Chapter 3 Deep Learning 1

Author: Weronika Hryniewska

Deep learning is one of the most rapidly developing field in artificial intelligence. Problems that previously required a lot of features engineering became easily solvable. New possibilities opened, and deep learning has started to adopt in various domains. One of the most demanding disciplines is medicine.

As a result of the outbreak of the COVID-19 pandemic, many scientists became interested in the possibilities of deep learning application in radiology. Many solutions have been created for classification, segmentation and detection based on computed tomography and radiographs of the lungs.

During classes, we explored deep learning methods for computer vision. If you would like to read more about them, please take a look at books: “Deep Learning with Python” (Chollet 2017) and “Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems” (Géron 2017). We focused on results reproduction and/or further development of the available code of the following papers::

  1. LungNet (Anthimopoulos et al. 2019) Adam Frej, Piotr Marciniak, Piotr Piątyszek
  2. BCDU-Net (Azad et al. 2019) (Asadi-Aghbolaghi et al. 2020) Maria Kałuska, Paweł Koźmiński, Mikołaj Spytek
  3. DeepCOVIDExplainer (Karim et al. 2020) Kacper Kurowski, Zuzanna Mróz, Aleksander Podsiad
  4. ERSCovid (S. Wang et al. 2020a) Bartłomiej Eljasiak, Tomasz Krupiński, Dominik Pawlak
  5. COVID-Net (L. Wang et al. 2020a) Jakub Kozieł, Tomasz Nocoń, Kacper Staroń

References

Anthimopoulos, M., Christodoulidis, S., Ebner, L., Geiser, T., Christe, A., & Mougiakakou, S. (2019). Semantic segmentation of pathological lung tissue with dilated fully convolutional networks. IEEE Journal of Biomedical and Health Informatics, 23(2), 714–722. https://doi.org/10.1109/JBHI.2018.2818620
Asadi-Aghbolaghi, M., Azad, R., Fathy, M., & Escalera, S. (2020). Multi-level context gating of embedded collective knowledge for medical image segmentation. https://arxiv.org/abs/2003.05056
Azad, R., Asadi-Aghbolaghi, M., Fathy, M., & Escalera, S. (2019). Bi-directional ConvLSTM u-net with densley connected convolutions. In 2019 IEEE/CVF international conference on computer vision workshop (ICCVW) (pp. 406–415). https://doi.org/10.1109/ICCVW.2019.00052
Chollet, F. (2017). Deep learning with python. Manning.
Géron, A. (2017). Hands-on machine learning with scikit-learn and TensorFlow : Concepts, tools, and techniques to build intelligent systems. O’Reilly Media.
Karim, M. R., Döhmen, T., Rebholz-Schuhmann, D., Decker, S., Cochez, M., & Beyan, O. (2020). DeepCOVIDExplainer: Explainable COVID-19 diagnosis from chest x-ray images. IEEE. https://doi.org/10.1109/BIBM49941.2020.9313304
Wang, L., Lin, Z. Q., & Wong, A. (2020a). COVID-net: A tailored deep convolutional neural network design for detection of COVID-19 cases from chest x-ray images. Scientific Reports, 10(1), 19549. https://doi.org/10.1038/s41598-020-76550-z
Wang, S., Zha, Y., Li, W., Wu, Q., Li, X., Niu, M., et al. (2020a). A fully automatic deep learning system for COVID-19 diagnostic and prognostic analysis. European Respiratory Journal, 56(2). https://doi.org/10.1183/13993003.00775-2020