Effective encoder-decoder network for pupil light reflex segmentation in facial photographs of ptosis patients

Accurate segmentation of pupil light reflexes is essential for the reliable assessment of ptosis severity, a condition characterized by the drooping of the upper eyelid. This study introduces a novel encoder-decoder network specialized in reflex segmentation by focusing on addressing issues related...

Full description

Saved in:
Bibliographic Details
Published in:Scientific reports Vol. 14; no. 1; pp. 26220 - 14
Main Authors: Lee, Sanghyuck, Song, Taekyung, Lee, Jeong Kyu, Lee, Jaesung
Format: Journal Article
Language:English
Published: London Nature Publishing Group UK 31.10.2024
Nature Publishing Group
Nature Portfolio
Subjects:
692/699/3161
692/699/3161/3166
692/699/3161/3173
Adult
Aged
Blepharoptosis - physiopathology
Clinical trials
Convolution
Design
Eyelid
Eyes & eyesight
Face - diagnostic imaging
Facial photograph
Female
Graves Ophthalmopathy - diagnostic imaging
Graves Ophthalmopathy - physiopathology
Humanities and Social Sciences
Humans
Image Processing, Computer-Assisted - methods
Light
Male
Middle Aged
multidisciplinary
Neural networks
Neural Networks, Computer
Paralysis
Photography - methods
Ptosis
Pupil light reflex
Reflex, Pupillary - physiology
Reflexes
Review boards
Science
Science (multidisciplinary)
Segmentation
Statistical analysis
Ptosis
Pupil light reflex
Segmentation
Facial photograph
ISSN:2045-2322, 2045-2322
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Accurate segmentation of pupil light reflexes is essential for the reliable assessment of ptosis severity, a condition characterized by the drooping of the upper eyelid. This study introduces a novel encoder-decoder network specialized in reflex segmentation by focusing on addressing issues related to very small regions of interest from an architectural perspective. Specifically, the proposed network is designed to exploit low-level features effectively by integrating a multi-level skip connection and a 1 × 1 convolution-enhanced initial encoding stage. Assessed using a photograph image dataset from Chung-Ang University Hospital, which includes 87 healthy subjects, 64 with ptosis, and 257 with Graves’ orbitopathy (collected between January 2010 and February 2023), the proposed network outperforms five conventional encoder-decoders. Over 30 trials, the proposed network achieved a mean Dice coefficient of 0.767 and an Intersection over Union of 0.653, indicating a statistically significant improvement in the segmentation of reflex. Our findings show that an elaborate design based on the lowest-level skip connection and 1 × 1 convolution at initial stage enhances the segmentation of pupil light reflexes. The source code of the proposed network is available at https://github.com/tkdgur658/ReflexNet .
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:2045-2322
2045-2322
DOI:10.1038/s41598-024-77001-9