Table 1.

Performance results of the AI-CDSS in the internal cross validation and the test data set: IoU and Dice Score for all categories as well as their means across all categories, pixel accuracy for complete frames and 95% CI in brackets

Internal cross validation
	Vessel detection	Tissue differentiation			Instrument detection
	Vessel	Submucosa	Muscularis	Background	Instrument	Knife	Mean
Dice Score	55.15 (54.10 to 56.18)	75.51 (74.88 to 76.12)	70.64 (69.32 to 71.88)	86.49 (85.99 to 86.99)	88.69 (87.57 to 89.83)	80.60 (79.61 to 81.49)	76.18 (75.73 to 76.57)
IoU	38.07 (37.08 to 39.07)	60.65 (59.85 to 61.44)	54.60 (53.05 to 56.10)	76.19 (75.43 to 76.98)	79.68 (77.89 to 81.54)	67.51 (66.13 to 68.77)	62.78 (62.18 to 63.31)
Pixel accuracy	80.99 (80.52 to 81.47)
Test
Dice Score	62.77 (60.08 to 65.12)	80.71 (79.50 to 81.82)	72.48 (69.40 to 74.99)	91.39 (90.45 to 92.10)	89.69 (87.09 to 91.96)	83.50 (82.06 to 84.87)	80.09 (79.14 to 80.92)
IoU	45.74 (42.94 to 48.28)	67.65 (65.97 to 69.24)	56.84 (53.14 to 59.99)	84.14 (82.56 to 85.36)	81.30 (77.14 to 85.11)	71.67 (69.58 to 73.72)	67.89 (66.61 to 69.04)
Pixel accuracy	86.89 (85.86 to 87.70)

AI-CDSS, artificial intelligence clinical decision support solution.