Application of artificial intelligence-driven three-dimensional imaging in preoperative planning for rectosigmoid colon cancer

Table 1 General patient information

Clinical factors	n (%)
Gender
Male	23 (59)
Female	16 (41)
Total	39
Age, median (range)	62 years (40-71 years)
Rectum	30 (76.9)
Sigmoid colon	9 (23.1)
Differentiation
High	0 (0)
Medium	29 (74.4)
Low-medium	10 (25.6)
Low	0 (0)
T1	3 (7.7)
T2	4 (10.2)
T3	29 (74.4)
T4	3 (7.7)
N0	23 (59)
N1	13 (33.3)
N2	3 (7.7)
N3	0 (0)
Tumor budding
High	13 (33.3)
Medium	8 (20.5)
Low	18 (46.2)
Vascular thrombus
Yes	16 (41)
No	23 (59)
Neural invasion
Yes	20 (51.3)
No	19 (48.7)
Intraoperative stoma
Yes	14 (35.9)
No	25 (64.1)
Preserve the anus
Yes	38 (97.4)
No	1 (2.6)
Postoperative adjuvant therapy
Yes	22 (56.4)
No	17 (43.6)

Table 2 Analysis of differences in tumor burden and the distance from the tumor’s lowest border to the anal verge measurements by different methods

	MD	SD	t	P value	ICC (95%CI)
Maximum tumor diameter (cm)
Pathology-AI-3D	-0.602	0.510	-7.367	< 0.001	0.921
AI-3D-CT	-0.106	1.400	-0.471	0.640	0.482
Pathology-CT	-0.708	1.327	-3.332	0.002	0.518
Maximum tumor cross-sectional area (cm2)
Pathology-AI-3D	-0.150	4.031	-0.233	0.817	0.846
AI-3D-CT	5.430	5.076	6.680	< 0.001	0.517
Pathology-CT	5.280	6.898	4.780	< 0.001	0.407
DTAV (cm)
AI-3D-colonoscopy	2.079	6.415	2.024	0.050	0.907

DTAV: Distance from the tumor’s lowest border to the anal verge; AI-3D: Artificial intelligence-driven digital three-dimensional imaging; ICC: Intraclass correlation coefficients; CT: Computed tomography; CI: Confidence interval.

Table 3 Assessment of perirectal lymph node invasion by artificial intelligence-driven digital three-dimensional imaging and computed tomography

	AI-3D (%)	CT (%)
Sensitivity	80.0	60.0
Specificity	62.5	29.2
Positive predictive value	57.1	34.6
Negative predictive value	83.3	53.8

AI-3D: Artificial intelligence-driven digital three-dimensional imaging; CT: Computed tomography.