Table 1 Quantitative description of categorical features of subjects at the time of enrollment.

From: Evaluating machine learning classifiers for glaucoma referral decision support in primary care settings

Features

Categories

Total (N = 3015)

Glaucoma Count (%)

Non-glaucoma Count (%)

Gender

Male

1353

167 (12.3%)

1186 (87.7%)

Female

1662

170 (10.2%)

1492 (89.8%)

Race

White

2913

315 (10.8%)

2598 (89.2%)

Black

84

18 (21.4%)

66 (78.6%)

Hispanic

9

1 (11.1%)

8 (88.9%)

Asian

4

1 (25%)

3 (75%)

Other

5

2 (40%)

3 (60%)

Diabetes

Positive

239

32 (13.4%)

207 (86.6%)

Negative

2776

305 (11%)

2471 (89%)

Arthritis

Positive

1354

157 (11.6%)

1197 (88.4%)

Negative

1661

180 (10.8%)

1481 (89.2%)

AMD*

Category 1

746

90 (12.1%)

656 (87.9%)

Category 2

673

65 (9.7%)

608 (90.3%)

Category 3

1054

119 (11.3%)

935 (88.7%)

Category 4

542

63 (11.6%)

479 (88.4%)

  1. *AMD category descriptions33.
  2. Category 1: A few small or no drusen.
  3. Category 2: Many small drusen or a few medium-sized drusen in one or both eyes.
  4. Category 3: Many medium-sized drusen or one or more large drusen in one or both eyes.
  5. Category 4: Breakdown of light-sensitive cells and supporting tissue in the central retinal.
  6. Area or abnormal and fragile blood vessels under the retina.
Back to article page