Assessment of muscularis propria invasion in urothelial bladder carcinoma using continuous‑time random walk and stretched‑exponential models: Comparison with VI‑RADS and apparent diffusion coefficient

Accurate assessment of muscle layer invasion in bladder cancer is required to distinguish muscle‑invasive bladder cancer (MIBC) from non‑MIBC (NMIBC). The present prospective study evaluated continuous‑time random‑walk (CTRW) and stretched‑exponential model (SEM) diffusion parameters against conventional apparent diffusion coefficient (ADC) and vesical imaging‑reporting and data system (VI‑RADS) scores. A total of 106 patients (62 MIBC and 44 NMIBC) who underwent 3.0T multiparametric magnetic resonance imaging, including multi‑b‑value diffusion‑weighted imaging, at The Second Affiliated Hospital of Kunming Medical University (Kunming, China) were analyzed. CTRW and SEM parameters were extracted, and VI‑RADS scores were assigned. Significant differences were found between groups for VI‑RADS ≥4, SEM‑ADC, CTRW‑α and CTRW‑β (P<0.05). Multivariate logistic regression identified VI‑RADS ≥4 and CTRW‑β as independent predictors of muscle invasion. While VI‑RADS alone achieved an area under the curve (AUC) of 0.85, the combined model incorporating SEM‑ADC, CTRW‑α and VI‑RADS achieved the highest overall performance (AUC=0.91; 95% CI: 0.86‑0.96), significantly outperforming VI‑RADS alone (raw P=0.002, Bonferroni‑corrected P=0.004). Additionally, diffusion kurtosis was positively correlated with invasion depth (r=0.62). These findings indicate that advanced diffusion parameters, particularly CTRW‑α and SEM‑ADC, effectively differentiate MIBC from NMIBC. When integrated with VI‑RADS, they significantly enhance diagnostic accuracy compared with VI‑RADS alone (AUC: 0.91 vs. 0.85; P=0.030). Decision curve analysis confirmed clinical utility, with the combined model providing a mean net benefit gain of 0.10 (95% CI: 0.06‑0.14) across the clinically relevant threshold range, corresponding with 10 additional correct diagnoses per 100 patients. These quantitative diffusion metrics represent promising non‑invasive biomarkers for guiding therapeutic decisions.