Corneal power evaluation after myopic corneal refractive surgery using artificial neural networks.

BACKGROUND: Efficacy and high availability of surgery techniques for refractive defect correction increase the number of patients who undergo to this type of surgery. Regardless of that, with increasing age, more and more patients must undergo cataract surgery. Accurate evaluation of corneal power is an extremely important element affecting the precision of intraocular lens (IOL) power calculation and errors in this procedure could affect quality of life of patients and satisfaction with the service provided. The available device able to measure corneal power have been tested to be not reliable after myopic refractive surgery. METHODS: Artificial neural networks with error backpropagation and one hidden layer were proposed for corneal power prediction. The article analysed the features acquired from the Pentacam HR tomograph, which was necessary to measure the corneal power. Additionally, several billion iterations of artificial neural networks were conducted for several hundred simulations of different network configurations and different features derived from the Pentacam HR. The analysis was performed on a PC with Intel® Xeon® X5680 3.33 GHz CPU in Matlab® Version 7.11.0.584 (R2010b) with Signal Processing Toolbox Version 7.1 (R2010b), Neural Network Toolbox 7.0 (R2010b) and Statistics Toolbox (R2010b). RESULTS AND CONCLUSIONS: A total corneal power prediction error was obtained for 172 patients (113 patients forming the training set and 59 patients in the test set) with an average age of 32 ± 9.4 years, including 67% of men. The error was at an average level of 0.16 ± 0.14 diopters and its maximum value did not exceed 0.75 dioptres. The Pentacam parameters (measurement results) providing the above result are tangential anterial/posterior. The corneal net power and equivalent k-reading power. The analysis time for a single patient (a single eye) did not exceed 0.1 s, whereas the time of network training was about 3 s for 1000 iterations (the number of neurons in the hidden layer was 400).

Intraocular pressure evaluation in healthy eyes and diseased ones using contact and non contact devices.

PURPOSE: To analyze and compare intraocular pressure (IOP) values measured in healthy subjects (HS), keratoconus (KC) patients and patients that underwent myopic photorefractive keratectomy (REF), using Goldmann applanation tonometry (GAT), dynamic contour tonometry (DCT), ocular response analyzer (ORA) and Corvis ST (CST). METHODS: The study included 76 eyes of 76HS, 15 eyes of 15 KC patients and 18 eyes of 18 subjects that underwent REF. Each participant underwent a complete ophthalmic evaluation, IOP measurement with GAT, DCT, ORA and CST. RESULTS: HS showed a mean GAT value of 15.62±2.33 mm Hg, a mean DCT value of 17.44±2.51 mm Hg, a mean ORA value of 15.99±3.58 mm Hg and a mean CST value of 17.24±3.44 mm Hg. KC showed a mean GAT value of 15.07±1.83 mm Hg, a mean DCT value of 17.01±1.96 mm Hg, a mean ORA value of 13.58±2.99 mm Hg and a mean CST value of 14.37±1.89 mm Hg. REF showed a mean GAT value of 14.06±1.51 mm Hg, a mean DCT value of 15.12±2.34 mm Hg, a mean ORA value of 16.85±2.4 mm Hg and a mean CST value of 15.57±1.77 mm Hg. CONCLUSION: Our data suggest that ORA and GAT could be used interchangeably in HS; GAT, ORA and CST could be used interchangeably in KC patients and that GAT provides lower IOP values compared to the other devices in eyes previously submitted to myopic PRK.

Central corneal thickness evaluation in healthy eyes with three different optical devices.

PURPOSE: To compare corneal pachymetry values measured by three different optical devices: Orbscan II, Pentacam HR and Sirius in healthy eyes. METHODS: The central corneal thickness (CCT) of 102 eyes of 102 healthy subjects (mean age of 33.09 ± 8.72 years and mean refractive defect -4.11 ± 4.74 D) was measured by three different physicians using Orbscan II, Pentacam HR and Sirius. The normality of the distribution was evaluated by with Kolmogorov-Smirnov test. The correlations between CCT obtained from each device and refractive defect and age were evaluated using the Pearson test. The differences were evaluated by the Student paired t-test using SPSS 18.0 (IBM Corp. Armonk, New York). RESULTS: Orbscan II provided significant (p < 0.0001) lower CCT measurements then both Pentacam HR (-13.66 ± 16.53 µm) and Sirius (-15.18 ± 17.16 µm); Sirius showed values slightly higher than Oculus Pentacam HR (+1.52 ± 6.21 µm) that appeared to be statistically significant (p < 0.015). CONCLUSIONS: The measurement of CCT by Sirius and Pentacam HR provides similar results. By contrast, the results obtained by Orbscan II are different from those obtained from both Sirius and Pentacam HR.

Comparison between Corvis and other tonometers in healthy eyes.

PURPOSE: To determine the agreement of intraocular pressure (IOP) values in healthy eyes among Goldmann applanation tonometer, dynamic contour tonometer, ocular response analyzer and Corvis. Additionally, to study the relationship between their differences with central corneal thickness (CCT) and corneal curvature (CK). METHODS: Seventy-six eyes of 76 healthy subjects were examined. Every subject underwent a complete ophthalmic evaluation, a Pentacam scan and three consecutive IOP measurements with each instrument (DCT, GAT, ORA and CST). IOP measurements provided by each device were compared with each other and the differences between them were correlated with morphological parameters obtained by Pentacam (CCT and CK). Statistical analysis was performed using SPSS software, version 18.0. RESULTS: The mean age of enrolled subjects was 36.8 ± 10.6 years old. The mean IOP measurements that were obtained with GAT, DCT, ORA and CST was 15.62 ± 2.33 mmHg, 17.44 ± 2.51 mmHg, 15.99 ± 3.58 mmHg and 17.24 ± 3.44 mmHg respectively. The mean CCT was 543.63 ± 36.15 µm, the mean CK was 43.35 ± 1.23 D. GAT and ORA provided IOP values not showing a statistical difference; CST and DCT IOP measurements did not show a statistical difference whereas CST provided statistically higher IOP values both than GAT and both ORA. CONCLUSIONS: According to our data, CST produces IOP values that are notably higher than GAT measures; therefore they cannot be used interchangeably. If CST should be used as the next gold standard, higher IOP values will come to be considered normal.

Evaluation of corneal deformation analyzed with a Scheimpflug based device.

PURPOSE: To evaluate the correlation between corneal biomechanical and morphological data in healthy eyes. METHODS: A complete clinical eye examination of naïve eyes was followed by tomographic (Pentacam, Oculus, Wetzlar, Germany) and biomechanical (Corvis ST, Oculus, Wetzlar, Germany) evaluation. Linear regression between central corneal thickness (CCT), corneal volume (CV) and anterior corneal curvature measured with Sim'K (SK), versus corneal deformation parameters measured with Corvis ST have been run using SPSS software version 18.0. RESULTS: Seventy-six eyes of 76 healthy subjects (44 women and 32 men) with a mean age of 36.84 ± 10.74 years and a mean refractive error of -0.55 ± 1.68 D (measured as spherical equivalent) were evaluated. Corneal deformation parameters were weakly correlated with corneal morphological parameters and with spherical equivalent. Although the correlations between deformation amplitude versus SK and between SK versus Velocity of Applanation 2, were higher than the others (R(2) = 0.28 and 0.26 respectively), none of them was statistically significant (p>0.01). CONCLUSIONS: According with these findings, Corvis ST seems to be able to provide an analysis of corneal deformation independent from corneal morphological characteristics. If these data will be confirmed in further studies, this device could be useful in the management and screening of eyes with corneal diseases.

Evaluation of corneal deformation analyzed with Scheimpflug based device in healthy eyes and diseased ones.

This study was designed to evaluate the correlation between corneal biomechanical and morphological data in healthy eyes, eyes that underwent myopic photorefractive keratectomy (PRK), keratoconus affected eyes, and keratoconus affected eyes that underwent corneal collagen crosslinking (CCC). Complete clinical eye examination of all eyes was followed by tomographic (Pentacam, Oculus, Wetzlar, Germany) and biomechanical (Corvis ST, Oculus, Wetzlar, Germany) evaluation. Differences among Corvis ST (CST) parameters in the different groups have been performed. Linear regression between central corneal thickness (CCT), intraocular pressure (IOP), and anterior corneal curvature measured with Sim'K (KM), versus corneal deformation parameters measured with Corvis ST in the different groups, has been run using SPSS software version 18.0. We evaluated 64 healthy eyes of 64 patients with a mean refractive error of -0.65 ± 1.68 D (measured as spherical equivalent), 17 eyes of 17 patients that underwent myopic PRK for a mean refractive defect of -4.91 ± 2.05 D (measured as spherical equivalent), 16 eyes of 16 patients affected by keratconus (stage 2-3 of Amsler Classification), and 13 eyes of 13 patients affected by keratoconus that underwent CCC. Our data suggest that corneal curvature would have a greater influence on corneal deformation than CCT; in fact KM values are more strongly associated with more CST parameters both about corneal change in shape and both about the corneal ability to come back at original shape.