Skip to content

Facial Analysis and Binary Classification of Acutely Ill Patients

License

Notifications You must be signed in to change notification settings

vandrw/acute_illness_classification

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Facial Analysis and Binary Classification of Acutely Ill Patients

Identifying whether a patient appears sick using Convolutional Neural Networks.

For in-depth information regarding the project, please refer to the paper.

How to run?

To extract the features, first place the images in their corresponding directories (e.g. data/unparsed/sick for images representing sick individuals) and run the command make create-data in a terminal.

Once the data is created, make train-individual or make train-stacked can be used to train all the individual networks (i.e. eyes, nose, mouth and skin) or, respectively, a stacked ensemble that parses each feature at once.

If one desires to remove the created data, make clean-data can be used. Furthermore, make clean-results will remove any saved models, histories and plots generated.

For python environment details, please check environment.py.

Project Structure

  • augment: folder containing the code of a neural style transfer network
  • categorization: folder containing a convolutional neural network that categorizes the images
  • data: folder containing the collected data set

Potential Data Sets for Augmentation

References for previous datasets

A. Sepas-Moghaddam, V. Chiesa, P.L. Correia, F. Pereira, J. Dugelay, “The IST-EURECOM Light Field Face Database”, International Workshop on Biometrics and Forensics, IWBF 2017, Coventry, UK, April 2017

Mahmoud Afifi and Abdelrahman Abdelhamed, "AFIF4: Deep gender classification based on an AdaBoost-based fusion of isolated facial features and foggy faces". Journal of Visual Communication and Image Representation, 2019.

PEER, Peter, EMERŠIČ, Žiga, BULE, Jernej, ŽGANEC GROS, Jerneja, ŠTRUC, Vitomir. Strategies for exploiting independent cloud implementations of biometric experts in multibiometric scenarios. Mathematical problems in engineering, vol. 2014, pp. 1-15, 2014.

About

Facial Analysis and Binary Classification of Acutely Ill Patients

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published