• Retrospective analysis of preoperative MRI scans from 90 patients with meningioma (45 grade I, 38 grade II, 7 grade III).
• Regions of Interest (ROIs) were drawn on T1WI, T2WI, and contrast-enhanced T1WI to generate 3D signal intensity histograms.
• Statistical analysis included t-tests, Kruskal-Wallis, univariate and multivariate logistic regression, and Spearman's correlation to identify predictive parameters and optimal classification models.

Main Results:

• Significant differences in histogram parameters were observed between HGM and LGM groups.
• Volume count and uniformity were identified as high-predictive parameters for distinguishing HGM from LGM.
• A logistic regression model incorporating contrast-enhanced T1WI histogram parameters achieved an AUC of 0.834 for grading, with a sensitivity of 83.9% and specificity of 77.4%.

Conclusions:

• Conventional MRI histogram analysis, based on 3D tumor measurements, is a valuable tool for clinical meningioma grading.
• This quantitative imaging technique can aid in differentiating high-grade from low-grade meningiomas, supporting treatment decisions.