Table 1 Baseline characteristics of participants according to median mammogram risk score

From: Investigating the relationship between breast cancer risk factors and an AI-generated mammographic texture feature in the Nurses’ Health Study II

Characteristic

All

MRS < the median level

MRS ≥ the median level

P

N = 853

N = 426

N = 427

Age, years

55.3 (5.45)

56.4 (5.21)

54.3 (5.48)

<0.001

Early life body size

2.56 (1.16)

2.64 (1.21)

2.49 (1.10)

0.059

Body mass index, kg/m2

26.8 (6.06)

27.4 (6.45)

26.2 (5.57)

0.003

Waist-to-hip ratio

0.83 (0.07)

0.83 (0.07)

0.82 (0.07)

0.035

Waist-to-hip ratio adjusted for BMI

0.00 (0.07)

0.00 (0.07)

0.00 (0.07)

0.648

Height, inch

64.9 (2.52)

64.9 (2.48)

65.0 (2.55)

0.523

Age at menarche, years

13.2 (2.95)

13.2 (2.98)

13.2 (2.92)

0.913

Age at first birth, years

26.3 (4.93)

26.0 (4.71)

26.7 (5.12)

0.075

Menopausal status

   

<0.001

Premenopausal

233 (27.8%)

86 (20.6%)

147 (34.8%)

 

Postmenopausal

538 (64.1%)

302 (72.4%)

236 (55.9%)

 

Unsure

68 (8.10%)

29 (6.95%)

39 (9.24%)

 

Age at natural menopause, years

49.7 (4.14)

49.5 (4.37)

50.0 (3.80)

0.094

Postmenopausal hormone use

   

0.108

Premenopausal or postmenopausal not on therapy

626 (77.4%)

322 (79.3%)

304 (75.4%)

 

Postmenopausal on therapy

183 (22.6%)

84 (20.7%)

99 (24.6%)

 

No. pregnancies ≥6 months

1.94 (1.24)

1.96 (1.25)

1.92 (1.22)

0.664

History of benign breast disease

   

0.013

Yes

525 (61.5%)

244 (57.3%)

281 (65.8%)

 

No

328 (38.5%)

182 (42.7%)

146 (34.2%)

 

Predicted BI-RADS densitya

   

<0.001

a, almost entirely fatty

40 (5.10%)

33 (8.44%)

7 (1.78%)

 

b, scattered areas of fibroglandular tissue

320 (40.8%)

209 (53.5%)

111 (28.2%)

 

c, heterogeneously dense

396 (50.4%)

140 (35.8%)

256 (65.0%)

 

d, extremely dense

29 (3.69%)

9 (2.30%)

20 (5.08%)

 

Family history of breast cancer

   

0.14

Yes

160 (18.8%)

71 (16.7%)

89 (20.8%)

 

No

693 (81.2%)

355 (83.3%)

338 (79.2%)

 

Breast cancer cases

   

<0.001

Yes

292 (34.2%)

88 (20.7%)

204 (47.8%)

 

No

561 (65.8%)

338 (79.3%)

223 (52.2%)

 

Mammogram risk score

0.00 (1.00)

−0.84 (0.58)

0.84 (0.49)

<0.001

  1. aPredicted BI-RADS density was assessed using a deep learning algorithm previously developed to predict mammographic breast density from digital mammograms. The algorithm categorizes breasts from a (almost entirely fatty) to d (extremely dense), matching an experienced mammographer’s evaluation.
Back to article page