• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.614-616
• /
• 2021

The world faces difficulties in terms of eye care, including treatment, quality of prevention, vision rehabilitation services, and scarcity of trained eye care experts. However, it is difficult to develop a method for classifying various ocular diseases because the existing dataset for retinal image disclosure does not consist of various diseases found in clinical practice. We propose a method for classifying ocular diseases using the Retinal Fundus Multi-disease Image Dataset (RFMiD), a dataset published in the ISBI-2021 challenge. Our goal is to develop a robust and generalizable model for screening retinal images into normal and abnormal categories. The performance of the proposed model shows a value of 0.9782 for the test dataset as an area under the curve (AUC) score.

• Journal of Korean Medical Science
• /
• v.33 no.43
• /
• pp.239.1-239.12
• /
• 2018

Background: We described a novel multi-step retinal fundus image reading system for providing high-quality large data for machine learning algorithms, and assessed the grader variability in the large-scale dataset generated with this system. Methods: A 5-step retinal fundus image reading tool was developed that rates image quality, presence of abnormality, findings with location information, diagnoses, and clinical significance. Each image was evaluated by 3 different graders. Agreements among graders for each decision were evaluated. Results: The 234,242 readings of 79,458 images were collected from 55 licensed ophthalmologists during 6 months. The 34,364 images were graded as abnormal by at-least one rater. Of these, all three raters agreed in 46.6% in abnormality, while 69.9% of the images were rated as abnormal by two or more raters. Agreement rate of at-least two raters on a certain finding was 26.7%-65.2%, and complete agreement rate of all-three raters was 5.7%-43.3%. As for diagnoses, agreement of at-least two raters was 35.6%-65.6%, and complete agreement rate was 11.0%-40.0%. Agreement of findings and diagnoses were higher when restricted to images with prior complete agreement on abnormality. Retinal/glaucoma specialists showed higher agreements on findings and diagnoses of their corresponding subspecialties. Conclusion: This novel reading tool for retinal fundus images generated a large-scale dataset with high level of information, which can be utilized in future development of machine learning-based algorithms for automated identification of abnormal conditions and clinical decision supporting system. These results emphasize the importance of addressing grader variability in algorithm developments.

• Korean Journal of Optics and Photonics
• /
• v.19 no.6
• /
• pp.439-447
• /
• 2008

There is a need for a finite schematic presbyopic eye that models vision and image quality under various conditions such as cataract or refractive surgery, as well as near vision corrections with an ophthalmic lens or contact lens. Using recently measured biometric data of presbyopic eyes, new model eyes were designed that are optically and anatomically close to real eyes. The parameters changing significantly with age were incorporated into models for four different age groups. The new model eyes have alpha angle, decentered pupil, aspheric GRIN lens and aspheric retinal surface. It is likely that the new finite presbyopic model eyes will be useful for designing visual instruments such as low vision aids, PALs, IOL and contact lenses, and for the clinical prediction of the retinal image quality of a presbyopic patient.

• Current Optics and Photonics
• /
• v.3 no.6
• /
• pp.597-604
• /
• 2019

This study was performed to evaluate objective visual quality and ocular scattering in myopic refractive-surgery patients, compared to emmetropes. Optical vision-quality parameters (modulation transfer function (MTF) cutoff and Strehl ratio) and objective scattering index (OSI) were measured using an optical quality analysis system (OQAS II) based on the double-pass technique. In all subjects, the higher the MTF cutoff and Strehl ratio, the lower the OSI and ocular higher-order aberrations (HOAs). The MTF cutoff and Strehl ratio for the laser-assisted subepithelial keratectomy (LASEK) group were lower than those for the emmetropia group, while the OSI, ocular HOAs, and spherical aberration (SA) for the LASEK group were higher than those for emmetropia group. Ocular scattering would be one of the important factors in regard to visual quality. Therefore, the quality of the retinal image in the LASEK patients has been shown to reduce the quality of vision more than in the emmetropes.

• Current Optics and Photonics
• /
• v.3 no.1
• /
• pp.72-77
• /
• 2019

This study was performed to analyze the optical aberrations of the cornea induced by myopic refractive surgery. Corneal total higher-order aberrations, spherical aberration and coma for 4-mm and 6-mm pupils were measured using a wave-front analyzer. The amount of aberrations of the oblate corneal optics by the achieved correction was found to be larger than for the prolate corneal shape with complete eye, in an emmetropia control group. The change in corneal shape acts as an optical factor that degrades the quality of the retinal image; it seems to be one of the important factors related to quality of vision.

• Korean Journal of Optics and Photonics
• /
• v.18 no.5
• /
• pp.337-344
• /
• 2007

In recent years, there has been rapid progress in different areas of vision science, such as refractive surgical procedures, contact lenses and spectacles, and near vision. This progress requires a highly accurate modeling of optical performance of the human eyes in different accommodation states. A new novel model-eye was designed using the Navarro accommodation-dependent finite model eye. For each of the vergence distances, ocular wavefront error, accommodative response, and visual acuity were calculated. Using the new model eye ocular wavefront error, accommodation dative response, and visual acuity are calculated for six vergence stimuli, -0.17D, 1D, 2D, 3D, 4D and -5D. Also, $3^{rd}\;and\;4^{th}$ order aberrations, modulation transfer function, and visual acuity of the accommodation-dependent model eye were analyzed. These results are well-matched to anatomical, biometric, and optical realities. Our corrected accommodation-dependent model-eye may provide a more accurate way to evaluate optical transfer functions and optical performances of the human eye.