Automatic Left Ventricle Segmentation from Short-Axis Cardiac MRI Images Based on Fully Convolutional Neural Network

Abstract

Background: Left ventricle (LV) segmentation using a cardiac magnetic resonance imaging (MRI) dataset is critical for evaluating global and regional cardiac functions and diagnosing cardiovascular diseases. LV clinical metrics such as LV volume, LV mass and ejection fraction (EF) are frequently extracted based on the LV segmentation from short-axis MRI images. Manual segmentation to assess such functions is tedious and time-consuming for medical experts to diagnose cardiac pathologies. Therefore, a fully automated LV segmentation technique is required to assist medical experts in working more efficiently. Method: This paper proposes a fully convolutional network (FCN) architecture for automatic LV segmentation from short-axis MRI images. Several experiments were conducted in the training phase to compare the performance of the network and the U-Net model with various hyper-parameters, including optimization algorithms, epochs, learning rate, and mini-batch size. In addition, a class weighting method was introduced to avoid having a high imbalance of pixels in the classes of image’s labels since the number of background pixels was significantly higher than the number of LV and myocardium pixels. Furthermore, effective image conversion with pixel normalization was applied to obtain exact features representing target organs (LV and myocardium). The segmentation models were trained and tested on a public dataset, namely the evaluation of myocardial infarction from the delayed-enhancement cardiac MRI (EMIDEC) dataset. Results: The dice metric, Jaccard index, sensitivity, and specificity were used to evaluate the network’s performance, with values of 0.93, 0.87, 0.98, and 0.94, respectively. Based on the experimental results, the proposed network outperforms the standard U-Net model and is an advanced fully automated method in terms of segmentation performance. Conclusion: This proposed method is applicable in clinical practice for doctors to diagnose cardiac diseases from short-axis MRI images.

Detection of Left Ventricular Cavity from Cardiac MRI Images Using Faster R-CNN

“…Overlap Ratio = ground truth box ∩ predicted box ground truth box ∪ predicted box (7) In terms of LV detection, the proposed network outperforms the R-CNN network. Tab.…”

“…The test results were evaluated using the overlap ratio index, which achieved efficient detection of the LV from cardiac MRI images using the labelled images as a reference, as in Eq. (7).…”

“…Thus, LV localization is extremely useful in reducing computational load, particularly for deep learning algorithms. The fundamental advantage of LV localization is that it improves the effectiveness of subsequent techniques like regression [4] and segmentation [5][6][7] by extracting only the LV area and ignoring the other organs, as illustrated in Fig. 1.…”

Convolutional Neural Network Model to Segment Myocardial Infarction from MRI Images

“…Several methods employed traditional segmentation techniques such as thresholding based on image intensity and resolution to distinguish healthy and pathological tissues. Recently, deep learning-based algorithms have achieved state-of-the-art performance of myocardium, and LV segmentation in cardiac MRI [22]- [25], and more recent studies in this field are summarized in [26]. Table 1 summarized the characteristics of the recent related works in LV and myocardium segmentation using deep learning algorithms.…”

Automatic left ventricle segmentation via edge‐shape feature‐based fully convolutional neural network

“…This objective indicates that the automation system should be trained with a greater number of samples. [19][20][21][22] The existing techniques for segmentation of LV from the cardiac images depend on prior structure and the features are unstructured. These features are applied to the deep learning-based segmentation task which gives over or under-segmentation results and gives poor accuracy.…”