Skip to main content
. 2024 Nov 13;14(12):8758–8770. doi: 10.21037/qims-24-836

Table 2. Univariate and multivariate analysis of MR feature in training set.

Variable Univariate analysis Multivariate analysis
OR 95% CI P value OR 95% CI P value Regression coefficient
Hepatic steatosis 0.995 0.561–1.765 0.987
Intralesional fat 0.602 0.196–1.849 0.375
Morphology 8.902 4.916–16.121 <0.001 4.155 2.023–8.536 <0.001 1.425
DWI 1.201 0.520–2.773 0.668
ADC 1.160 0.598–2.248 0.658
T1WI 1.793 1.060–3.031 0.028
T2WI 1.798 1.0526–3.072 0.030
Heterogeneous 6.410 2.630–15.622 <0.001
Enhancing capsule 7.637 3.322–17.556 <0.001
Central scar 8.552 4.696–15.574 <0.001 2.838 1.284–6.273 0.010 1.043
Feeding artery 11.182 5.258–23.782 <0.001 7.083 2.970–16.892 <0.001 1.958
Radiating septa 7.218 4.055–12.846 <0.001 3.747 1.682–8.347 0.001 1.321
Enhancement pattern
   Hyper/hyper/hyper 1
   Hyper/hyper-iso/iso 2.097 1.222–3.598 0.007 3.539 1.650–7.595 0.001 1.164
   Hyper/hypo/hypo 0.999

MR, magnetic resonance; OR, odds ratio; CI, confidence interval; DWI, diffusion-weighted imaging; ADC, apparent diffusion coefficient; T1WI, T1-weighted imaging; T2WI, T2-weighted imaging; hyper, hyperintense; iso, isointense; hypo, hypointense.