Novel Software Program to Improve Biometry Measurements Obtained by an Optical Low Coherence Reflectometry Biometer.

Abstract

Purpose: To evaluate a novel software program (SpikeFinder) that improves quality of measurements obtained by optical low coherence reflectometry (OLCR) biometry (Lenstar).

Setting: Tertiary care academic center.

Design: Retrospective case series.

Method: ology: After applying inclusion criteria, 12,988 eyes scanned with OLCR biometry and swept-source OCT (SS-OCT) biometry (IOLMaster700) were analyzed. Measurements, including axial length (AL), anterior chamber depth (ACD), and lens thickness (LT), were compared. SpikeFinder was applied to eyes with the most prominent ACD and LT measurement differences. Statistical testing was performed using Wilcoxon (two variables) and Kruskal-Wallis ANOVA tests (multiple variables) with Tukey post-hoc adjustments. Heteroscedastic testing (WHWK Statistics Package) was used to assess differences after software application.

Results: 99.1% of AL measurements were within 0.01 mm between the two biometers. SS-OCT LT was significantly longer than the OLCR LT: 33% differed by ≥ 0.50 mm (p<0.001). Correspondingly, SS-OCT ACD was significantly shorter than OLCR ACD: 14% differed by ≥ 0.50 mm (p<0.001). SS-OCT biometer - OLCR biometer values for ACD and LT before applying SpikeFinder were -0.64 ± 0.31 mm and 1.15 ± 0.38 mm, respectively (p<0.001). After applying SpikeFinder to a subgroup of 337 eyes, these differences were -0.02 ± 0.03 mm and 0.01 ± 0.11 mm, respectively. SpikeFinder ACD and LT values were similar to SS-OCT biometer-measured ACD (p=0.872) and LT (p=0.987) values.

Conclusion: The SpikeFinder program can help improve the accuracy of ACD and LT measurements obtained by the OLCR biometer, which may enhance the performance of formulas that utilize these measurements.