Advertisement

Automated Cardiovascular Pathology Assessment Using Semantic Segmentation and Ensemble Learning

  • Tony LindseyEmail author
  • Jin-Ju Lee
Article

Abstract

Cardiac magnetic resonance imaging provides high spatial resolution, enabling improved extraction of important functional and morphological features for cardiovascular disease staging. Segmentation of ventricular cavities and myocardium in cardiac cine sequencing provides a basis to quantify cardiac measures such as ejection fraction. A method is presented that curtails the expense and observer bias of manual cardiac evaluation by combining semantic segmentation and disease classification into a fully automatic processing pipeline. The initial processing element consists of a robust dilated convolutional neural network architecture for voxel-wise segmentation of the myocardium and ventricular cavities. The resulting comprehensive volumetric feature matrix captures diagnostic clinical procedure data and is utilized by the final processing element to model a cardiac pathology classifier. Our approach evaluated anonymized cardiac images from a training data set of 100 patients (4 pathology groups, 1 healthy group, 20 patients per group) examined at the University Hospital of Dijon. The top average Dice index scores achieved were 0.940, 0.886, and 0.849 for structure segmentation of the left ventricle (LV), myocardium, and right ventricle (RV), respectively. A 5-ary pathology classification accuracy of 90% was recorded on an independent test set using the trained model. Performance results demonstrate the potential for advanced machine learning methods to deliver accurate, efficient, and reproducible cardiac pathological assessment.

Keywords

2D U-Net Cardiac cine-MRI Semantic segmentation Feature selection Classification 

Notes

References

  1. 1.
    Cohn J, Ferrari R, Sharpe N: Cardiac remodeling concepts and clinical implications: A consensus paper from an international forum on cardiac remodeling. JACC 35:569–582, 2000CrossRefGoogle Scholar
  2. 2.
    Saeed M, Van T, Krug R, Hetts S, Wilson M: Cardiac MR imaging: Current status and future direction. Cardiovascular Diagnosis Therapy 5(4):290–310, 2015PubMedGoogle Scholar
  3. 3.
    Doi K.: Computer-aided diagnosis in medical imaging: Historical review, current status and future potential. CMIG 31(4):198–211, 2008Google Scholar
  4. 4.
    Ritter S, Margulies K: Emerging Tools for Computer-Aided Diagnosis and Prognostication. J Clin Trials 4(2):1–120, 2014Google Scholar
  5. 5.
    Ronneberger O, Fischer P, Brox T: U-net: Convolutional networks for biomedical image segmentation. Medical Image Computing and Computer-Assisted Intervention (MICCAI) 9351:234–241, 2015Google Scholar
  6. 6.
    Dubloit J (2017) Automated cardiac diagnosis challenge. Available from: https://www.creatis.insalyon.fr/Challenge/acdc
  7. 7.
    Sorensen T: A method of establishing groups of equal amplitude in plant sociology based on similarity of species and its application to analyses of the vegetation on Danish commons. K Dan Vidensk Selsk 5(4):1–34, 1948Google Scholar
  8. 8.
    Kingma D, Ba J (2015) Adam: A Method for Stochastic Optimization. In: 3rd International Conference for Learning Representations. vol. arXiv:1412.6980
  9. 9.
    Glorot X, Bengio Y Understanding the difficulty of training deep feedforward neural networks. In: Proceedings of the thirteenth international conference on artificial intelligence and statistics, 2010, pp 249–256Google Scholar
  10. 10.
    Ronneberger O, Fischer P, Brox T (2015) U-Net: Convolutional Networks for Biomedical Image Segmentation. In: arXiv:1505.04597

Copyright information

© Society for Imaging Informatics in Medicine 2020

Authors and Affiliations

  1. 1.Intelligent Systems, NASA Ames Research CenterMountain ViewUSA
  2. 2.Biomedical InformaticsStanford School of MedicineStanfordUSA

Personalised recommendations