Table 2 Linear mixed model predicting anomaly score.

From: Feasibility of anomaly score detected with deep learning in irradiated breast cancer patients with reconstruction

Variables

Coefficient

95% CI

P-value

Days after RT

0.004

 

0.007

Type of reconstruction

 Autologous

0.000

(base)

 

 Tissue expander

2.223

(0.431–4.016)

0.015

Type x Days after baseline

0.000

(−0.001–0.002)

0.563

RT fractionation

 Conventional

0.000

(base)

 

 Hypofractionated

0.485

(−2.149 to 3.120)

0.718

Fractionation x Days after RT

−0.001

(−0.004 to 0.002)

0.389

Mastectomy

 Total/Radical

0.000

(base)

 

 NSM/SSM

−0.309

(−1.556 to 0.938)

0.627

RT to SCL

 No

0.000

(base)

 

 Yes

1.094

(−0.412 to 2.600)

0.154

RT to IMN

 No

0.000

(base)

 

 Yes

−0.229

(−2.460 to 2.002)

0.840

Boost

 No

0.000

(base)

 

 Yes

0.909

(−0.792 to 2.610)

0.295

RT plan

 3D

0.000

(base)

 

 IMRT

−0.186

(−2.338 to 1.966)

0.865

Major complication

 No

0.000

(base)

 

 Yes

0.077

(−1.647 to 1.802)

0.930

Age (year)

0.073

(−0.008 to 0.153)

0.078

BMI (kg/m2)

 ≤23

0.000

(base)

 

 >23

0.570

(−0.660 to 1.800)

0.364

Intercept

−2.390

(−6.701 to 1.920)

0.277

  1. P-value by a linear mixed-effect model.
  2. CI Confidence interval, NSM Nipple-sparing mastectomy, SSM Skin-sparing mastectomy, SCL Subclavian lymph node, IMN Internal mammary node, 3D 3-dimensional, IMRT Intensity modulated radiation therapy, BMI Body mass index.