Table 4 Comparison of findings across previous studies.
From: CystNet: An AI driven model for PCOS detection using multilevel thresholding of ultrasound images
Authors | Year | Dataset | Feature extraction | Classification | Accuracy |
|---|---|---|---|---|---|
Nilofer et al.38 | 2021 | Kaggle PCOS US images | GLCM | ANN | 97.50% |
Gopalakrishnan et al.36 | 2021 | Kaggle PCOS US images | GIST-MDR | SVM | 93.82% |
Alamoudi et al.37 | 2023 | King Fahad Hospital US images | MobileNet | FC | 82.46% |
Rachana et al.19 | 2021 | Kaggle PCOS US images | Entropy, Contrast, Energy, Homogeneity | K-NN | 97% |
Nakhua et al.60 | 2024 | PCOSGen US images | GLCM | Hybrid ML | 92% |
Bedi et al.61 | 2023 | MMOTU images | Attention Residual Unit | AResUNet | 97% |
Paramasivam et al.62 | 2024 | Kaggle PCOS US images | SD_CNN | Hybrid ML | 96.43% |
Chitra et al.63 | 2023 | Kaggle PCOS US images | - | Hybrid DL | 95% |
Proposed Approach A | - | Kaggle PCOS US images | CystNet | FC | 96.54% |
Proposed Approach B | - | Kaggle PCOS US images | CystNet | RF | 97.75% |
Proposed Approach A | - | PCOSGen US images | CystNet | FC | 94.39% |
Proposed Approach B | - | PCOSGen US images | CystNet | RF | 96.12% |
Proposed Approach A | - | MMOTU images | CystNet | FC | 95.67% |
Proposed Approach B | - | MMOTU images | CystNet | RF | 97.23% |
