Automatic localization of the left ventricle in cardiac MRI images using deep learning

Automatic localization of the left ventricle (LV) in cardiac MRI images is an essential step for automatic segmentation, functional analysis, and content based retrieval of cardiac images. In this paper, we introduce a new approach based on deep Convolutional Neural Network (CNN) to localize the LV in cardiac MRI in short axis views. A six-layer CNN with different kernel sizes was employed for feature extraction, followed by Softmax fully connected layer for classification. The pyramids of scales analysis was introduced in order to take account of the different sizes of the heart. A publically-available database of 33 patients was used for learning and testing. The proposed method was able it localize the LV with 98.66%, 83.91% and 99.07% for accuracy, sensitivity and specificity respectively.

[1]  Milan Sonka,et al.  4-D Cardiac MR Image Analysis: Left and Right Ventricular Morphology and Function , 2010, IEEE Transactions on Medical Imaging.

[2]  Antonio Criminisi,et al.  Regression Forests for Efficient Anatomy Detection and Localization in CT Studies , 2010, MCV.

[3]  Dorin Comaniciu,et al.  Search strategies for multiple landmark detection by submodular maximization , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[4]  John K. Tsotsos,et al.  Efficient and generalizable statistical models of shape and appearance for analysis of cardiac MRI , 2008, Medical Image Anal..

[5]  Caroline Petitjean,et al.  A review of segmentation methods in short axis cardiac MR images , 2011, Medical Image Anal..

[6]  Dorin Comaniciu,et al.  Image based regression using boosting method , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[7]  Paul F. Whelan,et al.  A Novel Model-Based 3D ${+}$Time Left Ventricular Segmentation Technique , 2011, IEEE Transactions on Medical Imaging.

[8]  Ioannis A. Kakadiaris,et al.  Automated left ventricular segmentation in cardiac MRI , 2006, IEEE Transactions on Biomedical Engineering.

[9]  D. Hubel,et al.  Receptive fields and functional architecture of monkey striate cortex , 1968, The Journal of physiology.

[10]  S. Sotthivirat,et al.  Active Contours With Automatic Initialization For Myocardial Perfusion Analysis , 2005, 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference.

[11]  Alexander Dick,et al.  Segmentation of Left Ventricle in Cardiac Cine MRI: An Automatic Image-Driven Method , 2009, FIMH.

[12]  Dorin Comaniciu,et al.  Hierarchical parsing and semantic navigation of full body CT data , 2009, Medical Imaging.

[13]  Andreas Schöllhuber.  Fully Automatic Segmentation of the Myocardium in Cardiac Perfusion MRI , 2008 .

[14]  Kieran Clarke,et al.  Dystrophinopathies are characterised by impaired cardiac metabolism, contractile dysfunction and fibrosis in patients with and without coxsackie B3 exposure , 2011, Journal of Cardiovascular Magnetic Resonance.

[15]  Dorin Comaniciu,et al.  Constrained Marginal Space Learning , 2014 .

[16]  Dorin Comaniciu,et al.  A boosting regression approach to medical anatomy detection , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[17]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[18]  Xiaoguang Lu,et al.  Automatic LV localization and view planning for cardiac MRI acquisition , 2011, Journal of Cardiovascular Magnetic Resonance.

[19]  Luca Maria Gambardella,et al.  Proceedings of the Twenty-Second International Joint Conference on Artificial Intelligence Flexible, High Performance Convolutional Neural Networks for Image Classification , 2022 .

[20]  Yiqiang Zhan,et al.  Active Scheduling of Organ Detection and Segmentation in Whole-Body Medical Images , 2008, MICCAI.

[21]  Liang Zhong,et al.  Automatic Localization of the Left Ventricle from Cardiac Cine Magnetic Resonance Imaging: A New Spectrum-Based Computer-Aided Tool , 2014, PloS one.