Skip to content

Latest commit

 

History

History
94 lines (63 loc) · 4.14 KB

ACPS.md

File metadata and controls

94 lines (63 loc) · 4.14 KB

ACPS

Dataset Information

ACPS (Automatic Cone Photoreceptor Segmentation) is a dataset specifically designed for the task of automatic segmentation of cone photoreceptors. The dataset includes 840 images derived from the imaging results of the right eyes of 21 subjects. The average age of the 21 subjects is 25.9 ± 6.5 years, with one subject having green color blindness. The gold standard for each image is based on cone positions previously identified using semi-automatic detection methods, which were meticulously reviewed and corrected by experts. Additionally, any missing cones were manually annotated.

The significance of this dataset lies in its provision of a high-quality reference standard for the retinal cone photoreceptor locations, carefully reviewed and corrected by experts. It serves as a reliable benchmark for validating and optimizing the accuracy of automated cone detection algorithms. By comparing the results of automated algorithms with the manually corrected "gold standard" data, researchers can evaluate and improve the performance of these algorithms, thereby enhancing the precision of retinal image analysis. This dataset not only provides a foundation for validating automated image processing techniques but also lays a solid groundwork for research on retinal structures and the development of automated ophthalmic diagnostics.

Dataset Meta Information

Dimensions Modality Task Type Anatomical Structures Anatomical Area Number of Categories Data Volume File Format
2D Optical imaging Segmentation Cone photoreceptors Eye 1 840 .mat

Resolution Details

Dataset Statistics size
min 150x150
median 150x150
max 150x150

Label Information Statistics

Metric Cone Photoreceptor
Case Count 840
Coverage 100%

Visualization

The following figure shows the original image, the outline of the cone photoreceptor, and its center point from left to right:

File Structure

DAO-SLOCPASA
└── 2013_BOE_Chiu
    ├── Subject01_QuadrantBL_01.mat
    ├── Subject01_QuadrantBL_02.mat
    ├── Subject01_QuadrantBL_03.mat
    ├── Subject01_QuadrantBL_04.mat
    ├── Subject01_QuadrantBL_05.mat
    └── ...

Authors and Institutions

  • Stephanie J. Chiu (Duke University, Department of Biomedical Engineering)

  • Yuliya Lokhnygina (Duke University, Department of Biostatistics and Bioinformatics)

  • Adam M. Dubis (Duke University, Department of Ophthalmology)

  • Alfredo Dubra (Medical College of Wisconsin, Department of Ophthalmology)

  • Joseph Carroll (Medical College of Wisconsin, Department of Ophthalmology)

  • Joseph A. Izatt (Duke University, Department of Biomedical Engineering)

  • Sina Farsiu (Duke University, Department of Biomedical Engineering)

Source Information

Official Website: https://people.duke.edu/~sf59/Chiu_BOE_2013_dataset.htm

Download Link: https://people.duke.edu/~sf59/Chiu_BOE_2013_dataset.htm

Article Address: https://people.duke.edu/~sf59/Chiu_BOE_2013_dataset.htm

Publication Date: 2013

Citation

@article{chiu2013automatic,
  title={Automatic cone photoreceptor segmentation using graph theory and dynamic programming},
  author={Chiu, Stephanie J and Lokhnygina, Yuliya and Dubis, Adam M and Dubra, Alfredo and Carroll, Joseph and Izatt, Joseph A and Farsiu, Sina},
  journal={Biomedical optics express},
  volume={4},
  number={6},
  pages={924--937},
  year={2013},
  publisher={Optica Publishing Group}
}

Original introduction article is here.