The DDS4 Dataset is composed of 30300 digital fields images of 1040x1388 pixels from slides S16 to S25.

For each five diagnoses classes ( negative, scanty, 1+, 2+, and 3+), 50 sets were generated .

The sets are composed of digital field images and a text file that indicates the total bacilli in the set.

The text file is named ground_truth_x.txt, where x corresponds to the set number and can be a value between 0 and 49.

• Each set of negative, scanty, and 1+ diagnostics is composed of 164 digital fields and the ground truth.

• Each set of 2+ diagnostic is composed of 82 digital fields and ground truth.

• Each set of 3+ diagnostic is composed of 32 digital fields and the ground truth file. 


Figure 1. Example of digital field image of 1040x1388 pixels





S→ Slide
T→ Tile (a digital microscopy field)
TR→ Tile Region (a set of 164 Tiles or digital fields)
St→ Stack (a set of 11 digital images acquired from a microscopic field at different focal depths)
EDF→ Extended Depth of Focus image (a digital image obtained by the fusion of the stack of 11 images)
M→ Marking (Annotation)
NP→ Negative Patch
PP→ Positive Patch
MSC→  Mosaic Image


Figure 2. File Directory