Skip to main content

Table 3 Performance Metrics and Parameters of LASSO-based Logistic Regression Model

From: Development of a risk prediction model for radiation dermatitis following proton radiotherapy in head and neck cancer using ensemble machine learning

AUC

ACC

R2

Hosmer-Lemeshow

Omnibus

0.870

73.7%

0.383

0.727

< 0.001

Feature

\(\beta\)

p-value

OR (CI)

S.E.

Wald

Gender

1.427

0.178

4.166 (0.53–33.17)

1.06

1.82

Smoking

2.670

0.023

14.446 (1.47–143.30)

1.17

5.20

N

1.143

0.241

3.137 (0.46–21.23)

0.98

1.37

AJCC

2.335

0.065

10.325 (0.86–123.73)

1.27

3.39

V60− 5 mm

0.116

0.194

1.123 (0.94–1.34)

0.09

1.68

D100− 3 mm

-0.073

0.517

0.929 (0.745–1.160)

0.11

0.42

Constant

-2.223

    
  1. Abbreviations β: Coefficient; OR: Odds Ratio; CI: Confidence Interval; S.E.: Standard Error; AJCC: American Joint Committee on Cancer; N: Lymph Node Metastasis; Vx: Volume receiving x Gy dose; Dx: Dose received by x cc volume; ACC: Accuracy