DeepCADD

This repository contains the source files of our research on DeepCADD: a Deep Neural Network for Automatic Detection of Coronary Artery Disease (CAD). The study contains three main features: CAD segmentation using mainly the Frangi filter; CAD Classification using a ResNet-50; CAD detection using a Mask R-CNN-based architecture.

This repository contains the following algorithms:

Right Coronary Artery (RCA) Segmentation, introduced in [2], which provides a technique for image segmentation of the right coronary artery (RCA) main segment. See experiment at segmentation/CAD-segmentation.ipynb
CNN for coronary segment classification, introduced in [1], which evaluates the performance of ResNet-50 architecture on classifying CAD segments into stenosed and non-stenosed. See experiment at colab/DeepCADD-backbone-ResNet50.ipynb
Mask R-CNN for CAD detection, introduced in [1], which introduces the use of Mask-RCNN for CAD detection. See experiment at colab/DeepCADD-running-with-ResNet50-backbone.ipynb and colab/DeepCADD-training-with-ResNet50.ipynb

Publications

S. A. Freitas, F. A. Zeiser, C. A. Da Costa and G. De O. Ramos, “DeepCADD: A Deep Learning Architecture for Automatic Detection of Coronary Artery Disease,” 2022 International Joint Conference on Neural Networks (IJCNN), 2022, pp. 1-8. [link].
FREITAS, Samuel A.; RAMOS, Gabriel de O.; COSTA, Cristiano André da. DEEPCADD: A Deep Neural Network for Automatic Detection of Coronary Artery Disease. In: CONCURSO DE TESES E DISSERTAÇÕES - SIMPÓSIO BRASILEIRO DE COMPUTAÇÃO APLICADA À SAÚDE (SBCAS), 22. , 2022, Teresina/PI. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2022 . p. 25-32. ISSN 2763-8987. [link].
FREITAS, Samuel A.; NIENOW, Débora; COSTA, Cristiano A. da; RAMOS, Gabriel de O.. Functional Coronary Artery Assessment: a Systematic Literature Review. In: Wiener Klinische Wochenschrift, 2021. [link]
FREITAS, Samuel A.; COSTA, Cristiano A. da; RAMOS, Gabriel de O.. Coronary Artery Disease Automatic Classification. In: ESCOLA REGIONAL DE COMPUTAÇÃO APLICADA À SAÚDE (ERCAS), 8. , 2021, São Paulo. Anais […]. Porto Alegre: Sociedade Brasileira de Computação, 2021 . p. 26-29. [link]
FREITAS, Samuel; RAMOS, Gabriel; SCHMITH, Jean; COSTA, Cristiano. Nodal Analysis for Coronary Artery Ischemia Diagnosis. In: SIMPÓSIO BRASILEIRO DE COMPUTAÇÃO APLICADA À SAÚDE (SBCAS), 20. , 2020, Evento Online. Anais […]. Porto Alegre: Sociedade Brasileira de Computação, 2020 . p. 262-272. [link]
NIENOW, Débora; FREITAS, Samuel; SCHMITH, Jean. Software para auxílio no diagnóstico de estenoses em artérias coronárias. Arq. Bras. Cardiol: 113 (2 sup l. 1) p. 254. 2019. [link]

Requirements

Before you begin, ensure you have met the following requirements:

Python v3.6
Tensorflow v1.x
h5py v2.10.0
Ipython v7.31.0
OpenCV v4.1.2

Usage

We decided to execute all the experiments using Google Colab PRO, powered by the existing available libraries and avoiding to overload the local machine.

For CAD segmentation, copy the notebook available at segmentation/CAD-segmentation.ipynb.

For DeepCADD backbone classification, copy the notebook available at colab/DeepCADD-backbone-ResNet50.ipynb.

For DeepCADD detection, copy the notebook available at colab/DeepCADD-training-with-ResNet50.ipynb.

tbc

Contributors

Thanks to the following people who have contributed to this project:

How to cite this research

For citing this work, please use the following entries. For entries of the other works, go to https://armbrustsamuel.github.io/publications/.

@INPROCEEDINGS{freitas2022ijcnn,  
  author={Freitas, Samuel A. and Zeiser, Felipe A. and Da Costa, Cristiano A. and De O. Ramos, Gabriel},  
  booktitle={2022 International Joint Conference on Neural Networks (IJCNN)},   
  title={DeepCADD: A Deep Learning Architecture for Automatic Detection of Coronary Artery Disease},   
  year={2022},  
  pages={1-8},  
  doi={10.1109/IJCNN55064.2022.9892501}}

@article{freitas2021functional,
  title={Functional Coronary Artery Assessment: a Systematic Literature Review},
  author={Freitas, Samuel A and Nienow, D{\'e}bora and da Costa, Cristiano A and Ramos, Gabriel de O},
  journal={Wiener klinische Wochenschrift},
  pages={1--17},
  year={2021},
  publisher={Springer}
}

License

This project uses the following license: MIT.

schneiderl/DeepCADD