Voxelwise Prediction of Recurrent High-Grade Glioma via Proximity Estimation-Coupled Multidimensional Support Vector Machine

Abstract

Purpose: To provide early and localized glioblastoma (GBM) recurrence prediction, we introduce a novel postsurgery multiparametric magnetic resonance-based support vector machine (SVM) method coupling with stem cell niche (SCN) proximity estimation.

Methods and materials: This study used postsurgery magnetic resonance imaging (MRI) scans from 50 patients with recurrent GBM, obtained approximately 2 months before clinically diagnosed recurrence. The main prediction pipeline consisted of a proximity-based estimator to identify regions with high risk of recurrence (HRRs) and an SVM classifier to provide voxelwise prediction in HRRs. The HRRs were estimated using the weighted sum of inverse distances to 2 possible origins of recurrence-the SCN and the tumor cavity. Subsequently, multiparametric voxels (from T1, T1 contrast-enhanced, fluid-attenuated inversion recovery, T2, and apparent diffusion coefficient) within the HRR were grouped into recurrent (warped from the clinical diagnosis) and nonrecurrent subregions and fed into the proximity estimation-coupled SVM classifier (SVM_PE). The cohort was randomly divided into 40% and 60% for training and testing, respectively. The trained SVM_PE was then extrapolated to an earlier time point for earlier recurrence prediction. As an exploratory analysis, the SVM_PE predictive cluster sizes and the image intensities from the 5 magnetic resonance sequences were compared across time to assess the progressive subclinical traces.

Results: On 2-month prerecurrence MRI scans from 30 test cohort patients, the SVM_PE classifier achieved a recall of 0.80, a precision of 0.69, an F1-score of 0.73, and a mean boundary distance of 7.49 mm. Exploratory analysis at early time points showed spatially consistent but significantly smaller subclinical clusters and significantly increased T1 contrast-enhanced and apparent diffusion coefficient values over time.

Conclusions: We demonstrated a novel voxelwise early prediction method, SVM_PE, for GBM recurrence based on clinical follow-up MR scans. The SVM_PE is promising in localizing subclinical traces of recurrence 2 months ahead of clinical diagnosis and may be used to guide more effective personalized early salvage therapy.