Abstract
BACKGROUND: To investigate the value of virtual magnetic resonance elastography (vMRE), fractional order calculus (FROC) model, and diffusion-weighted imaging (DWI) in differentiating rectal cancer grades. METHODS: This prospective study included 74 patients with rectal cancer who underwent a pelvic MRI. The Mann-Whitney U test or independent samples t-test was employed to compare the parameters of vMRE (µ(MRE)), the FROC model (D, β, and µ), and DWI (ADC). Logistic regression analysis and area under the receiver operating characteristic curve (AUC) were utilized separately for multiparameter co-diagnosis and to assess diagnostic performance. Pearson's correlation analysis was conducted to evaluate the relationship of different parameters. RESULTS: Compared to the low-grade group, the high-grade group had higher µ(MRE) and µ values and lower D, β, and ADC values (all P < 0.05). In terms of the different parameters, the D value demonstrated the highest diagnostic efficacy with an AUC of 0.852(95% CI: 0.750-0.924). In terms of the various methods, the FROC model (D + β + µ) had the highest diagnostic performance with an AUC of 0.943(95% CI: 0.864-0.984), which was significantly higher than those of DWI and vMRE (Z = 3.586, 2.430, and 2.081, all P < 0.05). µ(MRE) showed moderately negative correlations with ADC, D, and β (r = - 0.553, - 0.683, and - 0.530, respectively, all P < 0.05) and a moderately positive correlation with µ (r = 0.443, P < 0.05). CONCLUSION: FROC, vMRE, and DWI can be utilized for assessing rectal cancer grades, with FROC offering comparatively better diagnostic performance.