Dataset

The dataset consists of 16 EDF real cervical cytology images and 945 sythetic images. The qualitative assessment is performed on the 16 EDF real cervical cytology images. The 945 synthetic images are used in the quantitative evaluation of participants’ codes.

• January 15th, 2014: 45 training synthetic images, 90 testing synthetic images (with annotations), 8 EDF real cervical cytology images and evaluation code written by Matlab.

• February 15th, 2014: Rest 810 testing synthetic images (without annotations) and 8 EDF real cervical cytology images.

Participants are encouraged to evaluate their algorithms on the 45 training + 90 testing synthetic images released in the first phrase. They need to run their algorithms on the rest 810 synthetic images and submit both the codes and segmentations results on the 810 images for evaluation.

Please cite the dataset by the following papers:

• “Zhi Lu, Gustavo Carneiro, Andrew P. Bradley, Daniela Ushizima, Masoud S. Nosrati, Andrea G. C. Bianchi, Claudia M. Carneiro, and Ghassan Hamarneh. Evaluation of Three Algorithms for the Segmentation of Overlapping Cervical Cells. IEEE Journal of Biomedical and Health Informatics (J-BHI). Jan 2015 (Accepted).”

  
  

• “Zhi Lu, Gustavo Carneiro, and Andrew P. Bradley. An Improved Joint Optimization of Multiple Level Set Functions for the Segmentation of Overlapping Cervical Cells. IEEE Transactions on Image Processing. Vol.24, No.4, pp.1261-1272, April 2015.”

Download

• Training Phase:

  • 45 Training synthetic images, 90 Test synthetic images.

  • 8 EDF real cervical cytology images.

  • Evaluation Code.

• Test Phase:

  • 810 Test synthetic images.

  • another 8 EDF real cervical cytology images.

• Annotation

  • Annotation for Train and Test set

How to Use Evaluation Function?

The evaluateion code is used to quantitatively assess the segmentation performance on the synthetic images. It consists of three files:

• evaluateCytoSegmentation.m: Function to evaluate the result.

• SegEvaluateJIDiceTPRFPR.m: Function about the evaluation measurement, called by evaluateCytoSegmentation.m.

• CytoScriptExample.m: Example script to call the function evaluateCytoSegmentation.m .

You need to edit the script CytoScriptExample.m to evaluate the result. Details about the function evaluateCytoSegmentation.m can be found by “help evaluateCytoSegmentation”