Skip to main content

Table 2 Agreement and correlation between automated and manual segmented choroidal thickness

From: Comparison of manual and artificial intelligence-automated choroidal thickness segmentation of optical coherence tomography imaging in myopic adults

Subjects

n

LOA

Proportion of outliers (n [%])

Fixed bias

95% CI

P value

Proportional bias (β)

P value

ICCa

95% CI

P value

r

P value

Lower bound LOA

(95% CI)

Upper bound LOA

(95% CI)

PLEX Elite 9000

All subjects

456

 − 47.787 (− 52.134 to − 43.440)

60.475 (56.128 to 64.822)

20 (4.4)

6.344

3.802 to 8.886

 < 0.001

0.120

 < 0.001

0.937

0.922 to 0.949

 < 0.001

0.946

 < 0.001

AL < 26 mm

188

 − 47.171 (− 54.530 to − 39.813)

69.932 (62.574 to 77.291)

9 (4.8)

11.380

7.083 to 15.679

 < 0.001

0.098

 < 0.001

0.921

0.878 to 0.947

 < 0.001

0.934

 < 0.001

AL ≥ 26 mm

268

 − 46.943 (− 52.168 to − 41.719)

52.565 (47.341 to 57.789)

8 (3.0)

2.811

 − 0.242 to 5.864

0.071

0.134

 < 0.001

0.931

0.913 to 0.946

 < 0.001

0.940

 < 0.001

ChT < 300 µm

371

 − 46.524 (− 50.955 to − 42.940)

52.923 (48.493 to 57.353)

13 (3.5)

3.199

0.610 to 5.789

0.016

0.158

 < 0.001

0.890

0.866 to 0.909

 < 0.001

0.901

 < 0.001

ChT ≥ 300 µm

85

 − 43.815 (− 55.877 to − 31.753)

83.952 (71.890 to 96.014)

6 (7.1)

20.068

13.038 to 27.099

 < 0.001

 − 0.011

0.881

0.749

0.492 to 0.864

 < 0.001

0.803

 < 0.001

Triton DRI-OCT

All subjects

456

 − 90.556 (− 95.683 to − 85.429)

37.127 (32.000 to 42.254)

30 (6.6)

 − 26.714

 − 29.711 to − 23.716

 < 0.001

 − 0.090

 < 0.001

0.887

0.608 to 0.950

 < 0.001

0.933

 < 0.001

AL < 26 mm

188

 − 105.076 (− 114.522 to − 95.629)

45.258 (35.811 to 54.704)

11 (5.9)

 − 29.908

 − 35.426 to − 24.391

 < 0.001

 − 0.108

0.001

0.852

0.555 to 0.931

 < 0.001

0.907

 < 0.001

AL ≥ 26 mm

268

 − 78.709 (− 84.403 to − 73.014)

29.762 (24.067 to 35.456)

14 (5.2)

 − 24.473

 − 27.801 to − 21.145

 < 0.001

 − 0.077

 < 0.001

0.884

0.545 to 0952

 < 0.001

0.934

 < 0.001

ChT < 300 µm

375

 − 77.249 (− 82.020 to − 72.479)

30.419 (25.648 to 35.189)

17 (4.5)

 − 23.415

 − 26.204 to − 20.626

 < 0.001

 − 0.011

0.666

0.830

0.453 to 0.924

 < 0.001

0.894

 < 0.001

ChT ≥ 300 µm

81

 − 134.339 (− 152.218 to − 116.461)

50.362 (32.484 to 68.241)

6 (7.4)

 − 41.987

 − 52.406 to − 31.569

 < 0.001

 − 0.328

 < 0.001

0.561

0.090 to 0.778

 < 0.001

0.723

 < 0.001

  1. P values in bold indicate statistical significance
  2. n = number of eyes; LOA = limits of agreement; β = unstandardized coefficient; r = Pearson correlation coefficient; ICC = intraclass correlation coefficient; AL = axial length; ChT = choroidal thickness
  3. aICC was calculated based on two-way mixed effects absolute agreement model
Back to article page

Eye and Vision

ISSN: 2326-0254