Skip to main content

ARIP, software to quantify bacterial resistance to antibiotics by analysing picture of phenotypic plates

Project description

This software is aimed to quantify bacterial resistance to antibiotics by analysing pictures of phenotypic plates. Currently it supports 96 well plates where different bacteria are cultured with different concentrations of antibiotics, but the application adapt to different plates size in rows and columns. Computer vision algorithms have been implemented in order to detect different levels of bacterial growth. As a result, the software generates a report providing quantitative information for each well of the plate. Pictures should be taken so that the plate is square with the picture frame, the algorithm should be able to cope with a slight rotation of the plate.

Key methods:

  • Hough Circles method to detect circles in an image doc
  • Wells segmentation using threshold feature of opencv doc combining binary and otsu threshold
  • Quality detection using a grid model by rows and columns and clustering them, robust to scale and sensible rotation.

Execution:

There are two ways for executing the process: binary or library * Binary using arip.py file allocated in the project:

python arip.py --image images/\<platename\>.png
  • Library installing as described below:

    import arip
    arip.process({'image': 'images/sinteticplate.jpg'})
    

input:

images/<platename>.png with a plate and ninety six wells

output:

  • Image with extracted wells: images/<platename>/outputXXX.png
  • Cropped image of extracted well: images/<platename>/<row><column><resistance>_<density>.png
  • Report in json format: images/<platename>/report.json
  • Log: images/<platename>/log.txt

description of schema: * row: well row index * column: well colmun index * total: well area in pixels * resistance: absolute resistance found in pixels * density: density of the resistance found

report example:

"7-J":{
   "density":0.17,
   "column":"A",
   "resistance":122,
   "total":706,
   "row":"4"
}

output images example:

4-A_122-0.23, is the well 4-A, with 122 pixels found as resistance with density of 17%

output log example:

customizing scale well: found False, num wells 93, min radius value 18, max radius value 23
customizing scale well: found False, num wells 96, min radius value 18, max radius value 24
customizing grid matching: found False, num wells recognized 96
Succesfully processed plate, found 96 wells

Installing dependencies

pip

sudo apt-get install python-pip ### opencv sudo apt-get install build-essential sudo apt-get install cmake git libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-dev sudo apt-get install python-opencv ### scilab sudo apt-get install python-scipy

Installing arip

There are two ways of installing pynteractive: * Cloning the project

$ git clone https://github.com/mazeitor/antibiotic-resistance-process.git
$ cd antibiotic-resistance-process
$ python setup.py install  ### (as root)

TODO

  • Normalizing radius by neighborhood instead of general average
  • Working with static grids or masks

Project details


Release history Release notifications

This version
History Node

1.0.2

History Node

1.0.1

History Node

1.0.0

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Filename, size & hash SHA256 hash help File type Python version Upload date
arip-1.0.2-py2-none-any.whl (10.6 kB) Copy SHA256 hash SHA256 Wheel py2 Feb 14, 2017
arip-1.0.2-py2.py3-none-any.whl (10.6 kB) Copy SHA256 hash SHA256 Wheel py2.py3 Feb 14, 2017
arip-1.0.2.tar.gz (9.0 kB) Copy SHA256 hash SHA256 Source None Feb 14, 2017

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging CloudAMQP CloudAMQP RabbitMQ AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page