Segmentation and Classification of Cine-MR Images Using Fully Convolutional Networks and Handcrafted Features
This work addresses the problem of precise cardiac condition diagnosis for medical professionals by automating segmentation and classification from MRI images, though it is incremental with hybrid methods.
The study tackled automated segmentation and classification of cardiac structures from 3D cine-MRI images to assess cardiac function, achieving Dice Coefficients of 0.95±0.03 for LV, 0.89±0.13 for RV, and 0.90±0.03 for MC in segmentation, and classification accuracies of 95.05% and 92.77% using ground truth and proposed segmentations, respectively.
Three-dimensional cine-MRI is of crucial importance for assessing the cardiac function. Features that describe the anatomy and function of cardiac structures (e.g. Left Ventricle (LV), Right Ventricle (RV), and Myocardium(MC)) are known to have significant diagnostic value and can be computed from 3D cine-MR images. However, these features require precise segmentation of cardiac structures. Among the fully automated segmentation methods, Fully Convolutional Networks (FCN) with Skip Connections have shown robustness in medical segmentation problems. In this study, we develop a complete pipeline for classification of subjects with cardiac conditions based on 3D cine-MRI. For the segmentation task, we develop a 2D FCN and introduce Parallel Paths (PP) as a way to exploit the 3D information of the cine-MR image. For the classification task, 125 features were extracted from the segmented structures, describing their anatomy and function. Next, a two-stage pipeline for feature selection using the LASSO method is developed. A subset of 20 features is selected for classification. Each subject is classified using an ensemble of Logistic Regression, Multi-Layer Perceptron, and Support Vector Machine classifiers through majority voting. The Dice Coefficient for segmentation was 0.95+-0.03, 0.89+-0.13, and 0.90+-0.03 for LV, RV, and MC respectively. The 8-fold cross validation accuracy for the classification task was 95.05% and 92.77% based on ground truth and the proposed methods segmentations respectively. The results show that the PPs increase the segmentation accuracy, by exploiting the spatial relations. Moreover, the classification algorithm and the features showed discriminability while keeping the sensitivity to segmentation error as low as possible.