Purchase this article with an account.
N.J. Sheen, R. Goyal, J.M. Wild, R.V. North, J.E. Morgan; Comparison of Sequential and Simultaneous Methods of Stereoscopic Image Capture in the Diagnosis of Glaucomatous Optic Neuropathy . Invest. Ophthalmol. Vis. Sci. 2003;44(13):3360.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: To compare sequential and simultaneously acquired optic disc images in glaucoma diagnosis, and to determine the stereopsis quality of both methods of image capture. Methods: Fifty-two glaucoma patients, diagnosed on the basis of a Hodapp grading of 2 serial Humphrey 24-2 SITA-Standard fields, and fifty-four normal subjects were imaged with a simultaneous Nidek 3-Dx, and a sequential Nikon NF-505 camera. Digitised images of the optic discs were viewed stereoscopically using a Z screen (Stereographics Corp., San Rafael, CA). A single observer, (resident in training) subjectively graded each disc image from the two cameras as normal or glaucomatous and then, using a cursor, outlined cup and disc borders. Neuroretinal rim and optic disc areas were calculated at 30-degree intervals using custom software to provide measurements for quantitative analysis. Additionally, three experienced observers, masked to the camera type used, graded the stereoscopic quality of the disc images as good, fair or poor. Results: Normals; average MD –0.4dB, PSD 1.9. Glaucoma; MD – 4.5dB, PSD 5.5. Subjective assessment gave sensitivity and specificity values of 81% and 94% respectively for simultaneously acquired optic disc images, and a sensitivity and specificity of 69% and 93% respectively for sequentially acquired images. Optimal diagnostic precision for a quantitative analysis was achieved using linear regression, plotting the disc area against Log transformed neuroretinal rim segment areas, with 95% prediction interval cut-offs for disc area. This gave a sensitivity of 81% and specificity of 83% for the simultaneous images, and a sensitivity of 73% and specificity of 93% for the sequential images. Stereoscopic quality for simultaneous images was poor in 5% of images, fair in 24% and good in 71%. For the sequential images 54% were poor, 35% were fair and 11% were good. Conclusions: Simultaneously acquired Nidek 3-Dx images are superior to sequentially acquired Nikon NF-505 images in qualitative and quantitative disease discrimination and in stereoscopic quality.
This PDF is available to Subscribers Only