Table 2 The Dataset type and performance of ML algorithms
From: Machine learning applications for early detection of esophageal cancer: a systematic review
Modality | Algorithm | Dataset | Accuracy |
|---|---|---|---|
CT | U-Net | 100 | - |
CT | V-Net | 414 | 65% |
CT | VGG16 | 457 | 84.20% |
WLI | SSD | 100 | - |
WLI | VGG16 | 805 | 93.43% |
WLI | FRCNN | 1051 | 96.29% |
WLI | CNN | 1272 | 85.83% |
WLI | ResNet-Unet | 1704 | 89% |
WLI | YOLO v5 | 4447 | 92.90% |
WLI | Neuro-T | 5162 | 95.60% |
WLI | SSD | 1780 | 96.10% |
NBI | SegNet | 6473 | 96.50% |
WLI & NBI | SSD | 8428 | 98% |
WLI & NBI | SSD | 155 | 84% |
WLI & NBI | SSD | 498 | 90.90% |
WLI & NBI | U-Net | 165 | 84.72% |