Biomedical subjects

Luis Alberto Carvalho

Publications and source records attributed to Luis Alberto Carvalho.

7 recordsLinked to original sources

PubMed2006

Quantitative comparison of different-shaped wavefront sensors and preliminary results for defocus aberrations on a mechanical eye.

PURPOSE: There is a general acceptance among the scientific community of Cartesian symmetry wavefront sensors (such as the Hartmann-Shack (HS) sensor) as a standard in the field of optics and vision science. In this study it is shown that sensors of different symmetries and/or configurations should also be tested and analyzed in order to quantify and compare their effectiveness when applied to visual optics. Three types of wave-aberration sensors were developed and tested here. Each sensor has a very different configuration and/or symmetry (dodecagonal (DOD), cylindrical (CYL) and conventional Hartmann-Shack (HS)). METHODS: All sensors were designed and developed in the Physics Department of the Universidade de São Paulo--São Carlos. Each sensor was mounted on a laboratory optical bench used in a previous study. A commercial mechanical eye was used as control. This mechanical eye has a rotating mechanism that allows the retinal plane to be positioned at different axial distances. Ten different defocus aberrations were generated: 5 cases of myopia from -1D to -5D and 5 cases of hyperopia, from +1D to +5D, in steps of 1D following the scale printed on the mechanical eye. For each wavefront sensor a specific image-processing and fitting algorithm was implemented. For all three cases, the wavefront information was fit using the first 36 VSIA standard Zernike polynomials. Results for the mechanical eye were also compared to the absolute Zernike surface generated from coefficients associated with the theoretical sphere-cylinder aberration value. RESULTS: Precision was analyzed using two different methods: first, a theoretical approach was used by generating synthetic Zernike coefficients from the known sphere-cylinder aberrations, simply by applying sphere-cylinder equations in the backward direction. Then comparisons were made of these coefficients with the ones obtained in practice. Results for DOD, HS and CYL sensors were, respectively, as follows: mean of root mean square (RMSE) for all aberrations, when theoretical Zernike coefficients were used as control, was 0.22, 0.66 and 0.26 microns; RMSE of sphere-cylinder values when compared to autorefractor measurements was 0.18D, 0.22D and 0.35D for sphere, 0.14D, 0.24D and 0.17D for cylinder, 34.36 degrees, 35.16 degrees and 26.36 degrees for axis; RMSE of sphere-cylinder values when theoretical values were used as control was 0.11D, 0.29D and 0.46D for sphere, 0.15D, 0.28D and 0.17D for cylinder, 19.71 degrees, 25.56 degrees and 18.56 degrees for axis. CONCLUSION: The main conclusion is that the symmetry of an optical sensor is not an important consideration when measuring typical eye aberrations such as defocus (myopic and hyperopic), but there are differences. In this sense, the polar symmetry sensors render results that are equivalent to the traditional Cartesian Hartmann-Shack sensor, but furnish an easier method for determining the optical center.

Calibration↗

PubMed2006

The placido wavefront sensor and preliminary measurement on a mechanical eye.

PURPOSE: The hardware and software of a novel wavefront sensor was developed (The sensor presented here is patent pending.). It has the same principal of the Hartmann-Shack (HS) and other sensors that are based on slope information for recovery of wavefront surface, but a different symmetry, and does not use individual microlenses. This polar symmetry might offer differences during practical measurements that may add value to current and well-established "gold standard" techniques. METHODS: The sensor consists of a set of concentric "half-donut" surfaces (longitudinally sectioned toroids) molded on an acrylic surface with a CCD located at the focal plane. When illuminated with a plane wavefront, it focuses a symmetric pattern of concentric discs on the CCD plane; for a distorted wavefront, a nonsymmetric disc pattern is formed (similar to images of a placido-based videokeratographer). From detection of shift in the radial direction, radial slopes are computed for a maximum of 2880 points, and the traditional least-squares procedure is used to fit these partial derivatives to a set of 15 conventional OSA-VSIA Zernike polynomials. Theoretical computations for several synthetic surfaces containing low-order aberration (LOA) and high-order aberration (HOA) were implemented for both the HS and the new sensor. RESULTS: Root mean square error (RMSE) in microns when theoretical data was taken as control, for HS sensor and new sensor, was 0.02 and 0.00003 for LOA (defocus, astigmatism) and 0.07 and 0.06 for HOA (coma, spherical, and higher terms), respectively. After this, practical preliminary measurements on a mechanical eye with a 5-mm pupil and 10 different defocus aberrations ranging from -5 D to 5 D, in steps of 1 D, were compared between sensors. RMSE for difference in measurements for HS and new sensor for sphere, cylinder, and axis, was 0.13 D, 0.07 D, and 11. Measurements were taken only on defocus aberrations. Qualitative images for astigmatism are shown. DISCUSSION: Although practical in vivo tests were not conducted in this first study, we also discuss certain possible alignment differences that may arise as a result of the different symmetry of the new sensor. To take any conclusive assumption regarding the accuracy and/or precision of this new sensor, when compared with other well-established sensors, statistically significant in vivo measurements will need to be conducted.

Cornea↗

PubMed2006

A new wavefront sensor with polar symmetry: quantitative comparisons with a Shack-Hartmann wavefront sensor.

PURPOSE: A novel wavefront sensor has been developed. It follows the same principle of the Shack-Hartmann wavefront sensor in that it is based on slope information. However, it has a different symmetry, which may offer benefits in terms of application. METHODS: The new wavefront sensor consists of a set of donut-shaped acrylic lenses with a charge coupled device located at the focal plane. From detection of shift in the radial direction, radial slopes are computed for 2880 points. Theoretical computations for higher order aberrations and lower order aberrations were implemented for the Shack-Hartmann wavefront sensor and the new wavefront sensor, and practical measurements were conducted on several sphere-cylinder trial lenses. RESULTS: The overall mean value of root mean square error (RMSE) (in microns) for theoretical computations was 0.03 for the Shack-Hartmann wavefront sensor and 0.02 for the new wavefront sensor. The mean value of RMSE for lower order aberrations (1-5) was 0.01 and 0.00003, and for higher order aberrations was 0.02 and 0.02, for the Shack-Hartmann and new wavefront sensors, respectively. For practical measurements (sphere, cylinder, axis), the standard deviation was 0.04 diopters (D), 0.04 D, and 4 degrees for the new wavefront sensor and 0.02 D, 0.02 D, and 5 degrees for the Shack-Hartmann wavefront sensor. CONCLUSIONS: Precision of the new wavefront sensor when measuring astigmatic and spherical surfaces is compatible with the Shack-Hartmann wavefront sensor. Centration with this new sensor is an absolute process using the center of the entrance pupil, which is where the line of site passes. This wavefront sensor, similar to the Shack-Hartmann sensor, does not eliminate the possibility of tilt. For more conclusive and statistically valid data, in vivo measurements are needed.

Cornea↗

PubMed2005

Preliminary results of neural networks and zernike polynomials for classification of videokeratography maps.

PURPOSE: Our main goal in this work was to develop an artificial neural network (NN) that could classify specific types of corneal shapes using Zernike coefficients as input. Other authors have implemented successful NN systems in the past and have demonstrated their efficiency using different parameters. Our claim is that, given the increasing popularity of Zernike polynomials among the eye care community, this may be an interesting choice to add complementing value and precision to existing methods. By using a simple and well-documented corneal surface representation scheme, which relies on corneal elevation information, one can generate simple NN input parameters that are independent of curvature definition and that are also efficient. METHODS: We have used the Matlab Neural Network Toolbox (MathWorks, Natick, MA) to implement a three-layer feed-forward NN with 15 inputs and 5 outputs. A database from an EyeSys System 2000 (EyeSys Vision, Houston, TX) videokeratograph installed at the Escola Paulista de Medicina-Sao Paulo was used. This database contained an unknown number of corneal types. From this database, two specialists selected 80 corneas that could be clearly classified into five distinct categories: (1) normal, (2) with-the-rule astigmatism, (3) against-the-rule astigmatism, (4) keratoconus, and (5) post-laser-assisted in situ keratomileusis. The corneal height (SAG) information of the 80 data files was fit with the first 15 Vision Science and it Applications (VSIA) standard Zernike coefficients, which were individually used to feed the 15 neurons of the input layer. The five output neurons were associated with the five typical corneal shapes. A group of 40 cases was randomly selected from the larger group of 80 corneas and used as the training set. RESULTS: The NN responses were statistically analyzed in terms of sensitivity [true positive/(true positive + false negative)], specificity [true negative/(true negative + false positive)], and precision [(true positive + true negative)/total number of cases]. The mean values for these parameters were, respectively, 78.75, 97.81, and 94%. CONCLUSION: Although we have used a relatively small training and testing set, results presented here should be considered promising. They are certainly an indication of the potential of Zernike polynomials as reliable parameters, at least in the cases presented here, as input data for artificial intelligence automation of the diagnosis process of videokeratography examinations. This technique should facilitate the implementation and add value to the classification methods already available. We also discuss briefly certain special properties of Zernike polynomials that are what we think make them suitable as NN inputs for this type of application.

Artificial Intelligence↗

PubMed2005

Accuracy of Zernike polynomials in characterizing optical aberrations and the corneal surface of the eye.

PURPOSE: Zernike polynomials have been successfully used for approximately 70 years in many different fields of optics. Nevertheless, there are some recent discussions regarding the precision and accuracy of these polynomials when applied to surfaces such as the human cornea. The main objective of this work was to investigate the absolute accuracy of Zernike polynomials of different orders when fitting several types of theoretical corneal and wave-front surface data. METHODS: A set of synthetic surfaces resembling several common corneal anomalies was sampled by using cylindrical coordinates to simulate the height output files of commercial videokeratography systems. The same surfaces were used to compute the optical path difference (wave-front [WF] error), by using a simple ray-tracing procedure. Corneal surface and WF error was fit by using a least-squares algorithm and Zernike polynomials of different orders, varying from 1 to 36 OSA-VSIA convention terms. RESULTS: The root mean square error (RMSE) ranged-from the most symmetric corneal surface (spherical shape) through the most complex shape (after radial keratotomy [RK]) for both the optical path difference and the surface elevation for 1 through 36 Zernike terms-from 421.4 to 0.8 microm and 421.4 to 8.2 microm, respectively. The mean RMSE for the maximum Zernike terms for both surfaces was 4.5 microm. CONCLUSIONS: These results suggest that, for surfaces such as that present after RK, in keratoconus, or after keratoplasty, even more than 36 terms may be necessary to obtain minimum accuracy requirements. The author suggests that the number of Zernike polynomials should not be a global fixed conventional or generally accepted value but rather a number based on specific surface properties and desired accuracy.

Cornea↗

PubMed2004

A simple and effective algorithm for detection of arbitrary Hartmann-Shack patterns.

The Hartmann-Shack sensor is well known for its capability of detecting optical aberrations from telescopic images, normally caused by the atmospheric turbulence. Since the mid-1990s, this sensor has been adapted to work with ophthalmic instrumentation to measure aberrations of the human eye. In astronomical applications, the image processing is implemented using robust closed-loop hardware systems, allowing for very high frequencies (usually in the range of 30-500 Hz). In ophthalmic instruments, this hardware solution is not proper, for two reasons: first, prices of ophthalmic instrumentation have to be at more accessible prices and, second, there is no need for ultra-high frequency closed-loop systems; in fact, as we will see along the text, recent studies of the dynamics of the eye's aberrations show that closed-loop systems of very much lower frequencies (from 10 to 40 Hz) may well eliminate the undesired micro-fluctuations in the eye's aberrations and allow for diffraction limited measurements. We have developed an effective algorithm for detection of HS patterns using an affordable commercial hardware configuration and easy to implement software. To our knowledge, this type of information has not been thoroughly disclosed elsewhere, probably due to proprietary reasons.

Algorithms↗

PubMed2004

Absolute accuracy of Placido-based videokeratographs to measure the optical aberrations of the cornea.

Advances in excimer laser spot size and ablation pattern capabilities, together with the development of Hartmann-Shack sensors for high-resolution eye aberration measurements, have taken refractive surgery to a new stage. Nevertheless, studies have shown that the lens contributes differently to total aberration and depends on accommodation and age factors. Therefore, the choice of which aberrations to correct is also related to the internal optics of the eye, although only the cornea undergoes surface changes. A simple model is to subtract the total aberrations of the eye from the aberrations of only the cornea, which are calculated based on its three-dimensional profile. This has been done successfully, in practice, for certain commercial videokeratographs (VKS), but it is important to analyze the consequences of general flaws associated with VKS systems when used to estimate the corneal aberrations. A computer simulation of Placido-based VKS has been developed and applied to several theoretical surfaces resembling corneal profiles. The resulting Placido images were used to retrieve the corneal profiles. Corneal surface data were input into an algorithm that calculates the wave-front aberration as the difference in optical path length of the principal ray of an on-axis object point and several marginal rays. Wave aberration was then fit by a set of Zernike polynomials using the least-square method. From the Zernike coefficients, the correlation of different corneal profiles to their optical aberrations could be estimated. Misalignment (decentralization), improper focusing, and tilt were also implemented to the VKS simulation, and a quantitative analysis of the consequences on precision of wave aberration estimation was undertaken.

Algorithms↗