Clustering-Based Multiparametric MRI for Differentiation Between True Tumor Progression and Pseudoprogression in Glioblastoma

Motivation: Clustering analysis in brain tumor aims to improve differentiation between true progression (TP) and pseudoprogression (PsP) in glioblastoma using multiparametric MRI-based pharmacokinetic and diffusion parameters, addressing a critical diagnostic challenge to enhance treatment assessment and planning. Goal(s): The objective is to distinguish between true progression (TP) from pseudoprogression (PsP) in glioblastoma using multiparametric MRI-based clustering analysis. Approach: K-means ++ Clustering based Multiparametric MRI data analysis for segregating the tumor into low and high intensity regions. Results: Mean Ktrans and Mean Kep are statistically significant DCE-MRI parameters for differentiating true progression from pseudoprogression of glioblastoma. Impact: These results demonstrate the importance of vascular permeability in tumor assessment and establish Ktrans and Kep along with tumor volume as essential parameters for differentiating true progression from treatment effects.