Skip to main content
. 2022 Aug 21;46(10):62. doi: 10.1007/s10916-022-01850-y

Table 8.

Benchmarking table

- C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13
R# Author # Patients # Images Image Dim #M Model Types Solo vs. HDL Dim AE DS JI BA ACC
R1 Paluru et al. [80] 69  ~ 4339 5122 1 AnamNet Solo 2D  ✖  ✔  ✖  ✖  ✔
R2 Saood and Hatem [85] -  ~ 100 2562 2 UNet, SegNet Solo 2D  ✖  ✔  ✖  ✖  ✔
R3 Cai et al. [86] 99  ~ 250 - 1 UNet Solo 2D  ✖  ✔  ✔  ✖  ✖
R4 Suri et al. [40] 72  ~ 5000 7682 4

NIH,

SegNet,

VGG-SegNet,

ResNet-SegNet

 Both 2D  ✔  ✔  ✔  ✔  ✔
R5 Suri et al. [39] 72  ~ 5000 7682 3

PSPNet,

VGG-SegNet,

ResNet-SegNet

 Both 2D  ✔  ✔  ✔  ✔  ✔
R6 Suri et al. [38] 79  ~ 5500 7682 2

VGG-SegNet,

ResNet-SegNet

 HDL 2D  ✔  ✔ ✔   ✔  ✔
R7 Suri et al. (Proposed) 152  > 10,000 5122 9

PSPNet,

SegNet,

UNet,

VGG-PSPNet, VGG-SegNet,

VGG-UNet,

ResNet-PSPNet,

ResNet-SegNet,

ResNet-UNet

 Both 2D  ✔  ✔  ✔  ✔  ✔

# number, HDL Hybrid Deep Learning, AE Area Error, DS Dice Similarity, JI Jaccard Index, BA Bland–Altman, ACC Accuracy, Dim Dimension (2D vs. 3D), R# Row number, #M number of AI models