Purchase this article with an account.
Y. Li, M.R. Chalita, J. Goldsmith, V. Westphal, B.A. Bower, R. Shekhar, A.M. Rollins, J.A. Izatt, D. Huang; Automated Anterior Chamber Biometry with High-speed Optical Coherence Tomography . Invest. Ophthalmol. Vis. Sci. 2003;44(13):3604.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: Accurate sizing of angle-supported anterior chamber intraocular lens (AC-IOL) is crucial in preventing complications. To accurate measure AC width and other dimensions, we developed a high-speed optical coherence tomography (OCT) system and automated image processing. Methods: The OCT prototype operated at 1.3 micron wavelength and was capable of 8 images/sec at 500 axial scans/image. Scan dimensions are 16mm wide and 6mm deep (in air). The OCT scanner was adapted to a slit-lamp with video camera. Horizontal cross-sectional OCT images of the anterior segment was obtained. Each of the 40 eyes from 20 healthy volunteers was scanned 3 times. A computer algorithm was developed to measure angle-to-angle AC width, AC depth, and lens vault. These measurements were also obtained manually by 3 ophthalmologist using computer calipers on screen displays of OCT images. Results: The computer algorithm successfully measured AC diameter and AC depth from all 120 OCT images. The difference between computer and human measurements was 0.13+/-0.14 mm (mean +/- SD) for AC width, 0.04+/-0.04 mm for AC depth, and 0.08+/-0.06mm for lens vault. The image-to-image reproducibility of computer measurements (pooled SD) is 0.11 mm for AC width, 0.04mm for AC depth, and 0.07mm for lens vault. The image-to-image reproducibility of human measurement was 0.13 mm for AC width, 0.05 mm for AC depth, and 0.06mm for lens vault. The agreement between the human raters (inter-rater SD by analysis of variance) is 0.30 mm for AC width, 0.05 mm for AC depth, and 0.08mm for lens vault. Conclusion: Due to its longer wavelength, the OCT system was able to penetrate and image the angles. The speed was sufficient high for reproducible AC width measurement. The automated computer algorithm agrees well with human raters. The use of a computer measurement algorithm avoids the relatively large disagreement between human raters for AC width. The use of OCT to directly measure AC width may improve the fitting of AC-IOL and avoid complications such as IOL dislocation and pupil ovalization.
This PDF is available to Subscribers Only