Table 3 Logistic regression analysis on the results of clustering, and CNV scoring system.

From: Deep phenotype unsupervised machine learning revealed the significance of pachychoroid features in etiology and visual prognosis of age-related macular degeneration

 

OR (95% CI)*

P value**

Score

Intercept

–

–

− 7.04

Age (every 10 years)

1.18 (0.79–1.76)

0.42

0.16

Sex (male: 0, female: 1)

1.67 (0.81–3.50)

0.17

0.52

CVH in either eye (+: 1, −: 0)

7.35 (3.23–17.89)

4.39 × 10–6

1.99

Retinal thickness in CNV-affected eye (every 100 μm)

0.71 (0.59–0.85)

2.07 × 10–4

− 0.34

Choroidal thickness in CNV-affected eye (every 100 μm)

3.91 (2.35–6.78)

4.25 × 10–7

1.37

Choroidal thickness in fellow eye (every 100 μm)

3.45 (2.08–5.90)

3.12 × 10–6

1.24

Type A drusen in either eye (+: 1, −: 0)

0.19 (0.059–0.53)

2.50 × 10–3

− 1.68

  1. CVH choroidal vascular hyperpermeability, CNV choroidal neovascularization, OR odds ratio, CI confidence interval, PNV pachychoroid neovasculopathy, AMD age related macular degeneration.
  2. *Odds ratio of PNV compared with AMD.
  3. **P value was derived using logistic regression analysis on the clustering result.
Back to article page