Transcript RTVue Overview Slides.pps

Fourier Domain OCT:
The RTVue
Michael J. Sinai, PhD
Director of Clinical Affairs
Optovue, Inc.
Rise of Structural Assessment with
Scanning Lasers
• Scanning lasers provide objective and quantitative
information for numerous ocular pathologies
• First appeared over 20 years ago as a research tool
• Today, structural assessment with retinal imaging
devices has become an indispensable tool for
clinicians
Role of imaging in clinical practice
• AAO preferred practice patterns recommends using scanning
laser imaging in routine clinical exams
• In glaucoma, studies show imaging results can be as good as
expert grading of high quality stereo-photographs1
• Pre-perimetric glaucoma is now commonly accepted
• In OHTS, most converted based on structural assessment
only (not fields) 2
• OHTS has shown that imaging results have a high positive and
negative predictive power for detecting glaucoma 3
1.
2.
3.
Wollstein et al. Ophthalmology 2000
Kass et al. Arch Ophthalmol 2001
Zangwill LM, Weinreb RN, et al. Archives of Ophthalmol. 2005.
3 Imaging technologies have been
shown to be effective in detecting and
managing ocular pathologies
Light
• Scanning Laser Polarimetry (SLP)
• Confocal Scanning Laser
Ophthalmoscopy (CSLO)
• Optical Coherence Tomography
(OCT)
Polarizer
Two polarized components
Birefringent structure
(RNFL)
Retardation
SLP – GDx VCC
Strengths
• Provides RNFL thickness
• Large database
• Easy to use/interpret (deviation map/automated classifier)
• Progression
Weaknesses
• Atypical Pattern Birefringence (RNFL artifact)1
• Converts retardation to thickness assuming uniform birefringence (not true) 2
• Only RNFL information (No Optic Disc info and no Retina info)
• Data not backwards compatible
Normal
Glaucoma
1. Bagga, Greenfield, Feuer. AJO, 2005: 139: 437.
2. Huang, Bagga, Greenfield, Knighton IOVS, 2004: 45: 3037.
Atypical
CSLO – HRT 3
Strengths
• Provides Optic Disc morphology
• Sophisticated Progression Analysis
• Large ethnic Specific Database comparisons
• Automated classifier
• Data backwards compatible
• Some retinal capabilities
• Cornea microscope attachment
Weaknesses
• Only Optic Disc assessment (poor RNFL)
• Manual Contour Line drawing
• Reference plane based on surface height (can change)
• Retina analysis confined to edema detection and sensitive to image quality
• Cornea scans very difficult and impractical
OCT – Time Domain
(Stratus from CZM and SLO/OCT from OTI)
Strengths
• Provides Cross Sectional images
• Useful to calculate RNFL thickness
• Cross section scans useful for retinal pathologies
• Database comparisons
Weaknesses
• Slow scan speed (400 A scans / second)
• Limited data for glaucoma, 768 pixel (A-scan) ring for RNFL
• Limited data for retina, 6 radial lines with 128 A scans (pixels) each
• Macula maps 97% interpolated
• No progression analysis
• Location of scan ring affects RNFL results
• Prone to motion artifacts because of slow scan speed
• Poor optic disc measurements
Time Domain OCT susceptible to
eye movements
• 768 pixels (A-scans)
captured in 1.92 seconds is
slower than eye movements
• Stabilizing the retina
reveals true scan path (white
circles)1
1. Koozekanani, Boyer and Roberts. “Tracking the Optic Nervehead in OCT Video Using Dual Eigenspaces
and an Adaptive Vascular Distribution Model”; IEEE Transactions on Medical Imaging, Vol. 22, No. 12, 2003
Scan location and eye movements
affects results
Properly centered
T
S
N
I
Poorly centered: too inferior Poorly centered: too superior
T
Normal Double Hump
T
S
N
I
T
Inferior RNFL “Loss”
T
S
N
I
T
Superior RNFL “Loss”
Time Domain OCT artifacts can be common
Percent Error in Stratus OCT Scans
100
90
92
80
70
60
50
40
42
43
Bartsch et al.
2004
Ray et al. 2005
30
20
10
13.5
0
Sadda et al.
2006
1.
2.
3.
(severe error)
Sadda, Wu, et al. Ophthalmology 2006;113:285-293
Ray, Stinnett, Jaffe . Am J Ophth 2005; 139:18-29
Bartsch, Gong, et al. Proc. of SPIE Vol. 5370; 2140-2151
The Future of OCT
• RTVue Fourier Domain OCT overcomes limitations of
Time Domain OCT Devices
–
–
–
–
–
–
–
Better resolution (5 micron VS 10 micron)
Faster scan speeds (26,000 A scans / sec VS 400)
3-D data sets (won’t miss pathology)
Large data maps (less interpolation)
Progression capabilities
Layer by layer assessment
Versatility (Anterior Chamber Imaging)
Retina
Glaucoma
Anterior Chamber
The Evolution of OCT Technology
40,000
RTVue
2006
26,000
20,000
Speed
(A-scans
per sec)
400
Time domain OCT
Zeiss OCT 1
and 2, 1996
100
16
Fourier domain OCT
•
•
~ 65 x faster
~ 2 x resolution
Zeiss Stratus
2002
10
Depth Resolution
7
5
(mm)
Comparison of OCT Images
OCT 1 / 2
(Time Domain)
1996
Stratus OCT
(Time Domain)
2002
RTVue
(Fourier Domain)
2006
Case 1: AMD
Stratus
(Time Domain)
RTVue
(Fourier Domain)
Drusen not visible in Stratus Time Domain OCT
Case 2: DME
Stratus
(Time Domain)
RTVue
(Fourier Domain)
Case 3: PED
Stratus
(Time Domain)
RTVue
(Fourier Domain)
Same eye, PED missed by Stratus
Case 4: Macula Hole
Stratus
(Time Domain)
RTVue
(Fourier Domain)
Time Domain OCT vs Fourier Domain OCT
Time Domain
Fourier Domain
• Entire A scan generated at once
• A-scan generated sequentially based on Fourier transform of
one pixel at a time in depth
spectrometer analysis
• Moving reference mirror
• Stationary reference mirror
• 400 A scans per second
• 26,000 A scans per second
• 10 micron depth resolution
• 5 micron depth resolution
• B scan (512 A scans) in 1.28 sec • B scan (1024 A-scans) in 0.04 sec
• Slower than eye movements
• Faster than eye movements
Summary of Fourier Domain OCT
Advantages
• High speed reduces eye motion artifacts present
in time domain OCT
• High resolution provides precise detail, allows
more structures to visualized
• Layer by layer assessment
• Larger scanning areas allow data rich maps &
accurate registration for change analysis
• 3-D scanning improves clinical utility
RTVue Clinical Applications
Retina
Glaucoma
Anterior
Chamber
Retina Analysis with the RTVue: Line Scans
Line Scan
• Data Captured: 1024 A scans
(pixels)
• Time: 39 msec
• Area covered: 6 mm line
(adjustable 2-12 mm)
Provides
•High resolution B scan
•Image averaging
increases S/N
Cross Line Scan
• Data Captured:
2048 A scans (pixels)
• Time: 78 msec
• Area covered: 2 x 6
mm lines (adjustable
2-12 mm)
Provides
• vertical and
horizontal high
resolution B scan
•Image averaging
increases S/N
Line Scan: Cystoid Macula Edema
Courtesy: Michael Turano, CRA
Columbia University.
Courtesy: Michael Turano, CRA
Columbia University.
Retina Analysis with the RTVue: 3-D Scans
Provides
•3 D map
• Comprehensive assessment
• Fly through review
• C scan view
• SLO OCT image simultaneously
captured
• Data Captured: 51,712 A scans (pixels)
• Time: 2 seconds
• Area covered: 4 x 4 X 2 mm (adjustable)
• 101 B scans each 512 A scans
3-D view reveals extent of
damage over large area
Top Image: En face view of retinal surface from 3-D scan
Bottom Image: B scan from corresponding location (green line)
Retina Analysis with the RTVue: Macula
Maps (MM5)
Provides:
• Layer specific thickness maps
• Detailed B scans
• ETDRS thickness grid
• Data Captured: 19,496 A scans (pixels)
• Time: 750 msec
• Area covered: 5 mm x 5 mm (grid pattern)
Full retinal
thickness
Inner retinal
thickness
Outer retinal
thickness
RPE/Choroid
Elevation
Surface
Topography
ILM to RPE
ILM to IPL
IPL to RPE
RPE height
ILM height
Glaucoma Analysis with the RTVue: Nerve Head Map
Provides
• Cup Area
• Rim Area
• RNFL Map
TSNIT graph
16 sector analysis
compares sector values
to normative database
and color codes result
based on probability
values (p values)
Color shaded regions
represent normative
database ranges based
on p-values
Glaucoma Analysis with the RTVue:
Nerve Head Map Parameters
RNFL Parameters
Optic Disc Parameters
All parameters color-coded based on
comparison to normative database
Glaucoma Analysis with the RTVue: Nerve Head Map
Nerve Head Map (NHM)
• Data Captured: 9,510 A scans
(pixels)
• Time: 370 msec
• Area covered: 4 mm diameter
circle Provides
•Cup Area
• Rim Area
• RNFL Map
TSNIT graph
Ganglion Cell Map (MM7)
• Data Captured: 14,810 A
scans (pixels)
• Time: 570 msec
• Area covered: 7 x 7 mm
Provides
• Ganglion cell complex
assessment in macula
• Inner retina thickness is:
• NFL
• Ganglion cell body
• Dendrites
3-D Optic Disc
• Data Captured: 51,712 A
scans (pixels)
• Time: 2 seconds
• Area covered: 4 x 4 X 2 mm
Provides
•3 D map
• Comprehensive
assessment
The ganglion cell complex (ILM – IPL)
Inner retinal layers provide complete Ganglion cell
assessment:
• Nerve fiber layer (g-cell axons)
• Ganglion cell layer (g-cell body)
• Inner plexiform layer (g-cell dendrites)
Images courtesy of Dr. Ou Tan, USC
Normal vs Glaucoma
Cup
Rim
RNFL
NHM4
Ganglion cell
assessment
with inner
retinal layer
map
GCC
Normal
Glaucoma
Glaucoma Cases
Optovue, RTVue
Glaucoma Patient Case BK
24-2 white on white visual field
64 year old
white male
Nerve Head Map on RTVue
Normal
Glaucoma Patient Case BK
Macula Inner Retina Map on RTVue
10-2 white on white visual field
Normal
RTVue Normative Database
• Age Adjusted comparisons for more
accurate comparisons
• Age and Optic Disc adjusted comparisons
for Nerve Head Map scans
• Over 300 eyes, ethnically mixed, collected
at 8 clinical sites worldwide
• IRB approved study from independent
agency
34
Nerve Head Map (NHM4)
with Database comparisons
Patient Information
RNFL Thickness Map
RNFL Sector Analysis
Optic Disc Analysis
Parameter Tables
TSNIT graph
Asymmetry Analysis
Ganglion Cell Complex (GCC)
with Database comparisons
Patient Information
GCC Thickness Map
Deviation Map
Parameter Table
Significance Map
Early Glaucoma
Borderline
Sector results
in Superiortemporal region
Abnormal
parameters
TSNIT dips
below normal
TSNIT shows
significant
Asymmetry
OS Normal
GCC Analysis may detect damage
before RNFL
GCC and RNFL analysis will be correlated,
however GCC analysis may be more sensitive
for detecting early damage
Glaucoma Progression Analysis
(Nerve Head Map of stable eye)
Thickness Maps
Change in
optic disc
parameters
TSNIT graph
comparisons
Change in
RNFL
parameters
RNFL trend
analysis
Glaucoma Progression Analysis
(GCC of stable glaucomatous eye)
Thickness Maps
Deviation Maps
Significance
Maps
GCC parameter
change analysis
Versatility: Scanning the Anterior Chamber
with the same device
Cornea
Adapter
Module
(CAM)
Higher resolution allows better
visualization of LASIK flap
2 years after LASIK with mechanical microkeratome
Image enhanced by frame averaging
056-CP
Post-LASIK interface fluid & epithelial
ingrowth
Epithelial ingrowth
Fibrosis
Fluid
Higher resolution helps visualize
pathogens
Acanthamoeba in 0.25% agar
Pachymetry Maps
Inferotemporal
thinning
Normal
Keratoconus
Angle Measurements
Normal
Narrow
LD044, OS
Narrow angle after peripheral iridotomy
Limbus
Angle
Opening
Distance
500 mm
anterior to
scleral
spur
(AOD 500)
Scleral spur
MaTa, OD
Normal Angle
Limbus
Trabecular
meshworkIris Space
750 mm
anterior to
scleral
spur
(TISA750)
Scleral spur
Advantages of the RTVue
• 5 micron resolution allows more structures and detail
to be visualized
• High speed allows larger areas to be scanned
• Layer by layer assessment
• Data-rich maps
• Volumetric analysis
• Comprehensive glaucoma assessment (Cup, Rim, RNFL,
ganglion cell complex)
• Normative Database
• Progression Analysis
• Anterior Chamber imaging
Thank You!