scholarly journals Ray tracing intraocular lens calculation performance improved by AI-powered postoperative lens position prediction

2021 ◽  
pp. bjophthalmol-2021-320283
Author(s):  
Tingyang Li ◽  
Aparna Reddy ◽  
Joshua D Stein ◽  
Nambi Nallasamy
Keyword(s):  
Machine Learning ◽  
Ray Tracing ◽  
Anterior Chamber Depth ◽  
Prediction Method ◽  
Absolute Error ◽  
Power Calculation ◽  
University Of Michigan ◽  
Mean Error ◽  
Lens Position ◽  
Improved Performance

AimsTo assess whether incorporating a machine learning (ML) method for accurate prediction of postoperative anterior chamber depth (ACD) improves cataract surgery refraction prediction performance of a commonly used ray tracing power calculation suite (OKULIX).Methods and analysisA dataset of 4357 eyes of 4357 patients with cataract was gathered at the Kellogg Eye Center, University of Michigan. A previously developed machine learning (ML)–based method was used to predict the postoperative ACD based on preoperative biometry measured with the Lenstar LS900 optical biometer. Refraction predictions were computed with standard OKULIX postoperative ACD predictions and ML-based predictions of postoperative ACD. The performance of the ray tracing approach with and without ML-based ACD prediction was evaluated using mean absolute error (MAE) and median absolute error (MedAE) in refraction prediction as metrics.ResultsReplacing the standard OKULIX postoperative ACD with the ML-predicted ACD resulted in statistically significant reductions in both MAE (1.7% after zeroing mean error) and MedAE (2.1% after zeroing mean error). ML-predicted ACD substantially improved performance in eyes with short and long axial lengths (p<0.01).ConclusionsUsing an ML-powered postoperative ACD prediction method improves the prediction accuracy of the OKULIX ray tracing suite by a clinically small but statistically significant amount, with the greatest effect seen in long eyes.

scholarly journals AI-powered effective lens position prediction improves the accuracy of existing lens formulas

2021 ◽  
pp. bjophthalmol-2020-318321
Author(s):  
Tingyang Li ◽  
Joshua Stein ◽  
Nambi Nallasamy
Keyword(s):  
Anterior Chamber Depth ◽  
Absolute Error ◽  
University Of Michigan ◽  
Lens Position ◽  
Testing Dataset ◽  
Iol Power ◽  
Position Prediction ◽  
The Mean ◽  
Postoperative Anterior Chamber Depth ◽  
Testing Set

AimsTo assess whether incorporating a machine learning (ML) method for accurate prediction of postoperative anterior chamber depth (ACD) improves the refraction prediction performance of existing intraocular lens (IOL) calculation formulas.MethodsA dataset of 4806 patients with cataract was gathered at the Kellogg Eye Center, University of Michigan, and split into a training set (80% of patients, 5761 eyes) and a testing set (20% of patients, 961 eyes). A previously developed ML-based method was used to predict the postoperative ACD based on preoperative biometry. This ML-based postoperative ACD was integrated into new effective lens position (ELP) predictions using regression models to rescale the ML output for each of four existing formulas (Haigis, Hoffer Q, Holladay and SRK/T). The performance of the formulas with ML-modified ELP was compared using a testing dataset. Performance was measured by the mean absolute error (MAE) in refraction prediction.ResultsWhen the ELP was replaced with a linear combination of the original ELP and the ML-predicted ELP, the MAEs±SD (in Diopters) in the testing set were: 0.356±0.329 for Haigis, 0.352±0.319 for Hoffer Q, 0.371±0.336 for Holladay, and 0.361±0.331 for SRK/T which were significantly lower (p<0.05) than those of the original formulas: 0.373±0.328 for Haigis, 0.408±0.337 for Hoffer Q, 0.384±0.341 for Holladay and 0.394±0.351 for SRK/T.ConclusionUsing a more accurately predicted postoperative ACD significantly improves the prediction accuracy of four existing IOL power formulas.

scholarly journals AI-Powered Effective Lens Position Prediction Improves the Accuracy of Existing Lens Formulas

2020 ◽  
Author(s):  
Tingyang Li ◽  
Joshua D. Stein ◽  
Nambi Nallasamy
Keyword(s):  
Anterior Chamber Depth ◽  
Absolute Error ◽  
University Of Michigan ◽  
Lens Position ◽  
Testing Dataset ◽  
Iol Power ◽  
Position Prediction ◽  
The Mean ◽  
Postoperative Anterior Chamber Depth ◽  
Testing Set

ABSTRACTAimsTo assess whether incorporating a machine learning (ML) method for accurate prediction of postoperative anterior chamber depth (ACD) improves the refraction prediction performance of existing intraocular lens (IOL) calculation formulas.MethodsA dataset of 4806 cataract patients were gathered at the Kellogg Eye Center, University of Michigan, and split into a training set (80% of patients, 5761 eyes) and a testing set (20% of patients, 961 eyes). A previously developed ML-based method was used to predict the postoperative ACD based on preoperative biometry. This ML-based postoperative ACD was integrated into new effective lens position (ELP) predictions using regression models to rescale the ML output for each of four existing formulas (Haigis, Hoffer Q, Holladay, and SRK/T). The performance of the formulas with ML-modified ELP was compared using a testing dataset. Performance was measured by the mean absolute error (MAE) in refraction prediction.ResultsWhen the ELP was replaced with a linear combination of the original ELP and the ML-predicted ELP, the MAEs ± SD (in Diopters) in the testing set were: 0.356 ± 0.329 for Haigis, 0.352 ± 0.319 for Hoffer Q, 0.371 ± 0.336 for Holladay, and 0.361 ± 0.331 for SRK/T which were significantly lower than those of the original formulas: 0.373 ± 0.328 for Haigis, 0.408 ± 0.337 for Hoffer Q, 0.384 ± 0.341 for Holladay, and 0.394 ± 0.351 for SRK/T.ConclusionUsing a more accurately predicted postoperative ACD significantly improves the prediction accuracy of four existing IOL power formulas.

Machine learning-based prediction of component self-alignment in vapour phase and infrared soldering

2019 ◽  
Vol 31 (3) ◽  
pp. 163-168 ◽  
Author(s):  
Oliver Krammer ◽  
Péter Martinek ◽  
Balazs Illes ◽  
László Jakab
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Artificial Neural Network ◽  
Prediction Method ◽  
Absolute Error ◽  
Vapour Phase ◽  
The Self ◽  
Phase Method ◽  
Content Type ◽  
Artificial Neural

Purpose This paper aims to investigate the self-alignment of 0603 size (1.5 × 0.75 mm) chip resistors, which were soldered by infrared or vapour phase soldering. The results were used for establishing an artificial neural network for predicting the component movement during the soldering. Design/methodology/approach The components were soldered onto an FR4 testboard, which was designed to facilitate the measuring of the position of the components both prior to and after the soldering. A semi-automatic placement machine misplaced the components intentionally, and the self-alignment ability was determined for soldering techniques of both infrared and vapour phase soldering. An artificial neural network-based prediction method was established, which is able to predict the position of chip resistors after soldering as a function of component misplacement prior to soldering. Findings The results showed that the component can self-align from farer distances by using vapour phase method, even from relative misplacement of 50 per cent parallel to the shorter side of the component. Components can self-align from a relative misplacement only of 30 per cent by using infrared soldering method. The established artificial neural network can predict the component self-alignment with an approximately 10-20 per cent mean absolute error. Originality/value It was proven that the vapour phase soldering method is more stable from the component’s self-alignment point of view. Furthermore, machine learning-based predictors can be applied in the field of reflow soldering technology, and artificial neural networks can predict the component self-alignment with an appropriately low error.

scholarly journals Improving the prediction of effective lens position for intraocular lens power calculations

2020 ◽  
Vol 17 (2) ◽  
pp. 233-242
Author(s):  
Juanita Noeline Chui ◽  
Keith Ong
Keyword(s):  
Cataract Surgery ◽  
Intraocular Lens ◽  
Sagittal Plane ◽  
Anterior Chamber Depth ◽  
Posterior Chamber ◽  
Power Calculation ◽  
Iol Power Calculation ◽  
Lens Position ◽  
Iol Power ◽  
Corneal Apex

Purpose: Achieving the desired post-operative refraction in cataract surgery requires accurate calculations for intraocular lens (IOL) power. Latest-generation formulae use anterior-chamber depth (ACD)—the distance from the corneal apex to the anterior surface of the lens—as one of the parameters to predict the post-operative IOL position within the eye, termed the effective lens position (ELP). Significant discrepancies between predicted and actual ELP result in refractive surprise. This study aims to improve the predictability of ELP. We hypothesise that predictions based on the distance from the corneal apex to the mid-sagittal plane of the cataractous lens would more accurately reflect the position of the principal plane of the non-angulated IOL within the capsular bag. Accordingly, we propose that predictions derived from ACD + ½LT (length thickness) would be superior to those from ACD alone. Design: Retrospective cohort study, comparing ELP predictions derived from ACD to aproposed prediction parameter. Method: This retrospective study includes data from 162 consecutive cataract surgery cases, with posterior-chamber IOL (AlconSN60WF) implantation. Pre- and postoperative biometric measurements were made using the IOLMaster700 (ZEISS, Jena, Germany). The accuracy and reliability of ELP predictions derived from ACD and ACD + ½LT were compared using software-aided analyses. Results: An overall reduction in average ELP prediction error (PEELP) was achieved using the proposed parameter (root-mean-square-error [RMSE] = 0.50 mm), compared to ACD (RMSE = 1.57 mm). The mean percentage PEELP, comparing between eyes of different axial lengths, was 9.88% ± 3.48% and −34.9% ± 4.79% for predictions derived from ACD + ½LT and ACD, respectively. A 44.10% ± 5.22% mean of differences was observed (p < 0.001). Conclusion: ACD + ½LT predicts ELP with greater accuracy and reliability than ACD alone; its use in IOL power calculation formulae may improve refractive outcomes.

Anterior chamber depth, lens thickness and intraocular lens calculation formula accuracy: nine formulas comparison

2020 ◽  
pp. bjophthalmol-2020-317822
Author(s):  
Diogo Hipólito-Fernandes ◽  
Maria Elisa Luís ◽  
Rita Serras-Pereira ◽  
Pedro Gil ◽  
Vitor Maduro ◽  
...  
Keyword(s):  
Anterior Chamber ◽  
Intraocular Lens ◽  
Prediction Error ◽  
Anterior Chamber Depth ◽  
Case Series ◽  
Absolute Error ◽  
Power Calculation ◽  
Retrospective Case ◽  
Lens Thickness ◽  
Mean Prediction Error

Background/AimsTo investigate the influence of anterior chamber depth (ACD) and lens thickness (LT) on 9 intraocular lens (IOL) power calculation formulas accuracy, in patients with normal axial lengths.MethodsRetrospective case series, including patients having uncomplicated cataract surgery with insertion of a single IOL model, divided into three groups according to preoperative ACD. Each group was further subdivided into three subgroups, according to the LT. Using optimised constants, refraction prediction error was calculated for Barrett Universal II, Emmetropia Verifying Optical (EVO) V.2.0, Haigis, Hill-RBF V.2.0, Hoffer Q, Holladay 1, Kane, PEARL-DGS and SRK/T formulas. Mean prediction error, mean and median absolute error (MedAE) and the percentage of eyes within ±0.25D, ±0.50D and ±1.00D were also calculated.ResultsThe study included 695 eyes from 695 patients. For ACD ≤3.0 mm and ≥3.5 mm, mean prediction error of SRK/T, Hoffer Q and Holladay 1 was significantly different from 0 (p<0.05). PEARL-DGS, Kane, EVO V.2.0 and Barrett Universal II were more accurate than the Hoffer Q in ACD ≤3.0 mm (p<0.05). Kane, PEARL-DGS, EVO V.2.0 and Barrett Universal II revealed the lowest variance of mean and MedAE by ACD and LT subgroup. Haigis and Hill-RBF V.2.0 were significantly influenced by LT, independently of the ACD, with a myopic shift with thin lenses and a hyperopic shift with thick lenses (p<0.05).ConclusionNew generation formulas, particularly Kane, PEARL-DGS and EVO V.2.0, seem to be more reliable and stable even in eyes with extreme ACD-LT combinations.

scholarly journals Correlation between Axial Length and Anterior Chamber Depth in Eyes with High, Low and Average Axial Length

QJM ◽  
2020 ◽  
Vol 113 (Supplement_1) ◽  
Author(s):  
J H F Aziz ◽  
R M F Elghazawy ◽  
A I Elawamry ◽  
R G Zaki
Keyword(s):  
Anterior Chamber ◽  
Axial Length ◽  
Anterior Chamber Depth ◽  
Power Calculation ◽  
University Hospitals ◽  
Corneal Power ◽  
Lens Position ◽  
Study Results ◽  
Iol Power ◽  
Biometric Measurements

Abstract Background To achieve optimal refractive outcomes after cataract surgeries, accurate IOL power calculation is mandatory. IOL power is calculated using preoperative biometric measurements such as AL, corneal power, ACD and an estimation of postoperative effective lens position. Choosing the formula of IOL calculation depends mainly on AL and ACD. Objective The objective of this study is to investigate the correlation between AL and ACD in short eyes, normal eyes and long eyes. Patients and Methods The study was conducted at Ain Shams University Hospitals after the approval of the research ethical committee in the Faculty of Medicine, Ain Shams University. Study Period: 6 months (from July 2018 to January 2019). Results The results of our study showed that the correlation between the axial length and the anterior chamber depth among short eyes was statistically significant and they were negatively correlated, while no statistically significant correlation existed between AL and ACD in normal and long eyes. Discussion IOL power is calculated using preoperative biometric measurements such as AL, corneal power, ACD and an estimation of postoperative effective lens position. In the study by Holladay, 82% had normal AL and only 0.9% of patients had short eyes. Holladay et al. found that among patients with long eyes, 90% have normal ACD, 10% have high ACD and none of them have short ACD. In the group of patients with normal AL, they found that 90% of patients had normal ACD, none of them had low ACD and 10% had high ACD. In patients with short eyes they found that 20% have low ACD, 80% have normal ACD and none of them have high ACD. Sedaghat et al. documented that linear relationship was only seen in patients with normal AL and not in short and long sighted patients. Chang and Lau have concluded in their study that there was no statistically significant correlation between AL and ACD in eyes with AL of 27.5 mm or greater while positive statistically significant correlation existed in eyes shorter than 27.5 mm. Our study included ninety eyes of patients presenting for IOL or phakic lens implantation in Ain Shams University Hospitals. Patients were divided into three groups according to the AL; the short, normal and long eyes with AL ≤ 22 mm, &gt; 22 mm and &lt;24.50 mm and ≥24.50 mm, respectively. AL and ACD of each patient were measured using The ZEISS IOL Master 500. The results shows that the correlation between the axial length and the anterior chamber depth among short eyes is statistically significant and they are negatively correlated (r = -0.458. P value 0.011), while no statistically significant correlation exists between AL and ACD in normal and long eyes, i.e.: when AL &gt; 22 mm. This lack of correlation might influence the ELP predictions in third generation formulas, which do not consider the preoperative ACD. Our study results agree with most of the previous studies regarding long eyes &gt;27.5 mm that there is no statistically significant correlation between AL and ACD. While in short eyes it disagrees with Chang and Lau who concluded that there was no statistically significant correlation in short eyes and Sedaghat et al. who showed positive correlation between AL and ACD in short eyes. Our study results disagrees with most of the previous studies regarding normal eyes showing that there is no statistically significant correlation between AL and ACD in normal eyes, while previous studies concluded that a positive linear correlation exists. Conclusion The AL and ACD are inversely related in short eyes with AL ≤ 22 mm while no correlation exists in normal and long eyes with AL &gt; 22 mm.

scholarly journals Nationwide multicentre comparison of preoperative biometry and predictability of cataract surgery in Japan

2021 ◽  
pp. bjophthalmol-2021-318825
Author(s):  
Kazutaka Kamiya ◽  
Ken Hayashi ◽  
Mao Tanabe ◽  
Hitoshi Tabuchi ◽  
Masaki Sato ◽  
...  
Keyword(s):  
Cataract Surgery ◽  
Prediction Error ◽  
Corneal Thickness ◽  
Anterior Chamber Depth ◽  
Absolute Error ◽  
Power Calculation ◽  
Clinical Trial Registry ◽  
Biometric Data ◽  
Intraocular Lens Power Calculation ◽  
The Absolute

AimTo compare the preoperative biometric data and the refractive accuracy of cataract surgery among major surgical sites in a nationwide multicentre study.MethodsWe prospectively obtained the preoperative biometric data of 2143 eyes of 2143 consecutive patients undergoing standard cataract surgery at major 12 facilities and compared the preoperative biometry as well as the postoperative refractive accuracy among them.ResultsWe found significant differences in most preoperative variables, such as axial length (one-way analysis of variance, p=0.003), anterior chamber depth (p<0.001), lens thickness (p<0.001) and central corneal thickness (p<0.001), except for mean keratometry (p=0.587) and corneal astigmatism (p=0.304), among the 12 surgical sites. The prediction error using the Sanders-Retzlaff-Kraff/Theoretical (SRK/T formula was significantly more hyperopic than that using the Barrett Universal II formula (paired t-test, p<0.001). The absolute error using the SRK/T formula was significantly larger than that using the Barrett Universal II formula (p=0.016). The prediction error using the SRK/T formula was significantly more hyperopic than that using the Barrett Universal II formula at 10 of 12 institutions, but significantly more myopic at one institution. The absolute error using the SRK/T formula was significantly larger than that using the Barrett Universal II formula at 4 of 12 institutions but significantly smaller at two institutions.ConclusionsRegional divergences of the preoperative biometry were not necessarily negligible, and the optimised intraocular lens power calculation formula was individually different among the 12 facilities. Our findings highlight the importance of individual optimisation of these formulas at each facility, especially in consideration of these biometric variations.Trial registration numberClinical Trial Registry; 000039976.

scholarly journals Accuracy of intraocular lens power calculation using Scheimpflug tomography and OKULIX ray tracing software in corneal scarring

2019 ◽  
Author(s):  
Karim Mahmoud Nabil
Keyword(s):  
Visual Acuity ◽  
Ray Tracing ◽  
Intraocular Lens ◽  
Absolute Error ◽  
Power Calculation ◽  
Corneal Scarring ◽  
Iol Power Calculation ◽  
Intraocular Lens Power ◽  
Iol Power ◽  
Lens Power

Abstract Background: To evaluate the accuracy of intraocular lens power (IOL) calculation using Scheimpflug tomography and OKULIX ray tracing software in corneal scarring. Methods: This study was conducted on 40 consecutive eyes, 20 cases with corneal scarring and coexisting cataract, and 20 controls with clear cornea, which underwent uneventful phacoemulsification and IOL implantation following Scheimpflug tomography and OKULIX ray tracing software and third generation IOL power calculation formulas for IOL power calculation. Accuracy of IOL power calculation was evaluated by subtracting expected and achieved spherical refraction 3 months postoperatively and was recorded as mean absolute error (MAE). Distance uncorrected visual acuity (UCVA) for each eye was measured using Snellen chart preoperatively and 3 months postoperatively; visual acuity was scored and converted to the logarithm of the minimum angle of resolution (LogMar). Results: In cases of corneal scarring, 20 eyes (100 %) yielded a postoperative spherical refraction which differed less than 1 diopter (D) from predicted, in 16 eyes (80 %) the postoperative spherical refraction was within 0.50 D from expected. The MAE was 0.2 D in cases, which did not differ significantly compared to controls (MAE 0.1 D). In corneal scarring cases, distance UCVA showed significant improvement from 1.3 Log Mar (Snellen equivalent 20/400) preoperatively to 0.5 Log Mar (Snellen equivalent 20/60) 3 months postoperatively. Conclusion: Scheimpflug tomography combined with OKULIX ray tracing software for calculation of IOL power provides accurate results in cases of corneal scarring.

Comparison of 13 formulas for IOL power calculation with measurements from partial coherence interferometry

2020 ◽  
pp. bjophthalmol-2020-316193
Author(s):  
Giacomo Savini ◽  
Marco Di Maita ◽  
Kenneth J Hoffer ◽  
Kristian Næser ◽  
Domenico Schiano-Lomoriello ◽  
...  
Keyword(s):  
Anterior Chamber Depth ◽  
Case Series ◽  
Absolute Error ◽  
Partial Coherence ◽  
Power Calculation ◽  
Partial Coherence Interferometry ◽  
Iol Power Calculation ◽  
Iol Power ◽  
The Mean ◽  
The Difference

Background/aimsTo compare the accuracy of 13 formulas for intraocular lens (IOL) power calculation in cataract surgery.MethodsIn this retrospective interventional case series, optical biometry measurements were entered into these formulas: Barrett Universal II (BUII) with and without anterior chamber depth (ACD) as a predictor, EVO 2.0 with and without ACD as a predictor, Haigis, Hoffer Q, Holladay 1, Holladay 2AL, Kane, Næser 2, Pearl-DGS, RBF 2.0, SRK/T, T2 and VRF. The mean prediction error (PE), median absolute error (MedAE), mean absolute error and percentage of eyes with a PE within ±0.25, ±0.50, ±0.75 and ±1.00 diopters (D) were calculated.ResultsTwo hundred consecutive eyes were enrolled. With all formulas, the mean PE was zero. The BUII with no ACD had the lowest standard deviation (±0.343 D), followed by the T2 (0.347 D), Kane (0.348 D), EVO 2.0 with no ACD (0.348 D) and BUII with ACD (0.353 D) formulas. The difference among the MedAEs of all formulas was statistically significant (p<0.0001); the lowest values were achieved with the Kane (0.214 D), RBF 2.0 (0.215 D), BUII with and without ACD (0.218 D) and SRK/T (0.223 D). A percentage ranging from 80% to 88.5% of eyes showed a PE within ±0.50 D and all formulas achieved more than 50% of eyes with a PE within ±0.25 D.ConclusionAll investigated formulas achieved good results; there was a tendency towards better outcomes with newer formulas. Traditional formulas can still be considered an accurate option.

Sign in / Sign up

Export Citation Format

Share Document