Understanding and Comparing Confocal Scanning Lasers to Optical Coherence Tomography for Optic Nerve Head Analysis (ONH) A presentation courtesy of Zeiss

1
Understanding and Comparing Confocal Scanning
Lasers to Optical Coherence Tomography for Optic
Nerve Head Analysis (ONH)A presentation
courtesy of Zeiss
2
Third Generation Products

• LDT abandoned TopSS in 1999 to continue
  develop-ment of GDx Access
• HDT released their HRT II instrument, which is
  smaller, faster, easier and cheaper

3
How they Work

• A 670-780nm LASER light source plus (2) Pinhole
  (20-25 micron) apertures detect light reflected
  off the retina

4
Signal

• Confocal Scanning Lasers sense the shape of the
  surface of the Hyloid Face or the Vitreous to
  Internal Limiting Membrane interface.
• At this point the peak reflection is recorded
• Difficulties with some common conditions as PVD
  can cause artifacts.
• Have excellent X-Y resolution, but coarse Z
  resolution.

5
Scanning Lasers make Slices

• Typically 32 Slices are acquired over a 4mm
  depth.
• Each slice has a depth of 25 microns.
• A non sampled gap of 100 microns is between
  slices.

Z
Y
X
6
Scanning

• Typically a 6mm by 6mm area of retinal surface is
  scanned.
• Each slice is about 20 25 microns thick.
• Each gap between slices is about 100 microns
  thick.
• Only a few slices fall on the ONH.

7
Examples of Data

• Two of the 32 slices are displayed at the right.
• Lets plot the paths of light of the 3
  data-points.
• At different areas on the disk in slice 14 of 32
  notice
• Yellow is on a blood Vessel
• Green is on the rim
• Purple is in the cup

Slice 14 of 32
Slice 18 of 32
8
How the slices determine topography

• A graph illustrates REFLECTANCE for the same
  location at each of the 32 slices. Since the
  Yellow curve peaked before the Green or Purple
  curve, a RELATIVE difference in Topography can be
  calculated.

9
Establishing a Reference Plane

• A Publication by Weinreb et al shows the
  Papilomacular bundle as an area adjacent to the
  rim showing the least amount of change. So these
  instruments use this area and other areas as a
  peripheral band concentric around the disk to
  manufacture ZERO or a reference plane.

10
Volumetric Analysis

• The operator must draw the outline of the disk.
• This is not easy as the image is not taken in
  white light.
• Notice how two different operators outline the
  same disk with different shapes.
• Volumetric parameters are operator dependent.

11
Optical Coherence Tomography

• To scan the ONH, Optical Coherence Tomography
  acquires (6) 4mm long by 2mm deep Slices in the
  X-Z axis.
• Datapoints are sampled every 2 microns providing
  excellent resolution in the Z axis.
• Confocal Scanning Lasers typically sample every
  100 microns.

Y
X
Z
12
Image Analysis

• Using a proprietary algorithm, OCT identifies the
  Retina Pigment Epithelium and places a marker.

13

• The markers show the boundaries of the disk in
  each individual slice. The next slide shows all
  six slices.

14
ONH Tomography Examples
Slice 1
Slice 4
Slice 2
Slice 5
Slice 3
Slice 6
15
ONH Analysis

• Using the algorithm OCT objectively finds the
  margin of the disk.
• OCT objectively finds the cup by using an offset
  value of 150 microns up from the RPE.
• No reference plane is required.

16
OCT Printout

• Details volumetric analysis of cup and Disk.
• Provides Direct Cross Section of anatomy.

17
Z Dimension Datapoints
Optical Coherence Tomography acquire 1024
datapoints over 2mm
Confocal Scanning Lasers acquire 32 datapoints
over 4mm
18
Datapoint Comparrison

• Confocal Scanning Lasers have 100 micron gaps
  between datapoints in Z dimension.
• Optical Coherence Tomography has only 2 micron
  gap.

100 Micron Gap
150 Microns
19
Conclusion

• Confocal Scanning Laser Technology is not capable
  of seeing small changes in the ONH.
• Glaucomatous change occurs very slowly.
• Progression may not be observed with sampling
  points with 100 micron gaps.
• OCT with higher Z resolution is more likely to
  see glaucomatous progression.
• Objective placement of Cup and Disk insures
  accuracy between operators.
• Viewing the anatomy confirms ONH Analysis from
  artifacts seen in confocal scanning lasers.