Skip to main content
. Author manuscript; available in PMC: 2020 Sep 18.
Published in final edited form as: IEEE Access. 2019 May 13;7:64583–64591. doi: 10.1109/access.2019.2916557

TABLE VII.

Effect of 2D/3D CNN architecture on the performance of models trained from scratch: p-values of pairwise compared ROCS between the 3D CNN model group and two 2D CNN model groups

Scratch Fine-tune
2D slice image transverse plane 0.712 0.766
2D slice image multi-view plane 0.733 0.838
3D volume image[19] 0.809 N/A