Fig 1.
Schematic drawing of eyelid parameters.
Table 1.
Patient demographics and severity.
Fig 2.
Eyelid measurements using AI-tool and Manual (ImageJ) tool.
(A) EER measurement using manual tool (ImageJ 1.46), (B) Eyelid parameters generated by the AI-tool (Anigma-View 1.0.6), (C) EER measurement using the AI-tool (Anigma-View 1.0.6).
Fig 3.
The image is localized to eye regions and segmentation label composes 6 different classes.
(A) Input image, (B) Segmentation image.
Fig 4.
EER values before and after surgery images from the same patient using AI-tool (Anigma-View 1.0.6).
(A) Before surgery EER value generated by the AI-tool, (B) After surgery EER value generated by the AI-tool.
Table 2.
Pre-operation and Post-operation repeated AI-based and Manual (ImageJ) measurements of EER.
Table 3.
The comparison of EER values based according to measurement method, with Pre- and Postoperative Averages for Total Eyes (n = 100).
Table 4.
Intraclass correlation coefficients(ICC) between two measurements of EER pre-operation and post-operation.
Fig 5.
Bland-Altman analysis between different EER measurements.
(A) Bland-Altman plot for AI-tool 1st and 2nd measurement, (B) Bland-Altman plot for AI-tool and Manual (ImageJ) measurement.
Table 5.
The comparison of EER values pre- and post-operation according to ptosis severity using 1st AI-tool.
Fig 6.
The comparison of EER values pre- and post-operation according to ptosis severity using 1st AI-tool.