snaildmy/fae

FeAture Explorer

FAE

Feature Analysis Explorer (FAE) can help researchers develop a classification model with comparison among different methods. This project was inspired on the Radiomics, and provides a GUI to help analyze the feature matrix, including feature matrix pre-process, model development, and results visualization.

If you publish any work which uses this package, I will appreciate that you could give the following link (https://github.com/salan668/FAE)

Welcome any issues and PR.

Release

The Windows 64 version was release here https://drive.google.com/open?id=1htts7YsfaxKtN1NeDcNU4iksXfjr_XyK (Alternative link is: https://pan.baidu.com/s/1ha66TajeoT6dA-a4Qdt8fA)

A short tutorial video may help: https://www.bilibili.com/video/BV1yt4y1S79S/ (Chinese Version)

If FAE could help in your research, please refer to

Y. Song, J. Zhang, Y. Zhang, Y. Hou, X. Yan, Y. Wang, M. Zhou, Y. Yao, G. Yang. FeAture Explorer (FAE): A tool for developing and comparing radiomics models. PLoS One. 2020. DOI: https://doi.org/10.1371/journal.pone.0237587

Getting Started

Pre-install

The below modules must be installed first to make the FAE work.

- imbalanced-learn=0.6.2
- matplotlib=3.2.0
- numpy=1.18.1
- pandas=1.0.1
- pdfdocument=3.3
- pillow=7.0.0
- PyQt5=5.14.1
- PyQtGraph=0.10.0
- pyradiomics=3.0
- reportlab=3.5.34
- scikit-learn=0.22.2
- scipy=1.4.1
- seaborn=0.10.0
- statsmodels=0.11.1

Installing

Just clone it by typing in:

git clone https://github.com/salan668/FAE.git

The .ui file has to be transfered to the .py file by pyuic manually. For example, GUI/HomePage.ui should be tranfered to GUI/HomePage.py file.

Architecture of Project

• DataContainer
  • DataContainer. The structure to contain the data, which also includes methods like saving/loading.
  • DataSeparate. Including functions to separate data into training part and testing part.
• Feature Analysis
  • DataBalance, Sample the cases to make the binary-labels balance.
  • Normalization. Normalize the data to avoid the scale effect of different features.
  • DimensionReduction. Reduce the feature dimension, including PCA.
  • Classifier. Map the features onto the labels.
  • CrossValidation. Estimate model by using cross-validation on the training data set.
  • FeatureSelector. Select the sub-features from the feature matrix.
  • Pipelines. The class to estimate the model with different feature selected method and classifier.
• Image2Feature
  • RadiomicsFeatureExtractor. This class help extract features from image and ROI with batch process. This class should be more "smart" in the future.
• Visulization.
  • DrawDoubleLine. This function helps draw double-y plot. e.g. plot accuracy and error against the number of iterations.
  • DrawROCList. This function helps draw different ROC curves. AUC will be calculated automaticly and labeled on the legend.
  • FeatureRelationship. This function helps draw the distribution of the values of different features. I can only show at most 3 features in one figure.
  • FeatureSort. This function helps draw the features and the weights of them on the classifiction model.
  • PlotMetricVsFeatureNumber. This function helps draw the AUC / Accuracy / other metrics against the number of chosen features. This can help find the adaptive number of the features.
• Report
  • To Generate the report with PDF format.

Document

TODO

Author

• Yang Song
• Jing Zhang
• Guang Yang
• Chengxiu Zhang

License

This project is licensed under the GPL 3.0 License - see the LICENSE.md file for details

Acknowledge

• Contributor:
  • Yi-lai Pei
  • jmtaysom
  • Zhiyong Zhao
• Demo data support.
  • Yu-dong Zhang, Xu Yan.