Skip to main content

Table 1 Demographics and baseline clinical characteristics of the study participants

From: Preliminary analysis of predicting the first recurrence in patients with neovascular age-related macular degeneration using deep learning

Variables

Training

Test

Total

p-value

Unique eyes after exclusion applied

1,295

149

1,444

 

Gender

 Female

527 (44.97)

59 (45.38)

586 (45.01)

1.00†

Age at first nAMD diagnosed (yrs)

71.95 ± 8.27

71.93 ± 8.31

71.95 ± 8.27

0.94‡

BCVA at baseline (logMAR)

0.70 ± 0.49

0.76 ± 0.51

0.71 ± 0.49

0.31‡

BCVA after loading phase (logMAR)

0.54 ± 0.48

0.58 ± 0.47

0.55 ± 0.48

0.09‡

Axial length (mm)a

23.54 ± 0.90

23.62 ± 1.03

23.54 ± 0.91

 

Oculus

 Oculus dexter

647 (49.96)

76 (51.01)

723 (50.07)

0.88†

nAMD subtype

   

0.26†

 Type 1 or 2 CNV

969 (74.83)

115 (77.18)

1,084(75.07)

 

 PCV

208 (16.06)

17 (11.41)

225 (15.58)

 

 RAP (Type 3 CNV)

118 (9.11)

17 (11.41)

135 (9.35)

 

Anti-VEGF used for loading phase

   

0.55†

 Ranibizumab

585 (45.17)

74 (49.66)

659 (45.64)

 

 Aflibercept

576 (44.48)

62 (41.61)

638 (44.18)

 

 Bevacizumab

134 (10.35)

13 (8.72)

147 (10.18)

 

Duration until the first recurrence (months)

7.87 ± 14.38

11.04 ± 23.28

8.20 ± 15.56

0.93‡

OCT system

   

0.84†

 Cirrus

798 (61.62)

90 (60.40)

888 (61.50)

 

 Spectralis

497 (38.38)

59 (39.60)

556 (38.50)

 
  1. Continuous variables are reported as mean value ± standard deviation. All other data are numbers (percentages)
  2. nAMD neovascular age-related macular degeneration, yrs years, BCVA best-corrected visual acuity, logMAR logarithm of the minimum angle of resolution, anti-VEGF anti-vascular endothelial growth factor, CNV choroidal neovascularization, PCV polypoidal choroidal vasculopathy, RAP retinal angiomatous proliferation, OCT optical coherence tomography
  3. †Chi-Square p-value
  4. ‡Kruskal–Wallis p-value
  5. aOnly available in 498 eyes in the training set and 62 eyes in the test set