Skip to main content

Table 1 Baseline clinical characteristics of the study population

From: Develop and validate a radiomics space-time model to predict the pathological complete response in patients undergoing neoadjuvant treatment of rectal cancer: an artificial intelligence model study based on machine learning

Characteristics

Study dataset (n = 83)

ALL patients

pCR

(n = 22)

No-pCR

(n = 61)

P

Age (years, SD)

63.83 (13.64)

61.76 (9.21)

64.54 (13.64)

0.21

Sex (N, %)

Male

60 (72.3)

15 (81.8)

45 (73.8)

0.208

Female

23 (27.7)

7 (18.2)

16 (26.2)

CEA (N, %)

Abnormal

48

(57.8)

15 (68.2)

33 (54.1)

0.251

Normal

35

(42.2)

7 (31.8)

28 (45.9)

CA199 (N, %)

Abnormal

22 (26.5)

9 (36.4)

13 (21.3)

0.235

Normal

59 (73.5)

13 (59.1)

46 (75.4)

DIS (cm, SD)

5.02 (2.28)

6.03 (1.92)

4.62 (2.28)

0.012*

CRM status (N, %)

Positive

50 (60.2)

12 (54.5)

38 (62.3)

0.524

Negative

33 (39.8)

10 (45.5)

23 (37.7)

mrEMVI status (N, %)

Positive

36 (22.2)

8 (36.4)

28 (45.9)

0.474

Negative

47 (77.8)

14 (63.6)

33 (54.1)

Tumor stage (N, %)

T1 − 2

9 (10.8)

2 (90.9)

7 (11.5)

0.736

T3 − 4

74 (89.2)

20 (90.9)

54 (88.5)

Lymph node (N, %)

N0

11 (13.3)

3 (13.6)

8 (13.1)

0.289

N1 − 2

72 (86.7)

19 (86.4)

53 (86.9)

  1. Note: CEA, carcinoembryonic antigen; CA199, carbohydrate antigen 199; DIS, the distance from the end of the convex edge of the tumor to the edge of the anus; CRM, circumferential resection margin; mrEMVI, MRI-based extramural vascular invasion. Data are presented as counts or means (standard deviations in parentheses)