11 Segmentation and Visualization of Drusen …
339
reliable way to identify and distinguish areas of drusen and GA is by inspecting
SD-OCT cubes B-scan by B-scan.
More so, the traditional SVP projection usually masks drusen and does not take
advantage of the ability of SD-OCT to resolve structures in the depth-axis. The proposed method takes advantage of this ability and allows the visualization of both
drusen and GA present in the macula in a single image, while clearly distinguishing
between them using a false color mapping. The examples of this improved visualization technique is presented in Fig. 11.37. Since we analyzed visually the results
from 82 different cases, we only showed the results from two of the cases given the
chapter length limitations. We also obtained satisfactory result by visual inspection
for the rest of the cases. We analyzed the results produced by three of the cubes
quantitatively, by comparing with SVP and CFP. Overall, the proposed method is
better for the drusen and GA visualization than SVP and CFP. A combination of two
readers was able to clearly identify all GA areas in our proposed method, while also
identifying the majority of drusen.
The aim of this section is to simultaneously display drusen and GA in a single
projection image from 3D SD-OCT images, not to segment drusen and GA. Using
the difference between the actual RPE segmentation and the RPE floor (or Bruch’s
membrane), drusen can be segmented from OCT images.
In conclusion, we present a new visualization method for drusen and GA. It
enhances GA visualization by utilizing the bright choroid and thin RPE characteristics of GA in the visualization method. To efficiently and effectively display drusen
and GA in a single image, we present a false color fusion strategy to combine the
drusen and GA projection images. Our experimental results show that the false color
image is more effective for the drusen and GA visualization than the SVP image and
CFP. Most of the drusen are not visible in SVP images, and the contrast of GA in
SVP images is lower than that in the GA projection image. Even though, drusen and
GA are visible in CFPs, they are difficult to distinguish because of low contrast. In
false color images, the color difference between drusen and GA is more obvious.
The proposed method may be used in improving the ability of ophthalmologist to
visualize and evaluate drusen and GA.
11.4 Conclusion
This chapter presents several novel algorithm for semi-automated, automated GA and
drusen segmentation and visualization in SD-OCT images to enable robust, accurate,
and objective quantitative measurements of both drusen and GA extent and location
automatically. The proposed method combines different novel algorithms and utilizes well known techniques in the implementation of our algorithms. Our technique
seems more robust, through quantitative and qualitative experimental results which
demonstrate that our algorithms show good agreement when compared to segmentation performed by other researcher’s algorithm. The proposed algorithms may be
clinically useful in providing relatively reliable GA and drusen qualitative and quan-
339
reliable way to identify and distinguish areas of drusen and GA is by inspecting
SD-OCT cubes B-scan by B-scan.
More so, the traditional SVP projection usually masks drusen and does not take
advantage of the ability of SD-OCT to resolve structures in the depth-axis. The proposed method takes advantage of this ability and allows the visualization of both
drusen and GA present in the macula in a single image, while clearly distinguishing
between them using a false color mapping. The examples of this improved visualization technique is presented in Fig. 11.37. Since we analyzed visually the results
from 82 different cases, we only showed the results from two of the cases given the
chapter length limitations. We also obtained satisfactory result by visual inspection
for the rest of the cases. We analyzed the results produced by three of the cubes
quantitatively, by comparing with SVP and CFP. Overall, the proposed method is
better for the drusen and GA visualization than SVP and CFP. A combination of two
readers was able to clearly identify all GA areas in our proposed method, while also
identifying the majority of drusen.
The aim of this section is to simultaneously display drusen and GA in a single
projection image from 3D SD-OCT images, not to segment drusen and GA. Using
the difference between the actual RPE segmentation and the RPE floor (or Bruch’s
membrane), drusen can be segmented from OCT images.
In conclusion, we present a new visualization method for drusen and GA. It
enhances GA visualization by utilizing the bright choroid and thin RPE characteristics of GA in the visualization method. To efficiently and effectively display drusen
and GA in a single image, we present a false color fusion strategy to combine the
drusen and GA projection images. Our experimental results show that the false color
image is more effective for the drusen and GA visualization than the SVP image and
CFP. Most of the drusen are not visible in SVP images, and the contrast of GA in
SVP images is lower than that in the GA projection image. Even though, drusen and
GA are visible in CFPs, they are difficult to distinguish because of low contrast. In
false color images, the color difference between drusen and GA is more obvious.
The proposed method may be used in improving the ability of ophthalmologist to
visualize and evaluate drusen and GA.
11.4 Conclusion
This chapter presents several novel algorithm for semi-automated, automated GA and
drusen segmentation and visualization in SD-OCT images to enable robust, accurate,
and objective quantitative measurements of both drusen and GA extent and location
automatically. The proposed method combines different novel algorithms and utilizes well known techniques in the implementation of our algorithms. Our technique
seems more robust, through quantitative and qualitative experimental results which
demonstrate that our algorithms show good agreement when compared to segmentation performed by other researcher’s algorithm. The proposed algorithms may be
clinically useful in providing relatively reliable GA and drusen qualitative and quan-