Table 20 Performance analysis of SOTA methods.
From: Feature fusion context attention gate UNet for detection of polycystic ovary syndrome
Model | Accuracy (%) |
|---|---|
AResUNet1 | 98.00 |
Enhanced U-Net + ResNet3 | 97.80 |
GAN + CNN4 | 96.00 |
PCOS-WaveConvNet6 | 97.90 |
PCONet23 | 98.12 |
VGGNet16 + Stacking Ensemble8 | 98.90 |
VGG16 (modified last 4 layers)10 | 92.11 |
ASPPNet + ResNet11 | 98.79 |
CNN + BiLSTM12 | 97.74 |
ESDPCOS (CNN + GLCM)13 | 96.06 |
AMCNN14 | 98.79 |
CNN + KNN clustering15 | 97.00 |
MLOD16 | 96.00 |
Ocys-Net17 | 95.93 |
HHO-DQN18 | 96.50 |
ITL-CNN19 | 98.90 |
Ensemble (VGG16, ResNet50, MobileNet)20 | 95.00 |
Watershed + contour analysis21 | 97.80 |
CR-UNet22 | 91.20 |
Hybrid CNN23 | 95.00 |
SqueezeNet26 | 97.63 |
InceptionV3 + TL28 | 98.48 |
2D CNN + SVM, DT, RF29 | 98.07 |
Sequential 2D CNN + wrapper FS31 | 98.67 |
Elman NN + Gabor Wavelet32 | 78.10 |
BPA (modified LM optimization)35 | 93.92 |
EfficientNetB6 + Attention UNet38 | 98.12 |
Threshold-based segmentation39 | 97.00 |
DLNNSVM40 | 97.32 |
GrabCut + FL-SNNM41 | 97.99 |
GIST-MDR42 | 93.82 |
QEI-SAM63 | 99.31 |
Deeplabv364 | 94.60 |
CystNet68 | 97.82 |
TL-CNN69 | 97.20 |
DC-UNet70 | 97.54 |
AdaResU-Net71 | 98.47 |
FCAU-Net | 99.89 |
