308
Q. Chen et al.
procedure, Li [65] presented a new formulation that forces the level set function to
be close to a signed distance function. The formulation proposed is:
E(φ) μP(φ) + E m (φ),
(11.3)
where
P(φ)
1
2
Ω
(|∇φ| − 1)
2 dxdy.
(11.4)
For full detail of our approach see [58].
11.3.1.5 Qualitative Evaluation
The first dataset consisted of 55 longitudinal SD-OCT cube scans from twelve eyes
in eight patients with GA (acquired with the CirrusOCT device, Carl Zeiss Meditec,
Inc., Dublin, CA.). Each cube consisted of 512 × 128 × 1024 voxels corresponding
to a 6 × 6 × 2 mm
3 volume centered at the macular region of the retina in the
lateral, azimuthal and axial directions, respectively). The second dataset consisted
in 56 SD-OCT cube scans from 56 eyes in 56 patients with GA (acquired with the
CirrusOCT device). Each cube consisted of 200 × 200 × 1024 voxels corresponding
to a 6 × 6 × 2 mm
3 volume centered at the macular region of the retina in the lateral,
azimuthal and axial directions, respectively). Figures 11.17 and 11.18 show the GA
segmentation results for two different patients with GA from the first dataset who
had multiple SD-OCT studies over the course of their disease.
11.3.1.6 Quantitative Evaluation: Our Method Versus Expert Graders
We evaluated the variability observed in the GA segmentations by different graders
(inter-observer agreement) and by the same grader at different sessions (intraobserver agreement) in the SD-OCT RSVP images using the first scan dataset.
Figure 11.16 indicates that the contrast for the visualization of GA in RSVP images
is better than that in SVP images. The results from this evaluation are summarized
in Table 11.5, where A 1 represents the segmentations of the first grader in the first
session, A 2 is first grader in the second session, B 1 is the second grader in the first
session and B 2 is the second grader in the second session.
11.3.1.7 Conclusions
This section presents a semi-automated segmentation algorithm for GA in SD-OCT
images. A projection image constructed from a sub-volume of the retina beneath the
RPE which shows the GA abnormalities most clearly appears to improve the visu-
Q. Chen et al.
procedure, Li [65] presented a new formulation that forces the level set function to
be close to a signed distance function. The formulation proposed is:
E(φ) μP(φ) + E m (φ),
(11.3)
where
P(φ)
1
2
Ω
(|∇φ| − 1)
2 dxdy.
(11.4)
For full detail of our approach see [58].
11.3.1.5 Qualitative Evaluation
The first dataset consisted of 55 longitudinal SD-OCT cube scans from twelve eyes
in eight patients with GA (acquired with the CirrusOCT device, Carl Zeiss Meditec,
Inc., Dublin, CA.). Each cube consisted of 512 × 128 × 1024 voxels corresponding
to a 6 × 6 × 2 mm
3 volume centered at the macular region of the retina in the
lateral, azimuthal and axial directions, respectively). The second dataset consisted
in 56 SD-OCT cube scans from 56 eyes in 56 patients with GA (acquired with the
CirrusOCT device). Each cube consisted of 200 × 200 × 1024 voxels corresponding
to a 6 × 6 × 2 mm
3 volume centered at the macular region of the retina in the lateral,
azimuthal and axial directions, respectively). Figures 11.17 and 11.18 show the GA
segmentation results for two different patients with GA from the first dataset who
had multiple SD-OCT studies over the course of their disease.
11.3.1.6 Quantitative Evaluation: Our Method Versus Expert Graders
We evaluated the variability observed in the GA segmentations by different graders
(inter-observer agreement) and by the same grader at different sessions (intraobserver agreement) in the SD-OCT RSVP images using the first scan dataset.
Figure 11.16 indicates that the contrast for the visualization of GA in RSVP images
is better than that in SVP images. The results from this evaluation are summarized
in Table 11.5, where A 1 represents the segmentations of the first grader in the first
session, A 2 is first grader in the second session, B 1 is the second grader in the first
session and B 2 is the second grader in the second session.
11.3.1.7 Conclusions
This section presents a semi-automated segmentation algorithm for GA in SD-OCT
images. A projection image constructed from a sub-volume of the retina beneath the
RPE which shows the GA abnormalities most clearly appears to improve the visu-