Table 3.
Interobserver agreement for interpretation of endocytoscopic BE images for unassisted and AI-assisted gastroenterologists at various testing moments.
| Test Set 1 | Test Set 2 | |||
|---|---|---|---|---|
| Before Training (Test Set 1A) |
After Training (Test Set 1B) |
Follow-Up | ||
| Kappa values (95% CI) |
Unassisted gastroenterologists | 0.491 (0.437–0.545) |
0.562 (0.501–0.623) |
0.526 (0.469–0.583) |
| AI-assisted gastroenterologists | 0.597 (0.537–0.657) |
0.687 (0.622–0.752) |
0.696 (0.631–0.761) |
|
Interpretation of the interobserver kappa values according to the thresholds of Landis and Koch: scores less than 0.00: poor; 0.00 to 0.20: slight; 0.21 to 0.40: fair; 0.41 to 0.60: moderate; 0.61 to 0.80: substantial; 0.81 to 1.00: almost perfect [28].