Table 3 Characteristics of NCR cases in both training and validation sets.

From: A nomogram for predicting non-curative resection in patients with early gastric cancer based on white light imaging

Variables

NCR of training set (n = 70)

n(%)

NCR of validation set (n = 21)

n(%)

p value

Tumor size (mm)

 ≤ 20

42 (60.0%)

5 (23.8%)

< 0.05

 20–30

14 (20.0%)

7 (33.3%)

 > 30

14 (20.0%)

9 (42.9%)

Depth of invasion

 T1a (Mucosa)

24 (34.3%)

11 (52.4%)

> 0.05

 T1b1 (SM1, < 500 μm)

20 (28.6%)

3 (14.3%)

 T1b2 (SM2, ≥ 500 μm)

26 (37.1%)

7 (33.3%)

Histology

 Differentiated

25 (35.7%)

3 (14.3%)

< 0.05

 Undifferentiated

45 (64.3%)

18 (85.7%)

 Positive lymphovascular invasion (LVI+)

6 (8.6%)

1 (4.8%)

> 0.05

 Positive vertical margin (VM1)

21 (30.0%)

5 (23.8%)

> 0.05

 Ulceration

22 (31.4%)

6 (28.6%)

> 0.05

  1. NCR non-curative resection, SM1 submucosal invasion < 500 μm, SM2 submucosal invasion ≥ 500 μm.