Skip to content

Latest commit

 

History

History
102 lines (76 loc) · 4.25 KB

BiDR.md

File metadata and controls

102 lines (76 loc) · 4.25 KB

BiDR

Dataset Information

The BiDR (Binary Diabetic Retinopathy) dataset is designed for the classification of Diabetic Retinopathy (DR). Each image in the dataset has been meticulously evaluated by medical professionals to determine the presence or absence of diabetic retinopathy. This dataset is significant for the early detection and grading of diabetic retinopathy. As the prevalence of diabetes continues to increase globally, there is an urgent need for efficient and accurate tools to assist medical professionals in screening and diagnosing this blinding disease. The BiDR dataset provides a valuable resource by offering a large number of precisely annotated retinal images, facilitating the development and testing of automated systems capable of performing these tasks with high accuracy and consistency. These systems have the potential to greatly improve early intervention strategies, thereby preventing vision loss and enabling personalized treatment plans for patients with diabetic retinopathy.

The labels in the dataset are binary, with 0 indicating the presence of diabetic retinopathy and 1 indicating its absence. The dataset comprises a total of 2838 entries, including 2076 images for training, 531 files for validation, and 231 files for testing.

Dataset Meta Information

Dimensions Modality Task Type Anatomical Area Number of Categories Data Volume File Format
2D fundus photography Classification Retina 2 2838 JPG

Resolution Details

Dataset Statistics size
min (224, 224)
median (224, 224)
max (224, 224)

Label Information Statistics

Classification No Diabetic Retinopathy Diabetic Retinopathy
Number of Images 1430 1408
Percentage 50.38% 49.62%
Average Volume (cm³) (224, 224) (224, 224)
Median Volume (cm³) (224, 224) (224, 224)
Total Volume (cm³) (224, 224) (224, 224)

Visualization

Diabetic Retinopathy.

No Diabetic Retinopathy.

File Structure

BIDR/
│
├── train/
│   ├── DR/
│   │   └── <JPG image file>
│   └── No_DR/
│       └── <JPG image file>
│
├── valid/
│   ├── DR/
│   │   └── <JPG image file>
│   └── No_DR/
│       └── <JPG image file>
│
└── test/
    ├── DR/
    │   └── <JPG image file>
    └── No_DR/
        └── <JPG image file>

Authors and Institutions

Parisa Karimi Darabi (K.N. Tuss University of Technology, Tehran)

Source Information

Official Website: https://www.kaggle.com/datasets/pkdarabi/diagnosis-of-diabetic-retinopathy?resource=download-directory

Download Link: https://www.kaggle.com/datasets/pkdarabi/diagnosis-of-diabetic-retinopathy?resource=download-directory

Article Address: https://www.researchgate.net/publication/382264856_Diagnosis_of_Diabetic_Retinopathy

Publication Date: 2023-11

Citation

@dataset{dataset,
author = {Darabi, Parisa},
year = {2024},
month = {07},
pages = {},
title = {Diagnosis of Diabetic Retinopathy},
doi = {10.13140/RG.2.2.13037.19688}
}

Original introduction article is here.