Model-based processing scheme for quantitative 4-D cardiac MRI analysis

Making Semi-Automatic Segmentation Method to be Automatic Using Deep Learning for Biventricular Segmentation

Ventricular Segmentation or Delineation of Cardiac Magnetic Resonance Imaging (CMRI) is significant in obtaining the cardiac contractile function, which in turn is taken as input for diagnosing Cardio Vascular Diseases (CVD). Many automatic and semi-automatic methods were evolved to meet the constraints of diagnosing CVDs. Among these, semi-automatic methods require user intervention for delineation of ventricles, which consumes time and leads to intra and inter-observability, as with manual delineation. Thus, the automatic method is suggested by most of the researchers to address the above-stated problem. We proposed Saliency-based Active contour U-Net (SACU-Net) for automatic bi-ventricular segmentation which is found to surpass the existing highest developed methods regarding closeness to the gold standard. Three schemes are used by our proposed algorithm, namely 1. Saliency Detection Scheme for Region of Interest (ROI) Localization to concentrate only on Object of Interest, 2. Drop-out embedded U-net for Initial Contour evolution that performs initial segmentation and 3. Local-Global-based Regional active Contour (LGRAC) to fine-tune and avoid leaking, merging of ventricles during Delineation. We used three datasets namely Automatic Cardiac Diagnosing Challenge (ACDC) of MICCAI 2017, Right Ventricular Segmentation Challenge (RVSC) of MICCAI 2012, and Sunny Brook (SB) of MICCAI 2009 dataset to test the adaptability nature of our algorithm over different scanner resolutions and protocols. 100 and 50 CMRI Images of ACDC were used for training and testing respectively which obtained average Dice Coefficient (DC) metric of 0.963, 0.934, and 0.948 for Left Ventricular Cavity (LVC), Left Ventricular Myocardium (LVM), and Right Ventricular Cavity (RVC) respectively. 32 and 16 CMRI Images of RVSC are used for preparing and experimenting respectively, which obtained an average DC metric of 0.95 for RVC.30 and 15 CMRI Images of SB are used for preparing and experimenting respectively, which obtained average DC metric of 0.96 and 0.97 for LVC and LVM, respectively. Hausdorff Distance (HD) Metrics are also calculated to learn the distance of proposed delineated ventricles to reach the gold standard. The above resultant metrics show the robustness of our proposed SACU-Net in the segmentation of ventricles of CMRI than previous methods.

TSU-net: Two-stage multi-scale cascade and multi-field fusion U-net for right ventricular segmentation

Accurate segmentation of the right ventricle from cardiac magnetic resonance images (MRI) is a critical step in cardiac function analysis and disease diagnosis. It is still an open problem due to some difficulties, such as a large variety of object sizes and ill-defined borders. In this paper, we present a TSU-net network that grips deeper features and captures targets of different sizes with multi-scale cascade and multi-field fusion in the right ventricle. TSU-net mainly contains two major components: Dilated-Convolution Block (DB) and Multi-Layer-Pool Block (MB). DB extracts and aggregates multi-scale features for the right ventricle. MB mainly relies on multiple effective field-of-views to detect objects at different sizes and fill boundary features. Different from previous networks, we used DB and MB to replace the convolution layer in the encoding layer, thus, we can gather multi-scale information of right ventricle, detect different size targets and fill boundary information in each encoding layer. In addition, in the decoding layer, we used DB to replace the convolution layer, so that we can aggregate the multi-scale features of the right ventricle in each decoding layer. Furthermore, the two-stage U-net structure is used to further improve the utilization of DB and MB through a two-layer encoding/decoding layer. Our method is validated on the RVSC, a public right ventricular data set. The results demonstrated that TSU-net achieved an average Dice coefficient of 0.86 on endocardium and 0.90 on the epicardium, thereby outperforming other models. It effectively assists doctors to diagnose the disease and promotes the development of medical images. In addition, we also provide an intuitive explanation of our network, which fully explain MB and TSU-net's ability to detect targets of different sizes and fill in boundary features.

The 4D B-spline method of calculating left ventricular functional parameters of cardiac MRI to evaluate myocardial injury of the apical segment in patients with myocarditis: a case-controlled observational study

Background

Myocarditis does not have typical clinical manifestations and thus is difficult to accurately diagnose by virtue of infection history, and electrocardiogram (EKG) and peripheral blood abnormalities. Endomyocardial biopsy is the gold standard for diagnosis of myocarditis, but is invasive, high risk, and has an observational blind area. Cardiac magnetic resonance imaging (CMRI) is multiparameter and multidirectional with high spatial resolution and high contrast of soft tissue. However, the optimal method of calculating left ventricular (LV) function in patients with apical-segment-injured myocarditis is unresolved. We compared and analyzed the differences between two different methods (Simpson and 4D B-spline surface model (known as the 4D method)) of measuring LV function by CMRI in patients with myocarditis in the 17th segment of the left ventricle.

Methods

The basic clinical data of two groups (myocarditis and non-myocarditis) were statistically analyzed, and differences in the LV function parameters by the two imaging methods were compared in the myocarditis group. Receiver-operating characteristic curves of single parameters and combined parameters based on the Simpson and 4D methods were drawn and the area under the curve, diagnostic threshold, maximum sensitivity interval, and maximum specificity interval were calculated.

Results

In the myocarditis and non-myocarditis groups the respective number of patients was 22 and 17, the percentage of males was 54.55% and 47.06%, and the average age was 32.20±11.59 and 43.06±11.62 years. The difference in LV ejection fraction (LVEF) (P=0.033) and LV end systolic volume (LVESV) (P=0.030) in the myocarditis group was statistically significant. The respective AUCs based on the Simpson and 4D methods were LVEF 0.602 vs. 0.778, LVESV 0.556 vs. 0.751, LVEF-and-LVESV 0.634 vs. 0.775. Based on the 4D method, the diagnostic thresholds of LVEF and LVESV were 34.965 (sensitivity 0.882, specificity 0.591) and 69.090 (sensitivity 0.727, specificity 0.706), the maximum sensitivity intervals of LVEF and LVESV were (24.610, 27.450) and (35.355, 37.200), and the maximum specificity intervals of LVEF and LVESV were (60.530, 65.625) and (91.625, 95.835), respectively.

Conclusions

Compared with the Simpson method, the 4D method might be more effective for CMRI diagnosis of apical-segment-injured myocarditis. When the Simpson method is used, LVEF combined with LVESV is recommended for comprehensive evaluation to improve diagnostic efficiency. When the 4D method is used, LVEF might be the preferred parameter for evaluation of LV function.

Dilated-Inception Net: Multi-Scale Feature Aggregation for Cardiac Right Ventricle Segmentation

Segmentation of cardiac ventricle from magnetic resonance images is significant for cardiac disease diagnosis, progression assessment, and monitoring cardiac conditions. Manual segmentation is so time consuming, tedious, and subjective that automated segmentation methods are highly desired in practice. However, conventional segmentation methods performed poorly in cardiac ventricle, especially in the right ventricle. Compared with the left ventricle, whose shape is a simple thick-walled circle, the structure of the right ventricle is more complex due to ambiguous boundary, irregular cavity, and variable crescent shape. Hence, effective feature extractors and segmentation models are preferred. In this paper, we propose a dilated-inception net (DIN) to extract and aggregate multi-scale features for right ventricle segmentation. The DIN outperforms many state-of-the-art models on the benchmark database of right ventricle segmentation challenge. In addition, the experimental results indicate that the proposed model has potential to reach expert-level performance in right ventricular epicardium segmentation. More importantly, DIN behaves similarly to clinical expert with high correlation coefficients in four clinical cardiac indices. Therefore, the proposed DIN is promising for automated cardiac right ventricle segmentation in clinical applications.

Decision Support Tools, Systems, and Artificial Intelligence in Cardiac Imaging

Noninvasive cardiac imaging is widely used for the diagnosis and management of cardiac patients. The increasing demand for cardiac imaging begins to exceed the number of available interpreting physicians, leaving less time to interpret studies. In addition, the busy clinician is facing the increasingly daunting task of keeping abreast of current medical advancements and the ongoing changes in disease diagnosis and therapy. Committing to memory and recalling such large volumes of information is challenging and is responsible for difficulties in adopting the rapid changes in imaging practice, and is likely partially responsible for errors in patient diagnosis and management. Diagnostic errors rank high in the cause of death in the United States, and are more common than any other medical error and are responsible for most malpractice claims. Most of these errors are related to cognitive errors. The use of artificial intelligence systems that can serve as complementary methods to assist humans with decision making can potentially prevent these errors. The past decades witnessed the development and integration of these tools, which can assist physicians with image interpretation. These tools work to optimize image quality for better visualization and accompany all imaging modalities, starting from patient selection for the appropriate test, patient preparation, image acquisition, processing, and finally interpretation. Current and future directions for technologies that support cardiac imaging physicians are discussed in this review.

A deep Boltzmann machine-driven level set method for heart motion tracking using cine MRI images

Heart motion tracking for radiation therapy treatment planning can result in effective motion management strategies to minimize radiation-induced cardiotoxicity. However, automatic heart motion tracking is challenging due to factors that include the complex spatial relationship between the heart and its neighboring structures, dynamic changes in heart shape, and limited image contrast, resolution, and volume coverage. In this study, we developed and evaluated a deep generative shape model-driven level set method to address these challenges. The proposed heart motion tracking method makes use of a heart shape model that characterizes the statistical variations in heart shapes present in a training data set. This heart shape model was established by training a three-layered deep Boltzmann machine (DBM) in order to characterize both local and global heart shape variations. During the tracking phase, a distance regularized level-set evolution (DRLSE) method was applied to delineate the heart contour on each frame of a cine MRI image sequence. The trained shape model was embedded into the DRLSE method as a shape prior term to constrain an evolutional shape to reach the desired heart boundary. Frame-by-frame heart motion tracking was achieved by iteratively mapping the obtained heart contour for each frame to the next frame as a reliable initialization, and performing a level-set evolution. The performance of the proposed motion tracking method was demonstrated using thirty-eight coronal cine MRI image sequences.

Multi-Views Fusion CNN for Left Ventricular Volumes Estimation on Cardiac MR Images

OBJECTIVE

Left ventricular (LV) volume estimation is a critical procedure for cardiac disease diagnosis. The objective of this paper is to address a direct LV volume prediction task.

METHODS

In this paper, we propose a direct volume prediction method based on the end-to-end deep convolutional neural networks. We study the end-to-end LV volume prediction method in items of the data preprocessing, network structure, and multiview fusion strategy. The main contributions of this paper are the following aspects. First, we propose a new data preprocessing method on cardiac magnetic resonance (CMR). Second, we propose a new network structure for end-to-end LV volume estimation. Third, we explore the representational capacity of different slices and propose a fusion strategy to improve the prediction accuracy.

RESULTS

The evaluation results show that the proposed method outperforms other state-of-the-art LV volume estimation methods on the open accessible benchmark datasets. The clinical indexes derived from the predicted volumes agree well with the ground truth ( ${\rm{EDV:R}}^{{\rm 2}}={\text{0.974}}$, ${\rm{RMSE\,}}= {\text{9.6}}{\rm{\,ml}}$; ${\rm{ESV:R}}^{{\rm 2}}={\text{0.976}}$, ${\rm{RMSE}}= {\text{7.1}}\,{\text{ml}}$; ${\rm{EF:R}}^{{\rm 2}} ={\text{0.828}}$, ${\rm{RMSE}}= {\text{4.71}}\% $).

CONCLUSION

Experimental results prove that the proposed method may be useful for the LV volume prediction task.

SIGNIFICANCE

The proposed method not only has application potential for cardiac diseases screening for large-scale CMR data, but also can be extended to other medical image research fields.

A review of heart chamber segmentation for structural and functional analysis using cardiac magnetic resonance imaging

Cardiovascular magnetic resonance (CMR) has become a key imaging modality in clinical cardiology practice due to its unique capabilities for non-invasive imaging of the cardiac chambers and great vessels. A wide range of CMR sequences have been developed to assess various aspects of cardiac structure and function, and significant advances have also been made in terms of imaging quality and acquisition times. A lot of research has been dedicated to the development of global and regional quantitative CMR indices that help the distinction between health and pathology. The goal of this review paper is to discuss the structural and functional CMR indices that have been proposed thus far for clinical assessment of the cardiac chambers. We include indices definitions, the requirements for the calculations, exemplar applications in cardiovascular diseases, and the corresponding normal ranges. Furthermore, we review the most recent state-of-the art techniques for the automatic segmentation of the cardiac boundaries, which are necessary for the calculation of the CMR indices. Finally, we provide a detailed discussion of the existing literature and of the future challenges that need to be addressed to enable a more robust and comprehensive assessment of the cardiac chambers in clinical practice.

Medical image analysis with artificial neural networks

Given that neural networks have been widely reported in the research community of medical imaging, we provide a focused literature survey on recent neural network developments in computer-aided diagnosis, medical image segmentation and edge detection towards visual content analysis, and medical image registration for its pre-processing and post-processing, with the aims of increasing awareness of how neural networks can be applied to these areas and to provide a foundation for further research and practical development. Representative techniques and algorithms are explained in detail to provide inspiring examples illustrating: (i) how a known neural network with fixed structure and training procedure could be applied to resolve a medical imaging problem; (ii) how medical images could be analysed, processed, and characterised by neural networks; and (iii) how neural networks could be expanded further to resolve problems relevant to medical imaging. In the concluding section, a highlight of comparisons among many neural network applications is included to provide a global view on computational intelligence with neural networks in medical imaging.

3D left ventricular segmentation using double active contours and double active surfaces

We propose a 3D left ventricular segmentation method for cardiac MR images. There are two steps in our method. In the first step, we improve our double active contours for segmenting the endocardial and the epicardial boundaries simultaneously without requiring any training sets. In the second step, we designed a double active surface method to segment the volume within the endocardium and the thickness of the myocardium simultaneously from a set of 2D contours generated from the first step. Our results show successful simultaneous segmentation of the 3D left ventricular model of the heart.

Myocardial delineation via registration in a polar coordinate system1

RATIONALE AND OBJECTIVES

Cardiovascular disease is the number one cause of premature death in the western world. Analysis of cardiac function provides clinically useful diagnostic and prognostic information; however, manual analysis of function via delineation is prohibitively time consuming. This article describes a technique for analysis of dynamic magnetic resonance images of the left ventricle using a non-rigid registration algorithm. A manually delineated contour of a single phase was propagated through the dynamic sequence.

MATERIALS AND METHODS

Short-axis cine magnetic resonance images were resampled into polar coordinates before all the time frames were aligned using a non-rigid registration algorithm. The technique was tested on 10 patient data sets, a total of 1,052 images were analyzed.

RESULTS

Results of this approach were investigated and compared with manual delineation at all phases in the cardiac cycle, and with registration performed in a Cartesian coordinate system. The results correlated very well with manually delineated contours.

CONCLUSION

A novel approach to the registration and subsequent delineation of cardiac magnetic resonance images has been introduced. For the endocardium, the polar resampling technique correlated well with manual delineation, and better than for images registered without radial resampling in a Cartesian coordinate system. For the epicardium, the difference was not as apparent with both techniques correlating well.

Detecting objects in image sequences using rule-based control in an active contour model

A method is presented for tracking object boundaries in sequential images using an active contour model, based on fuzzy reasoning rule-based control. Evolution of contour segments is controlled by separate processes based on whether the segment is judged to be inside, outside, or near the boundary of the object, leading to robust boundary detection.