Automatic Segmentation and ROI detection in cardiac MRI of Cardiomyopathy using q-Sigmoid as preprocessing step

  • Eduardo Coltri Centro Universitário FEI
  • Gabriel S. Figueredo Costa Centro Universitário FEI
  • Kelvin Lins Silva Centro Universitário FEI
  • Pedro Zigante Martim Centro Universitário FEI
  • Leila Cristina C. Bergamasco Centro Universitário FEI

Resumo


The growth of data volume is a reality in all such as segments of our society. Despite of personalized experiences, accurate and fast information, new challenges had arisen. For healthcare industry, for example, it was noted an increase of radiologists workload which may cause visual fatigue and, consequently, errors during diagnosis. intelligence was pointed as an option to support Artificial physicians analysis and reduce the visual fatigue. Thus, this paper focus on the proposal of a novel strategy to enhance cardiac magnetic resonance images (MRI) and automatically detect their region-of-interest (ROI) using a convolutional neural network (CNN). Our object of study is the disease of Cardiomyopathy and the desirable ROI is the left ventricle from axial slices. We evaluated q-Sigmoid performance using it as a preprocessing step and validate the results through modified CNNs: U-Net and ResNet.

Palavras-chave: Cardiac MRI, cardiomyopathy, ROI detection, CNN

Referências

M. Abadi, A. Agarwal, and P. B. and. TensorFlow: Large-scale machine learning on heterogeneous systems, 2015. Software available from tensorflow.org.

F. M. Albanesi Filho. Cardiomiopatias. Arquivos Brasileiros de Cardiologia, 71(2), 1998.

S. Bach. Afinal, o que é big data? https://administradores.com.br/artigos/ bigdata, 05 2019.

C. F. Baumgartner, L. M. Koch, M. Pollefeys, and E. Konukoglu. An exploration of 2d and 3d deep learning techniques for cardiac mr image segmentation. ArXiv, abs/1709.04496, 2017.

L. C. C. Bergamasco. Recuperação de objetos médicos 3d utilizando harmônicos esféricos e redes de fluxo. 2015.

O. Bernard, A. Lalande, and C. Zotti. Deep learning techniques for automatic mri cardiac multi-structures segmentation and diagnosis: Is the problem solved? IEEE Transactions on Medical Imaging, 37:2514-2525, 2018.

G. Bradski. The OpenCV Library. Dr. Dobb's Journal of Software Tools, 2000.

M. Brett, C. J. Markiewicz, and M. Hanke. Nibabel. https://github.com/nipy/nibabel.

F. Chollet et al. Keras. https://github.com/fchollet/keras, 2015.

S. Fletcher and M. Z. Islam. Comparing sets of patterns with the jaccard index. Australasian Journal of Information Systems, 22, 2018.

C. R. Harris, K. J. Millman, S. J. van der Walt, R. Gommers, P. Virtanen, D. Cournapeau, E. Wieser, J. Taylor, S. Berg, N. J. Smith, and et al. Array programming with numpy. Nature, 585(7825):357-362, Sep 2020.

K. He, X. Zhang, S. Ren, and J. Sun. Deep residual learning for image recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 770-778, 2016.

F. Isensee, P. F. Jaeger, and P. M. Full. Automatic cardiac disease assessment on cine-mri via time-series segmentation and domain specific features. ArXiv, abs/1707.00587, 2017.

Z. Liu, P. Li, J. Li, Q. Xie, and X. Wang. Left ventricular full segmentation from cardiac magnetic resonance imaging via multi-task learning. 2021 IEEE 2nd International Conference on Big Data, Artificial Intelligence and Internet of Things Engineering (ICBAIE), pages 71-75, 2021.

B. J. Maron. Hypertrophic cardiomyopathy. Jama, 287(10), 2002.

I. e. a. Martire. A novel probabilistic jaccard distance measure for classification of sparse and uncertain data. 2017.

A. Pah, L. Rasmussen-Torvik, S. Goel, P. Greenland, and A. Kho. Big data: What is it and what does it mean for cardiovascular research and prevention policy. Current Cardiovascular Risk Reports, 9, 12 2014.

P. Rodrigues, G. A. Wachs-Lopes, R. M. Santos, E. Coltri, and G. Giraldi. A q-extension of sigmoid functions and the application for enhancement of ultrasound images. Entropy, 21, 2019.

O. Ronneberger, P. Fischer, and T. Brox. U-net: Convolutional networks for biomedical image segmentation. ArXiv, abs/1505.04597, 2015.

V. Sekar, M. Zhang, C. Shu, and B. C. Khoo. Inverse design of airfoil using a deep convolutional neural network. AIAA Journal, 57(3):993-1003, 2019.

K. Simonyan and A. Zisserman. Very deep convolutional networks for large-scale image recognition. CoRR, abs/1409.1556, 2015.

C. Tsallis. Nonextensive statistics: theoretical, experimental and computational evidences and connections. Brazilian Journal of Physics, 29:1 - 35, 03 1999.

P. Yakubovskiy. Segmentation models. https://github.com/qubvel/segmentation_models, 2019.
Publicado
22/11/2021
COLTRI, Eduardo; COSTA, Gabriel S. Figueredo; SILVA, Kelvin Lins; MARTIM, Pedro Zigante; BERGAMASCO, Leila Cristina C.. Automatic Segmentation and ROI detection in cardiac MRI of Cardiomyopathy using q-Sigmoid as preprocessing step. In: WORKSHOP DE VISÃO COMPUTACIONAL (WVC), 17. , 2021, Online. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2021 . p. 143-147. DOI: https://doi.org/10.5753/wvc.2021.18904.

Artigos mais lidos do(s) mesmo(s) autor(es)

Obs.: Esse plugin requer que pelo menos um plugin de estatísticas/relatórios esteja habilitado. Se o seu plugins de estatísticas oferece mais que uma métrica, então, por favor, também selecione uma métrica principal na página de configurações administrativas do site e/ou da revista.