This content is only available for ASCRS Members
This content from the 2020 ASCRS Virtual Annual Meeting is only available to ASCRS members. To log in, click the teal "Login" button in the upper right-hand corner of this page.
Purpose
To assess the accuracy of an iris registration algorithm of a femtosecond laser (CATALYS®, Johnson and Johnson Surgical Vision) for measuring ocular cyclotorsion after the eye has been docked with a liquid optics interface.
Methods
In a development data set of 52 eyes, we measured the amount of cyclotorsion from the preoperative measurements obtained with a color LED topographer (Cassini System, Cassini Technologies BV) and the intraoperative position when docked with the femtosecond laser. The development data set was used to refine the automated iris registration algorithm, and a verification data set of 50 eyes will be used to confirm its performance. The amount of cyclotorsion between the preoperative and intraoperative positions was manually compared by 3 image graders. The manually calculated cyclotorsion values were then compared to the automated cyclotorsion calculated by the Iris Registration algorithm.
Results
In the development data set, the arithmetic mean of cyclotorsion calculated by the automated system was -0.2° for right eyes and +3.1° for left eyes, with a range of +15° to -11°. The average standard deviation between the 3 manually graded cyclotorsion values for each eye was 0.57° degrees. The mean difference in cyclotorsion between automated calculation and the manual calculation was 0.71°; this difference was 2° or less for 98% of the eyes.
Conclusion
In our development data set, we found high agreement between the amount of cyclotorsion measured with manual and automated methods. We will report outcomes in a validation data set of 50 eyes.
To assess the accuracy of an iris registration algorithm of a femtosecond laser (CATALYS®, Johnson and Johnson Surgical Vision) for measuring ocular cyclotorsion after the eye has been docked with a liquid optics interface.
Methods
In a development data set of 52 eyes, we measured the amount of cyclotorsion from the preoperative measurements obtained with a color LED topographer (Cassini System, Cassini Technologies BV) and the intraoperative position when docked with the femtosecond laser. The development data set was used to refine the automated iris registration algorithm, and a verification data set of 50 eyes will be used to confirm its performance. The amount of cyclotorsion between the preoperative and intraoperative positions was manually compared by 3 image graders. The manually calculated cyclotorsion values were then compared to the automated cyclotorsion calculated by the Iris Registration algorithm.
Results
In the development data set, the arithmetic mean of cyclotorsion calculated by the automated system was -0.2° for right eyes and +3.1° for left eyes, with a range of +15° to -11°. The average standard deviation between the 3 manually graded cyclotorsion values for each eye was 0.57° degrees. The mean difference in cyclotorsion between automated calculation and the manual calculation was 0.71°; this difference was 2° or less for 98% of the eyes.
Conclusion
In our development data set, we found high agreement between the amount of cyclotorsion measured with manual and automated methods. We will report outcomes in a validation data set of 50 eyes.
View More Presentations from this Session
This presentation is from the session "SPS-103 Femtosecond Laser Assisted Cataract Surgery" from the 2020 ASCRS Virtual Annual Meeting held on May 16-17, 2020.