DistNet

Deep Bacteria Segmentation and Tracking with DistNet: Bacmman Module

This tutorial is based on the work described in this publication

Download example dataset and configure

1. If necessary, install bacmman. Install Tensorflow 2.
2. Download images from example dataset 1
3. If using bacmman for the first time: choose a working directory through right-click on the panel below Working Directory
4. From bacmman software, choose menu command: Dataset>New Dataset from github , set jeanollion as username in the Github Credentials panel, and select the configuration ExampleDatasets>dataset1_distnet
5. Download the trained model: choose menu command Import>DL Model library, set jeanollion as username in the Github Credentials panel. In the DL Models panel double-click on the model DistNet>DistNet_base_tf2>URL to download it. Unzip it and place it in a folder named model_bact_phase in the working directory (note: this directory can be changed in the configuration of the DLEngine the object class's tracker).

Run processing

The processing pipeline is the following:

1. Images are cropped during pre-processing
2. Microchannels are segmented
3. Bacteria are segmented and tracked in each microchannel

To run step 1) and 2): from the Home tab select Pre-Processing & Segment and Track in the Task panel and Microchannels in the Objects panel.

See this page to adapt DistNet to a dataset