Abstract:Background: Three-dimensional (3D) surface imaging and morphometrics are becoming the preferred craniofacial surface imaging modality. However, as a unique advantage of 3D imaging, areal and volumetric measurements have been rarely conducted and validated for evaluating soft tissue change in the periocular region, especially the upper eyelids. Therefore, based on an existing periocular landmark identification strategy, we proposed a novel modified method to define a standardized upper eyelid region for areal and volumetric measurements and validate its reliability for future clinical application.Methods: Forty-four healthy adult volunteers were enrolled in this cross-sectional study. Three-dimensional facial images were taken with a 3D imaging system. Subsequently, the upper eyelid region selection and areal and volumetric measurements were conducted using a modified landmarks localization strategy to evaluate their intrarater, interrater, and intramethod reliability.Results: Areal measurement of the upper eyelid revealed highly reliable outcomes for intraclass correlation coefficient (ICC), mean absolute difference (MAD), relative error measurement (REM), technical error of measurement (TEM), and relative technical error of measurement (%TEM) of 0.982, 0.1620 cm2, 2.9%, 0.1510 cm2, and 2.7% for intrarater reliability, respectively; 0.969, 0.2076 cm2, 3.7%, 0.1930 cm2, and 3.5% for interrater reliability, respectively; and 0.917, 0.3636 cm2, 6.5%, 0.3354 cm2, and 6.0% for intramethod reliability, respectively. Unsatisfactory results were found for the volumetric measurement of the upper eyelid: the ICC, MAD, REM, TEM, and %TEM estimates for intrarater reliability were 0.992, 0.2299 mL, 10.3%, 0.2414 mL, and 10.8%, respectively; for interrater reliability, these values were 0.985, 0.2749 mL, 12.3%, 0.3253 mL, and 14.6%, respectively; and for intramethod reliability, these values were 0.433, 1.6716 mL, 77.9%, 2.0615 mL, and 96.1%, respectively.Conclusions: This is the first study to propose a standardized upper eyelid region selection strategy and simultaneously validate its reliability for 3D areal and volumetric measurements. This study confirmed the high-level reliability of areal measurement and poor reliability of volumetric measurement based on direct measurements using a single image, which may provide better results when this method is combined with the image overlapping and registration procedure. However, this is subject to further validation. Nonetheless, this method could provide quantitative areal and volumetric data on the upper eyelids and might have widespread application potential in the future.

Artificial intelligence to automate assessment of ocular and periocular measurements

A Novel Automatic Morphologic Analysis of Eyelids Based on Deep Learning Methods

Automated Measurement of Ocular Movements Using Deep Learning-Based Image Analysis

A Novel Approach for Automated Eyelid Measurements in Blepharoptosis Using Digital Image Analysis

Automated Photographic Analysis of Inferior Oblique Overaction Based on Deep Learning.

A Fully Automatic Postoperative Appearance Prediction System for Blepharoptosis Surgery with Image-based Deep Learning

Toward an Automatic System for Computer-Aided Assessment in Facial Palsy

Automated 3D Perioral Landmark Detection Using High-Resolution Network: Artificial Intelligence-Based Anthropometric Analysis

Calculation of ophthalmic diagnostic parameters on a single eye image based on deep neural network

Reliability of Periocular Anthropometry Using Three-Dimensional Digital Stereophotogrammetry

Accuracy and reliability of automated three-dimensional facial landmarking in medical and biological studies. A systematic review

Ocular biomarkers: useful incidental findings by deep learning algorithms in fundus photographs

Automatic biometry of the anterior segment during accommodation imaged by optical coherence tomography

State-of-the-Art Periorbital Distance Prediction and Disease Classification Using Periorbital Features

Validation of the Portable Next-Generation VECTRA H2 3D Imaging System for Periocular Anthropometry

Fully automated landmarking and facial segmentation on 3D photographs

Smartphone-Acquired Anterior Segment Images for Deep Learning Prediction of Anterior Chamber Depth: A Proof-of-Concept Study

A novel standardized approach for the 3D evaluation of upper eyelid area and volume

Validation of an autonomous artificial intelligence-based diagnostic system for holistic maculopathy screening in a routine occupational health checkup context

Development of intra-oral automated landmark recognition (ALR) for dental and occlusal outcome measurements

Deep Learning for the Assessment of Facial Nerve Palsy: Opportunities and Challenges