Skip to main content
Join the official 2019 Python Developers SurveyStart the survey!

A package to study skin lesion's symmetry and help diagnose diseases like menalomas.

Project description

dermoscopic_symmetry

dermoscopic_symmetry is a Python package aiming at study skin lesion's symmetry (regarding shape and textures) and help the diagnose of diseases like melanomas.
Basically, the symmetry study is divided into two parts: shapes symmetry and textures symmetry. Here, shapes means the aspect of the outskirts of a lesion and its global form whereas textures stand for colors and types of perceived textures.

Note : The package has been built referring to the PH² Dataset. See :

Teresa Mendonça, Pedro M. Ferreira, Jorge Marques, Andre R. S. Marcal, Jorge Rozeira. PH² - A dermoscopic image database for research and benchmarking, 35th International Conference of the IEEE Engineering in Medicine and Biology Society, July 3-7, 2013, Osaka, Japan.

To be able to use it directly and properly, you must download and have access to it.

Installation

Use pip to install the package.

pip install dermoscopic_symmetry

These are the Python files used to study the symmetry of skin lesions from the PH² Dataset (see Usage section):

  1. shape_symmetry.py : containing functions to study the symmetry of shapes in a lesion's image.

  2. classifier_feeder.py : containing functions to create a classifier able to recognize if 2 patches taken in a lesion's image are similar or not.
    A new dataset called patchesDataSet, derivating from the PH² Dataset, has been designed to train this classifier. It is composed of patches pairs taken in the PH² Dataset images with one half similar and the other non similar.

  3. patches_for_texture_symmetry.py : containing functions to take patches from a dermoscopic image and extract features from them (textures and color).

  4. texture_symmetry.py : containing functions using the previous classifier and features to study the symmetry of textures in a lesion's image.

  5. combined_classifier.py : containing functions using only shape features, only textures features or both of them to train classifiers and be able to know which one is the best according to expert diagnose in the PH² Dataset.
    Those classifiers are trained according to the ShapesScores.csv, TextureScores.csv and ShapeAndTextureScores.csv files contained in the data repository. The final models are saved as shapeModel.pkl, textureModel.pkl and shapeAndTextureModel.pkl in the data/models repository.

Note : The code used to create the patchesDataSet is given in the patches_dataset_creator.py file. The utils.py file contains the utilities functions.

Usage

Each code script has an : example() function at the beginning aiming at presenting its functionalities. This function is run as a default main.

License

MIT

Project details


Download files

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

Files for dermoscopic-symmetry, version 0.0.2
Filename, size File type Python version Upload date Hashes
Filename, size dermoscopic_symmetry-0.0.2-py2-none-any.whl (11.1 MB) File type Wheel Python version py2 Upload date Hashes View hashes
Filename, size dermoscopic_symmetry-0.0.2.tar.gz (10.9 MB) File type Source Python version None Upload date Hashes View hashes

Supported by

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