Wavefront coding with Jacobi-Fourier phase masks for retinal imaging.

Wavefront coding is a technique that combines optical phase elements and digital signal processing in order to increase the effective depth of focus of optical systems. The success of wavefront coding lies in the design of a suitable phase mask placed at the system's aperture. This element allows for image formation invariant under the effects of different second-order optical aberrations. In optical systems limited by temporally or spatially varying high-order aberrations, the use of wavefront coding has not been fully demonstrated. Here we propose the choice of Jacobi-Fourier shaped phase masks to produce sharp and clear retinal images of living eyes. To demonstrate the potential use of the technique, we analyze the performance of the Jacobi-Fourier phase masks through experimental simulations to alleviate aberrations for different eye aberrations. We will show that the best mask choice is robust to noise while keeping acceptable resolution and reducing image artefacts.

Optimal parameters for retinal illumination and imaging in fundus cameras.

A fundus camera is a complex optical system for imaging the retina of the eye. Designing a fundus camera requires the combination of an imaging system and an illumination system to share common optics. This combination of systems results in the need to find an optimal balance between imaging and illuminating the retina. We present a series of parameters and methods used to optimize the illumination system of a fundus camera while maintaining excellent image quality.

Detection of improper fixation in MTI photoscreening images.

PURPOSE: To determine the effect of fixation shift on photoscreening crescents, the ability of human interpreters to detect fixation shift, and the potential improvement by image processing. METHODS: MTI photoscreening (Medical Technologies & Innovations, Inc, Lancaster, PA) images, measured at 11 positions of gaze, were obtained from 10 subjects (9 with refractive error warranting spectacle correction). Photographs were taken with subjects fixating at 20, 15, 10, and 5 cm to the left and the right of the camera fixation target (1 m distant); 5 cm above and below the camera fixation target; and on-axis. Photographs were inspected by 11 experienced raters, who indicated if the subject appeared to be looking directly at the camera. The photographs were digitized, enlarged, contrast enhanced, and measured by 3 raters. For each photograph, distance from the corneal light reflex to the nasal limbus was measured and a measure of asymmetry computed. RESULTS: Raters could reliably detect off-axis fixation greater than 10 cm away from the intended fixation target. Raters correctly identified on-axis subject viewing 73% of the time. Crescents became larger when the fixation shifted off-axis for both the myopic and hyperopic subjects. Image analysis correctly classified 10 of 10 on-axis measurements and 34 of 39 off-axis measurements. CONCLUSION: Direct inspection of photoscreening images by trained raters can result in the failure to detect small but relevant errors of fixation. These fixation shifts can cause crescents to become larger than expected, resulting in false-positive classification. Image analysis offers a potential improvement in the detection of off-axis fixation in MTI photoscreening images.

Theoretical limits to visual performance.

Wavefront sensors and scanning laser technology are enabling the correction of the aberrations of the eye. The effects of aberrations on visual performance are reviewed, and the theoretical limit of visual performance is predicted to understand the ultimate endpoint of these new technologies. A schematic eye model that incorporates diffraction, chromatic aberration, photopic response, the Stiles-Crawford effect, and pupil size is ray-traced to determine its limiting optical properties. These properties are compared to the detection requirements of the retina and brain to determine the theoretical limit of foveal vision. The theoretical limits on foveal vision are found to be between 20/12 and 20/5, depending on pupil diameter. It is concluded that emerging refractive surgery technologies may provide substantial increases in visual performance.

Acular as a single agent for use as an antimiotic and anti-inflammatory in cataract surgery.

PURPOSE: To assess the safety and effectiveness of ketorolac tromethamine 0. 5% (Acular) as a cost-efficient single agent to prevent intraoperative miosis and postoperative inflammation in cataract surgery. METHODS: Both eyes of 26 patients were randomized to receive Acular preoperatively and postoperatively or flurbiprofen sodium (Ocufen) preoperatively and prednisolone acetate 1% (Pred Forte) postoperatively. Time scheduled between procedures was from 2 weeks to 1 month. Pupil dilation was measured preoperatively, intraoperatively, and at the end of surgery. Cell and flare were measured 1 day, 1 week, and 1 month postoperatively. RESULTS: A comparison of the Acular and the Ocufen/Pred Forte groups (n=22) showed no statistically significant differences in dilation (preoperative versus postpostoperative) or cell and flare postoperatively. CONCLUSIONS: Using Acular as a single agent was as effective as the combination of preoperative Ocufen and postoperative Pred Forte in preventing intraoperative miosis and postoperative inflammation in cataract surgery. The use of Acular as a single agent could save the expense of using separate anti-inflammatory and antimiotic preparations preoperatively and postoperatively, enhancing convenience for the surgeon and surgical facility.

Eye movement during laser in situ keratomileusis.

PURPOSE: To measure eye motion in patients having laser in situ keratomileusis (LASIK) using a video technique and determine centration and variance of the eye position during surgery. SETTING: Laser refractive surgery center. METHODS: The procedure was videotaped in 5 consecutive eyes having LASIK performed by a single surgeon with the VISX Star S2 excimer laser. Following surgery, video images of the eyes were digitized and stored in a computer for processing. Digitized images were obtained at a rate of 25 images per second during the laser procedure. The pupil margin and a visual landmark, such as a scleral blood vessel, were identified in the initial image of each eye. Custom software was used to track the location of the landmark and the pupil center in subsequent images. RESULTS: Three of the 5 eyes were well centered on average. The remaining 2 eyes were decentered inferiorly by approximately 0.25 mm. The standard deviation in all eyes was approximately 0.10 mm. CONCLUSIONS: With these techniques, the position of the entrance pupil center relative to the excimer laser axis could be determined. Although the system is not fast enough to be used during surgery, it does allow quantification of centration and intraoperative motion after surgery.

Corneal ablation patterns to correct for spherical aberration in photorefractive keratectomy.

PURPOSE: To determine the spherical aberration introduced by photorefractive keratectomy (PRK) and customize ablation patterns to compensate for this aberration and improve post-PRK visual performance. SETTING: Department of Ophthalmology, University of Arizona, Tucson, Arizona, USA. METHODS: Presurgical and postsurgical corneal topography of 16 patients who had PRK with the Summit OmniMed laser were obtained. The data were applied to a schematic eye model, and exact ray tracing was used to determine the introduction of spherical aberration from the procedure. Optimization routines were used to determine the ideal ablation pattern. RESULTS: The magnitude of the spherical aberration introduced into the eyes after PRK increased with the level of attempted correction. The theoretical ideal ablation pattern requires additional flattening of the ablation periphery to avoid the introduction of spherical aberration. CONCLUSIONS: Current PRK ablations introduce spherical aberration into the eye. Modifying the existing ablation algorithms to compensate for spherical aberration may boost postoperative visual performance.

Customized corneal ablation and super vision.

PURPOSE: To review the early development of new technologies that are becoming available through customized corneal ablation techniques. METHODS: The authors describe the early development of two diagnostic methods to perform customized corneal ablation as well as a variety of new treatment modalities in development. RESULTS AND CONCLUSION: Results using the wavefront sensors indicate that these techniques have the potential to be more sensitive than traditional refraction and keratometry. Subtle defects such as coma and spherical aberration can be detected and treated. A whole series of new technologies are being incorporated to treat patients with customized corneal ablation. [J Refract Surg 2000;16(suppl):S230-S235].

Customized and low spherical aberration corneal ablation design.

PURPOSE: The purpose of this study was to use the Arizona Eye Model to help guide customization of corneal excimer ablation and reduce spherical aberration. METHODS: Myopic eyes were treated with the Nidek EC-5000 excimer laser with a 5.5-mm diameter optic ablation zone and a 7.0-mm diameter transition ablation zone. We analyzed preoperative and postoperative corneal topographies using height mapping. From this data, refractive error profiles and maps were constructed using the Arizona Eye Model. The first group of patients had refractions between -2.00 and -5.00 D. Data was obtained by subtracting postoperative topography from preoperative topography. We then plotted the ideal ablation pattern if no additional spherical aberration was introduced when compared to preoperative topographies. RESULTS: We found that in the central 4 mm, the ablation pattern was highly acceptable, with negligible spherical aberration. As the ablation moved out toward 6 mm, there was increasing spherical aberration. Newer ablation designs require more flattening in the midperiphery of the cornea. These flatter peripheral designs require more blending in the periphery and larger transition zones. CONCLUSION: The use of computerized corneal topography in eye modeling is helpful in designing new ablation patterns to reduce optical and spherical aberration. Ablation zone design is critical to maximizing optical and biologic tolerance.

Custom photorefractive keratectomy ablations for the correction of spherical and cylindrical refractive error and higher-order aberration.

Photorefractive keratectomy is an evolving refractive procedure for correcting myopia, hyperopia, and astigmatism. Earlier descriptions of the patterns required for this surgery are based on paraxial optics. In this investigation the required pattern is generalized to account for spherical refractive error (defocus), axial astigmatism of arbitrary orientation, and fourth-order aberrations of the eye. The patterns described in this study can be used to customize photorefractive keratectomy and to provide corrections that account for aberration content as well as paraxial values. Furthermore, a description of the pattern along the boundary of the optical zone is given, which may prove useful in designing blending zones. An example of the use of these techniques is given for a schematic eye model.

Using corneal height maps and polynomial decomposition to determine corneal aberrations.

PURPOSE: To review the use of corneal videokeratoscopic height date, elaborate on the advantages and disadvantages of such data, describe techniques for overcoming the limitations of height data, and demonstrate its use in quantifying the optical properties and aberrations of the cornea. METHODS: The steep sag of the cornea hides fine variations in corneal height that arise naturally or due to disease or surgery. The dynamic range, or ratio of the overall sag to the feature height, is the primary limitation of videokeratoscopic height data. Techniques for removing single or multiple reference surfaces are described in detail, and applications of the methodology to wavefront and raytracing analysis of corneal aberrations arising from radial keratotomy (RK), photorefractive keratectomy (PRK), and keratoconus are described. RESULTS: Removing a single reference surface from the raw corneal height data begins to reveal subtle variations in corneal height. However, expansion of surface height data into a complete set of basis functions provides a sophisticated method for extracting high-order corneal variations. Choosing an orthogonal basis set provides a robust least-squares fit and forms unique expansions of the surface. The resulting coefficients are uncorrelated and form a simple measure of the optical quality. CONCLUSION: Videokeratoscopic height data are useful for analyzing and quantifying corneal deformity arising from disease or refractive surgery and they provide a sophisticated alternative or complement to dioptric power maps.

Cone dimensions in keratoconus using Zernike polynomials.

PURPOSE: To determine the physical dimensions and location of the cone in keratoconic corneas from videokeratoscopic height data. METHODS: Corneal height date from keratoconus patients are obtained with a commercial videokeratoscope and decomposed into the set of orthogonal Zernike polynomials. The low-order Zernike terms are removed from the height data to expose the cone. The location of the peak of the cone and the cone's height and lateral dimensions are measured from the residual height data. This technique is compared to previous dioptric power-based methods of locating the cone. RESULTS: The corneal protrusion due to keratoconus takes on a variety of shapes and locations. These dimensions are easily measured with the Zernike decomposition technique and provide more detailed and accurate information regarding the cone than dioptric power data. CONCLUSIONS: The proposed method for quantifying cone dimensions and locale in keratoconic corneas provides practical and detailed information which may be useful in tracking the progression of the disease, contact lens fitting, and surgical planning.

Optical modeling of radial keratotomy incision patterns.

PURPOSE: To determine the optical effects of higher-order corneal shape variations resulting from radial keratotomy. METHODS: Videokeratoscopic height data were obtained postoperatively from several patients who had undergone radial keratotomy. For each of clear central zone sizes 3.00 mm, 4.00 mm, and 4.75 mm, two patients were chosen randomly from the larger study group. Data obtained 2 weeks postoperatively from these six patients were decomposed into the Zernike polynomials, and the low-order expansion terms were removed to disclose corneal height variations (the radial keratotomy artifact). The artifact was applied to a schematic eye model, and exact ray-tracing was used to evaluate visual performance, which was defined as a function of pupil diameter, optical zone (central clear zone) size, and radial keratotomy artifact centration. RESULTS: The radial keratotomy artifact degrades visual performance at midspatial frequencies more than it does at high spatial frequencies. This effect is most pronounced for smaller optical zones and for a pupil diameter of 4 mm. Visual performance remains nearly constant for small decentration (0.5 mm or less) of the radial keratotomy optical zone from the corneal apex. CONCLUSIONS: Residual refractive error, corneal asphericity, and the radial keratotomy artifact all affect visual performance after radial keratotomy. Isolated effects of the radial keratotomy artifact degrade visual performance, with the level of degradation dependent on pupil size, optical zone size, and centration of the procedure. More research is necessary to combine the radial keratotomy artifact with changes in corneal asphericity and to further quantify the optical effects of radial keratotomy.

Keratoconus detection based on videokeratoscopic height data.

PURPOSE: To develop a videokeratoscopic-based keratoconus detection scheme that avoids the ambiguity of dioptric power definitions and videokeratoscope design. METHODS: Corneal height data obtained with a commercial videokeratoscope are decomposed into the set of orthogonal Zernike polynomials. Expansion coefficients of a "normal" group and a keratoconus group are compared to find significant differences. Elevated Zernike terms are used to detect the disease in these populations. The performance of this detection scheme is compared to other videokeratoscopic keratoconus Indices. RESULTS: Two low-order Zernike polynomial terms are identified as being elevated in keratoconus patients and combined to form a new detection index. This index performed at least as well as keratoconus detection schemes based on the inferior-superior (I-S) value, the steepest radial axes (SRAX), and the Surface Asymmetry Index (SAI) for the samples studied. CONCLUSION: The proposed Zernike scheme offers a potentially viable algorithm for detecting keratoconus that avoids the ambiguities of dioptric power definitions and is independent of videokeratoscope design.

Comparison of three videokeratoscopes in measurement of toric test surfaces.

PURPOSE: We compared the accuracy of the Computed Anatomy TMS-1 (1.41), the EyeSys Laboratories Corneal Analysis System (2.1), and the Visioptic EH-270 (3.0) videokeratoscopes in measuring toric surfaces. These non-rotationally symmetric aspheric surfaces served as models of corneal astigmatism. METHODS: Precision diamond-turned toric surfaces modeling 0.00 diopter (D) to 7.00 D of astigmatism were fabricated. A three-dimensional contact profiler was developed to calibrate the aspheric surfaces. Videokeratoscopic data taken at "best focus" were compared to the theoretical shape to quantify device measurement errors. RESULTS: The Computed Anatomy system measurement accuracy shows no statistically significant correlation between measurement error and surface toricity (r2 < 0.13). Measurement error increased linearly with surface astigmatism for the EyeSys Laboratories system (0.12 D rms error per D of astigmatism, r2 > 0.96, p < 0.001 and the Visioptic system (0.03 D error per D of astigmatism, r2 = 0.88, p < 0.001). CONCLUSIONS: This study found systematic performance differences among the three machines. Under ideal alignment conditions, the Computed Anatomy TMS-1 is more accurate at detecting astigmatism. The EyeSys Laboratories Corneal Analysis System apparently underestimates the amount of surface astigmatism because of excessive data smoothing. The Visioptic EH-270 errors are primarily in the central zones and may be due to ring localization errors.

Representation of videokeratoscopic height data with Zernike polynomials.

Videokeratoscopic data are generally displayed as a color-coded map of corneal refractive power, corneal curvature, or surface height. Although the merits of the refractive power and curvature methods have been extensively debated, the display of corneal surface height demands further investigation. A significant drawback to viewing corneal surface height is that the spherical and cylindrical components of the cornea obscure small variations in the surface. To overcome this drawback, a methodology for decomposing corneal height data into a unique set of Zernike polynomials is presented. Repeatedly removing the low-order Zernike terms reveals the hidden height variations. Examples of the decomposition-and-display technique are shown for cases of astigmatism, keratoconus, and radial keratotomy.

Visual acuity modeling using optical raytracing of schematic eyes.

PURPOSE: We developed a methodology to predict changes in visual performance that result from changes in the optical properties of the eye. METHODS: Exact raytracing of schematic eyes was used to calculate the point spread function and the modulation transfer function of the visual system. The Stiles-Crawford effect, photopic response, diffraction, and the retinal contrast sensitivity are included in the model. Visual acuity was predicted by examining the modulation of the resultant retinal image of a bar target and by determining when the modulation falls below a threshold value. Visual acuity was predicted for refractive errors ranging from 0 to 5 diopters and for pupil diameters ranging from 0.5 to 8 mm. RESULTS: Visual acuity predictions were compared to clinically found Snellen visual acuities and were found to be highly correlated (r2 = .909). CONCLUSIONS: This modeling technique shows promise as a means of evaluating clinical and surgical procedures before undertaking clinical trails.

Modeling soft contact lenses in raytrace code.

The surface interferogram feature of raytrace codes is used to model the optical performance of soft contact lenses, especially in situations where the lens is decentered. The contact lens is totally defined by a thickness data file, which can easily be shifted relative to the corneal vertex of a schematic eye model.