Comparison of Computer Aided Planimetry Between Simultaneous and Non-Simultaneous Stereo Optic Disc Photographs

Abstract

 

Purpose:

 

Planimetry of stereo color photographs of the optic disc is an essential component of the evaluation of glaucoma. We wanted to determine whether the method of acquiring stereo photographs, namely simultaneous and non-simultaneous stereo imaging has an effect on the evaluation of the images by experts.

 

Methods:

 

284 optic disc stereo images (left+right) were obtained from 44 eyes of 44 patients with open-angle glaucoma by imaging them three times sequentially, realigned and refocused each times, with both the Nidek 3DX simultaneous stereo camera and a manually operated standard Zeiss 30⁰ sequential stereo camera, so there were 6 stereo images per eye. Computer aided planimetry was performed on each stereo image in random order by three masked independent glaucoma experts to segment the image into cup, rim and background. The three planimetries for each image were combined into one majority vote planimetry image (figure 1). Linear Cup to Disc ratio (lcdr) was computed by dividing the area of the cup by the area of the disc. Overall mean lcdr was calculated for per patient per camera, and compared between cameras by Tukey corrected t-test.

 

Results:

 

Mean lcdr for Nidek was 0.64 (95% CI, 0.64-0.69) and 0.64 for Zeiss (95% CI, 0.59-0.69).

 

Conclusions:

 

Mean linear cup to disc ratio as evaluated by three glaucoma experts was the same with simultaneous and sequential manual stereo imaging. (Figure 1)  

 

Keywords: imaging/image analysis: clinical • optic disc • image processing