Indian Diabetic Retinopathy Image Dataset (IDRiD)

Citation Author(s):
Prasanna
Porwal
Samiksha
Pachade
Ravi
Kamble
Manesh
Kokare
Girish
Deshmukh
Vivek
Sahasrabuddhe
Fabrice
Meriaudeau
Submitted by:
Prasanna Porwal
Last updated:
Tue, 11/12/2019 - 10:38
DOI:
10.21227/H25W98
Data Format:
Links:
License:
Creative Commons Attribution
4.8
5 ratings - Please login to submit your rating.

Abstract 

Diabetic Retinopathy is the most prevalent cause of avoidable vision impairment, mainly affecting working age population in the world. Recent research has given a better understanding of requirement in clinical eye care practice to identify better and cheaper ways of identification, management, diagnosis and treatment of retinal disease. The importance of diabetic retinopathy screening programs and difficulty in achieving reliable early diagnosis of diabetic retinopathy at a reasonable cost needs attention to develop computer-aided diagnosis tool. Computer-aided disease diagnosis in retinal image analysis could ease mass screening of population with diabetes mellitus and help clinicians in utilizing their time more efficiently. The recent technological advances in computing power, communication systems, and machine learning techniques provide opportunities to the biomedical engineers and computer scientists to meet the requirements of clinical practice. Diverse and representative retinal image sets are essential for developing and testing digital screening programs and the automated algorithms at their core. To the best of our knowledge, the database for this challenge, IDRiD (Indian Diabetic Retinopathy Image Dataset), is the first database representative of an Indian population. Moreover, it is the only dataset constituting typical diabetic retinopathy lesions and also normal retinal structures annotated at a pixel level. This dataset provides information on the disease severity of diabetic retinopathy, and diabetic macular edema for each image. This makes it perfect for development and evaluation of image analysis algorithms for early detection of diabetic retinopathy.

This dataset was available as a part of "Diabetic Retinopathy: Segmentation and Grading Challenge" organised in conjuction with IEEE International Symposium on Biomedical Imaging (ISBI-2018), Washington D.C.

Instructions: 
The dataset is divided into three parts:
A. Segmentation: It consists of
1. Original color fundus images (81 images divided into train and test set - JPG Files)
2. Groundtruth images for the Lesions (Microaneurysms, Haemorrhages, Hard Exudates and Soft Exudates divided into train and test set - TIF Files) and Optic Disc (divided into train and test set - TIF Files)
B. Disease Grading: it consists of
1. Original color fundus images (516 images divided into train set (413 images) and test set (103 images) - JPG Files)
2. Groundtruth Labels for Diabetic Retinopathy and Diabetic Macular Edema Severity Grade (Divided into train and test set - CSV File)
C. Localization: It consists of
1. Original color fundus images (516 images divided into train set (413 images) and test set (103 images) -
JPG Files)
2. Groundtruth Labels for Optic Disc Center Location (Divided into train and test set - CSV File)
3. Groundtruth Labels for Fovea Center Location (Divided into train and test set - CSV File)
 
For more information visit idrid.grand-challenge.org

Comments

2. Groundtruth Labels for Diabetic Retinopathy and Diabetic Macular Edema Severity Grade (Divided into train and test set - CSV File
Cannot be located 
please share csv files 

Submitted by Siddharth Chabukswar on Tue, 03/05/2019 - 09:50

Hi, Once you log-in to IEEE Dataport, just search for "DATASET FILES" and from there you should be able to download files easily.

Based on your message, I further crosschecked the folders that are available for download. I found all CSV files are available as uploaded to this repository and everything looks fine at our side.

Please let me know if you're still unable to locate or download.

Submitted by Prasanna Porwal on Wed, 03/13/2019 - 07:15

Hello, I could not find the link to download the IDRid dataset,. If you could like to provide the download link, I would very appreciate it. Thank you!

Submitted by Yanmin Xiong on Mon, 11/25/2019 - 05:19

Hello, Please follow directions in the following video to download the IDRiD dataset. Video Link: https://www.youtube.com/watch?v=PFk5fU4u0QY

If you are still having any difficulty to download the data then please send me an email at porwalprasanna@sggs.ac.in

Submitted by Prasanna Porwal on Tue, 12/03/2019 - 23:38

Sir I have downloaded the dataset for academic purpose but in DR grading folder ,label in Training csv file in Ground Truth folder is missing for some of the images that is present in Training images folder,thatswhy labeling the images becomes difficult.Kindly Provide me full dataset

 

Submitted by Abhilasha Yadav on Thu, 06/20/2024 - 07:46

Hi,

There were Online and On-Site tests in the original challenge. Just want to know how the current Train and Test sets were used in the challenge? Were they used for Online or On-Site testing?

Submitted by mengliu zhao on Thu, 06/06/2019 - 17:15

Complete details of the IDRiD dataset and the first results based on it are available in challenge summary paper. Link: https://www.sciencedirect.com/science/article/abs/pii/S1361841519301033

Submitted by Prasanna Porwal on Mon, 10/21/2019 - 11:03

According to which basis the images are classified into level 0,1,2,3,4?

Submitted by Anwayee Roy Cho... on Fri, 07/05/2019 - 07:23

Complete details of the IDRiD dataset and the first results based on it are available in challenge summary paper. Link: https://www.sciencedirect.com/science/article/abs/pii/S1361841519301033

Submitted by Prasanna Porwal on Mon, 10/21/2019 - 11:04

Accoding to csv file,what is grade 4 indicate?

Submitted by suganya B on Mon, 10/21/2019 - 05:43

Complete details of the IDRiD dataset and the first results based on it are available in challenge summary paper. Link: https://www.sciencedirect.com/science/article/abs/pii/S1361841519301033

Submitted by Prasanna Porwal on Mon, 10/21/2019 - 11:00

Thank you for providing this wonderful dataset.  I could not find the test set annotations for the Lesion Segmentation Sub-ChallengeAre these available anywhere or could you release them?

 

Test set annotations are necessary in order to compare against the leaderboard and more recent works using this dataset, and also useful for the research community.

 

Thank you, Alex

Submitted by Alex Gaudio on Mon, 12/02/2019 - 13:19

Hello Alex,

Based on your message, I crosschecked the folders that are available for download. 

I found all that all test set ground-truth files are available as uploaded to this repository and everything looks fine at our side.

See the video to locate files: https://youtu.be/PFk5fU4u0QY. Please let me know if you're still unable to locate or download the data by sending an email at porwalprasanna@sggs.ac.in.

Best,

Prasanna

Submitted by Prasanna Porwal on Tue, 12/03/2019 - 11:16

Thank you for your prompt response!   I downloaded again and can confirm I see the test files.  My apologies for causing any inconvenience.  The video link was invalid, but this works for me: https://www.youtube.com/watch?v=PFk5fU4u0QY

Submitted by Alex Gaudio on Tue, 12/03/2019 - 14:42

My pleasure! Happy to know that you have now access to test data as well. Thank you for the correction in the weblink.

Submitted by Prasanna Porwal on Tue, 12/03/2019 - 23:49

Excuse me I can not download this dataset

Please help 

 

Submitted by Pin Kuntha on Thu, 01/30/2020 - 07:47

If you already have an IEEE account then please follow directions in the video to download the IDRiD dataset. Video Link: https://www.youtube.com/watch?v=PFk5fU4u0QY

If you don't have it follow the steps to create an account for accessing data from IEEE Dataport:

1. Go to https://ieee-dataport.org/join

2. Click to Create a new IEEE account

3. Fill in Name, email ID and password for your account.

4. Now, Login and download the data by following directions in the video link provided above. 

Submitted by Prasanna Porwal on Thu, 01/30/2020 - 08:43

Can i use this dataset for academis purpose. 

Submitted by Anas Bilal on Thu, 06/18/2020 - 06:25

Yes, you can use it for academic purposes.

Submitted by Prasanna Porwal on Mon, 06/22/2020 - 06:58

Hello Sir, I want a csv dataset files for working in machine learning, Give the details of it

Submitted by kalpana devi on Wed, 02/03/2021 - 01:33

HOW CAN I GET THE LABEL FOR THE IMAGES?

Submitted by ANITHA NAIR on Fri, 05/14/2021 - 05:23

good afternoon, sir, I downloaded the database which includes original and ground truth images. Can you provide labelled ground truth images and label Ids for semantic segmentation? please do needful. My mail ID :ramakn@gmail.com

Submitted by Ramakrishna Nib... on Wed, 05/19/2021 - 07:05

HOW CAN I GET THE LABEL FOR THE IMAGES?

Submitted by Ahmed Galal on Tue, 03/29/2022 - 07:16

How can I get this dataset

Submitted by SANTOSH SHARMA on Tue, 08/23/2022 - 23:05

if i'm gonna make a binary classification how can i use the jpeg images ?

Submitted by sadok zdig on Sun, 04/16/2023 - 21:21

Sir
Is segmentation data (81)is a subset of classification data(516) ?Is it possible to get the severity(mild,moderate,severe) level of this segmentation data?

Submitted by ANITHA NAIR on Tue, 01/02/2024 - 01:01

Dataset Files

LOGIN TO ACCESS DATASET FILES
Open Access dataset files are accessible to all logged in  users. Don't have a login?  Create a free IEEE account.  IEEE Membership is not required.