Derive reference segmentations from segmentations of multiple experts.

1 - Expert Segmentations

Required Steps:

  1. Select parent folder containing sub-folders with segmentation masks, one folder per expert
  2. Click Load Data

Input Details: deepflash2 fuses

  • binary segmentations of an image, that is, there must be a single foreground value that represents positively classified pixels
    • Segmentation pixel values: background-class: 0; foreground-class: 1 or 255
  • instance segmentations of an image (instances represent positively classified pixels)
    • Segmentation pixel values: background-class: 0; foreground-instances: 1,2,...,I

Examplary input folder structure:

expert_segmentations  -> one parent folder
│                     
│───expert1           -> one folder per expert
│   │   mask1.png     -> segmentation masks
│   │   mask2.png
│   
└───expert2
    │   mask1.png
    │   mask2.png

All common image formats (tif, png, etc.) are supported. See imageio docs.

2 - Ground Truth Estimation

Required Steps:

  1. Click Run for STAPLE or Majority Voting
  • Simultaneous truth and performance level estimation (STAPLE). The STAPLE algorithm considers a collection of segmentations and computes a probabilistic estimate of the true segmentation and a measure of the performance level represented by each segmentation. Source: Warfield, Simon K., Kelly H. Zou, and William M. Wells. "Simultaneous truth and performance level estimation (STAPLE): an algorithm for the validation of image segmentation." IEEE transactions on medical imaging 23.7 (2004): 903-921
  • Majority Voting. Use majority voting to obtain the reference segmentation. Note that this filter does not resolve ties. In case of ties it will assign the indicated MV undecided label to the result.

3 - Expert Performance Scores

Required Steps:

  1. Results Table: Click Open and Update
    • Filter the and sort the results
    • Download the results