The value of an apparent diffusion coefficient histogram model in predicting meningioma recurrence

Area of Science:

• Neuroradiology
• Oncology
• Medical Imaging Analysis

Background:

• Meningiomas are common primary brain tumors.
• Predicting meningioma recurrence is crucial for effective patient management.
• Conventional MRI features alone have limitations in predicting recurrence.

Purpose of the Study:

• To develop and validate a predictive model for meningioma recurrence.
• To combine conventional MRI features with apparent diffusion coefficient (ADC) histogram parameters.
• To assess the model's predictive efficacy and clinical utility.

Main Methods:

• Retrospective analysis of 72 meningioma patients.
• Extraction of conventional MRI features and ADC histogram parameters using MaZda software.
• Development of a nomogram using logistic regression analysis.
• Validation using calibration curves, decision curve analysis, and ROC curves.

Main Results:

• Four independent risk factors for recurrence identified: enhancement uniformity, age, Simpson grade, and ADC first percentile (ADCp1).
• The combined model demonstrated high predictive efficacy (AUC 0.965).
• Excellent accuracy (90.3%), sensitivity (92.6%), and specificity (88.9%) were achieved.

Conclusions:

• A combined model of MRI features and ADC histogram parameters reliably predicts meningioma recurrence.
• This model offers a valuable tool for guiding treatment decisions and personalizing patient care.
• The findings support the clinical applicability of this predictive model in neuro-oncology.