Data on OCT and fundus images for the detection of glaucoma

This paper presents the data set of Optic coherence tomography (OCT) and fundus Images of human eye. The OCT machine TOPCON'S 3D OCT-1000 camera is employed to acquire the images. The dataset is comprised of 50 images which includes control and glaucomatous images. For each OCT Image there is a corresponding fundus Image with annotation. Cup to disc ratio (CDR) values annotated by glaucoma specialists through fundus Images are provided in excel file. OCT images are optic nerve head (ONH) centred. Manually annotation is performed for the delineation of the Inner Limiting Membrane (ILM) Layer and Retinal pigmented epithelium (RPE) layer with the help of ophthalmologist. The data is valuable for the development of automated algorithm for glaucoma diagnosis.


a b s t r a c t
This paper presents the data set of Optic coherence tomography (OCT) and fundus Images of human eye. The OCT machine TOP-CON'S 3D OCT-1000 camera is employed to acquire the images. The dataset is comprised of 50 images which includes control and glaucomatous images. For each OCT Image there is a corresponding fundus Image with annotation. Cup to disc ratio (CDR) values annotated by glaucoma specialists through fundus Images are provided in excel file. OCT images are optic nerve head (ONH) centred. Manually annotation is performed for the delineation of the Inner Limiting Membrane (ILM) Layer and Retinal pigmented epithelium (RPE) layer with the help of ophthalmologist. The data is valuable for the development of automated algorithm for glaucoma diagnosis. © 2020 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons. org/licenses/by/4.0/).

Data description
Fundus Images has been widely used for the initial examination of ophthalmic abnormalities. Ophthalmologist recommends appropriate treatment by observing subtle changes in ONH through fundus images [4]. OCT is relatively fastest imaging technique that provides the quantitative assessment of retinal layers [5]. OCT images are used to observe the morphological changes in retinal layers which provide the detail picture of ocular disease.
The data set include both controlled and glaucomatous case images of Fundus and OCT Images of humans. The data encompass is acquired from 26 subjects scanned on TOPCON'S 3D OCT system. It includes both eye data for 23 subjects and one eye data for three subjects. So, the presented data is comprised of 50 OCT and Fundus images, including 18 controlled and 32 glaucomatous affected cases. The subjects are selected in such a way that they form a diverse dataset as it includes male and female subjects belonging to different age groups. Local ethical committee approved data collection procedure and hospital ethical board approved data collection after getting consent from subjects and proper anonymization of data.
The provided OCT Images in data set are B-scan and ONH centred with resolution of 951 Â 456 [6]. Fig. 1 shows the fundus and its corresponding OCT image of a subject from the data set. OCT image is ONH centred; the retinal layers mostly considered for the detection of glaucoma are highlighted in Fig. 1(A). Whereas, Fig. 1(B) shows the fundus image of the same subject that helps in viewing the ONH abnormalities. As OCT images undergoes pre-processing steps for further analysis, so images are cropped and resized to original image 951 Â 456. Manually annotation is performed on rescaled image. Manually delineation of the Inner Limiting Membrane (ILM) Layer and Retinal pigmented epithelium (RPE) layer is achieved using Illustrator CS6 with the help of ophthalmologist. The Specifications Table   Subject Ophthalmology

Value of the Data
The data is valuable for the development and improvement of automated algorithm for glaucoma detection. The provided data is expedient for the analysis of retinal layers especially optic nerve head (ONH) region. The data can be useful for retinal layer analysis for other ocular diseases related to ONH. The data provides Optical coherence tomography (OCT) as well as fundus images for each subject which will help in automated correlation of finding from both image modalities.
manually outlined ILM and RPE layers through OCT Images of controlled and glaucomatous subject are shown in Fig. 2(B, E) and Fig. 2(C, F) respectively. For each OCT image we provided its Fundus image along with Cup to disc ratio (CDR) annotations performed by four glaucoma specialists. In addition, glaucoma specialist classifies the images into controlled, suspect and glaucoma through analysis of fundus and OCT images. The CDR annotation and labels for each image are provided in excel file. The fundus and OCT images for controlled and glaucomatous subjects are shown in Fig. 2(A, B, C, G) and Fig. 2(D, E, F, H) respectively. It is evident from Fig. 2 the cup size increased in glaucoma case thus as result CDR value also raised.

Experimental design, materials, and methods
The OCT machine TOPCON'S 3D OCT-1000 system is employed to acquire the OCT and fundus Images. Examination and image acquisition are performed after pupil dilations with Ø4.0mm (45 ) diameter. B-scan acquisition frequency is 5Hz thus reducing the eye movement effects. Scanning Speed is 27,000 to 50,000 A-scans per second with the depth of 2.3mm, B-scan is comprised of 1024 A-scans.  The Lateral and Vertical resolution are kept 5.9mm (±0.2) and 5.9mm (±0.2) respectively. The OCT scans of retina are optic nerve head (ONH) centred with 951 Â 456 resolution.
The data set had been used for the evaluation of automated segmentation algorithm for the extraction of retinal layers [1e3].