MRI-based Quantification of Perfusion and Microstructure

Sample size and power considerations

Aim 3. Diagnostic performance of IVIM parameters

By Table 4 of Ren et al. (2024), the mean \(\pm\) SD of Dfast in F0-F1 and F2-F4 are 63.2 \(\pm\) 9.2 and 51.8 \(\pm\) 8.2, respectively. Those of F in F0-F1 and F2-F4 are 0.4 \(\pm\) 0.07 and 0.32 \(\pm\) 0.06, respectively. Assuming equal number of patients in the two groups, the total numbers \(N\) needed to detect AUC > 0.5 (AUC=0.5 for no association) at significance level 0.05 with power 80% and 90% are tabulated below.

Power	N (Dfast)	N (F)
80%	20	23
90%	26	31

Aim 2. Repeatability of of IVIM parameters

Simchick et al. (2022) reported intra-class correlation coefficients (ICC) for repeatability of V_b and F as 0.80 and 0.94, respectively. Using the method of Zou (2012), the power for detecting ICC > 0.6¹, with each measurement repeated twice, at significance level 0.05 is plotted below.

The power for detecting ICC > 0.57, with each measurement repeated twice, at significance level 0.05 is plotted below.

Conclusion

With 40-50 patients, the tests for AUC (Aim 3) and ICC (Aim 2) are well powered.

References

Ren, Hao, Hui Xu, Dawei Yang, Xiaofei Tong, Xinyan Zhao, Qianyi Wang, Yameng Sun, et al. 2024. “Intravoxel Incoherent Motion Assessment of Liver Fibrosis Staging in MASLD.” Abdominal Radiology, March. https://doi.org/10.1007/s00261-024-04207-w.

Simchick, Gregory, Ruiqi Geng, Yuxin Zhang, and Diego Hernando. 2022. “B Value and First-Order Motion Moment Optimized Data Acquisition for Repeatable Quantitative Intravoxel Incoherent Motion DWI.” Magnetic Resonance in Medicine 87 (6): 2724–40. https://doi.org/10.1002/mrm.29165.

Zou, G. Y. 2012. “Sample Size Formulas for Estimating Intraclass Correlation Coefficients with Precision and Assurance.” Statistics in Medicine 31 (29): 3972–81. https://doi.org/10.1002/sim.5466.

Footnotes

1. Can change this threshold