Table 3 Processing time of DL algorithms
From: Machine learning applications for early detection of esophageal cancer: a systematic review
Algorithm | Time | Modality | Type |
|---|---|---|---|
Faster R-CNN | 74 s | - | segmentation |
Faster R-CNN | 5.3 s | - | Detection |
U-Net | 4.24 s | CT | segmentation |
CNN | 113 frames/s | ultrasound endoscopy | Prediction |
Cascade RCNN | 42 frames/s | ultrasound endoscopy | Prediction |
U-Net | 46.3 s | WLI | Detection |
R-CNN | 13.38–37.81 s | WLI | Detection |
Fast R-CNN | 0.65–2.1 s | WLI | Detection |
Faster R-CNN | 0.3–0.45 s | WLI | Detection |
SSD | 0.1–0.2 s | WLI | Detection |
SSD | 1.0 s | WLI&NBI | Detection |