Measurement And Prediction Of Tilt And Decentration Of A Novel Glistening-Free Hydrophobic Single-Piece Intraocular Lens Using A Machine Learning Algorithm

Aim of this study was to measure the post-operative intraocular lens (IOL) tilt and decentration and to predict IOL misalignment from pre-operative biometric data using a machine learning algorithm.

Department of Ophthalmology, Kepler University Clinic and Johannes Kepler University, Linz, Austria

This prospective non-randomized single-center study included patients scheduled for cataract surgery in both eyes. Assessments were performed preoperatively and 8 weeks postoperatively including optical biometry (IOLMaster 700, Carl Zeiss Meditec AG, Germany), high definition swept source OCT imaging (CASIA II, Tomey, Japan and MS-39, CSO, Italy) and Scheimpflug imaging (Pentacam HR, Oculus, Germany). Additionally, autorefraction, subjective refraction and visual acuity assessments for distance and intermediate vision were performed at the last follow-up.

In total, 100 eyes of 50 patients were included in the study. At the time of analysis 48 complete datasets were available. The mean age was 72.28 years (SD: 1.11). Axial eye length for right and left eyes was 23.12 mm (SD: 0.14) and 23.05 mm (SD: 0.14), respectively. There were no surgical, or post-surgical complications and the glistening grading was zero (no glistening) at all follow-ups. Mean post-operative tilt and decentration were 4.92 ° (SD: 0.81) and 0.26 mm (SD: 0.11), respectively.

The amount of post-operative tilt and decentration of the novel Clareon intraocular lens (Alcon, USA) was within the physiological range. The IOL showed a very good capsular bag stability. The machine learning-based prediction of post-operative tilt and decentration will be presented at the ESCRS in Barcelona.