Automatic cardiac LV segmentation in MRI using modified graph cuts with smoothness and interslice constraints

Quantification and 3D localization of magnetically navigated superparamagnetic particles using MRI in phantom and swine chemoembolization models

OBJECTIVE

Superparamagnetic nanoparticles (SPIONs) can be combined with tumor chemoembolization agents to form magnetic drug-eluting beads (MDEBs), which are navigated magnetically in the MRI scanner through the vascular system. We aim to develop a method to accurately quantify and localize these particles and to validate the method in phantoms and swine models.

METHODS

MDEBs were made of Fe₃O₄ SPIONs. After injected known numbers of MDEBs, susceptibility artifacts in three-dimensional (3D) volumetric interpolated breath-hold examination (VIBE) sequences were acquired in glass and Polyvinyl alcohol (PVA) phantoms, and two living swine. Image processing of VIBE images provided the volume relationship between MDEBs and their artifact at different VIBE acquisitions and post-processing parameters. Simulated hepatic-artery embolization was performed in vivo with an MRI-conditional magnetic-injection system, using the volume relationship to locate and quantify MDEB distribution.

RESULTS

Individual MDEBs were spatially identified, and their artifacts quantified, showing no correlation with magnetic-field orientation or sequence bandwidth, but exhibiting a relationship with echo time and providing a linear volume relationship. Two MDEB aggregates were magnetically steered into desired liver regions while the other 19 had no steering, and 25 aggregates were injected into another swine without steering. The MDEBs were spatially identified and the volume relationship showed accuracy in assessing the number of the MDEBs, with small errors (8.8%).

CONCLUSION AND SIGNIFICANCE

MDEBs were able to be steered into desired body regions and then localized using 3D VIBE sequences. The resulting volume relationship was linear, robust, and allowed for quantitative analysis of the MDEB distribution.

Recent Advances in Fibrosis and Scar Segmentation From Cardiac MRI: A State-of-the-Art Review and Future Perspectives

Segmentation of cardiac fibrosis and scars is essential for clinical diagnosis and can provide invaluable guidance for the treatment of cardiac diseases. Late Gadolinium enhancement (LGE) cardiovascular magnetic resonance (CMR) has been successful in guiding the clinical diagnosis and treatment reliably. For LGE CMR, many methods have demonstrated success in accurately segmenting scarring regions. Co-registration with other non-contrast-agent (non-CA) modalities [e.g., balanced steady-state free precession (bSSFP) cine magnetic resonance imaging (MRI)] can further enhance the efficacy of automated segmentation of cardiac anatomies. Many conventional methods have been proposed to provide automated or semi-automated segmentation of scars. With the development of deep learning in recent years, we can also see more advanced methods that are more efficient in providing more accurate segmentations. This paper conducts a state-of-the-art review of conventional and current state-of-the-art approaches utilizing different modalities for accurate cardiac fibrosis and scar segmentation.

Convexity preserving level set for left ventricle segmentation

In clinical applications of cardiac left ventricle (LV) segmentation, the segmented LV is desired to include the cavity, trabeculae, and papillary muscles, which form a convex shape. However, the intensities of trabeculae and papillary muscles are similar to myocardium. Consequently, segmentation algorithms may easily misclassify trabeculae and papillary muscles as myocardium. In this paper, we propose a level set method with a convexity preserving mechanism to ensure the convexity of the segmented LV. In the proposed level set method, the curvature of the level set contours is used to control their convexity, such that the level set contour is finally deformed as a convex shape. The experimental results and the comparison with other level set methods show the advantage of our method in terms of segmentation accuracy. Compared with the state-of-the-art methods using deep-learning, our method is able to achieve comparable segmentation accuracy without the need for training, while the deep-learning based method requires a large set of training data and high-quality manual segmentation. Therefore, our method can be conveniently used in situation where training data and their manual segmentation are not available.

Multitask Learning for Estimating Multitype Cardiac Indices in MRI and CT Based on Adversarial Reverse Mapping

The estimation of multitype cardiac indices from cardiac magnetic resonance imaging (MRI) and computed tomography (CT) images attracts great attention because of its clinical potential for comprehensive function assessment. However, the most exiting model can only work in one imaging modality (MRI or CT) without transferable capability. In this article, we propose the multitask learning method with the reverse inferring for estimating multitype cardiac indices in MRI and CT. Different from the existing forward inferring methods, our method builds a reverse mapping network that maps the multitype cardiac indices to cardiac images. The task dependencies are then learned and shared to multitask learning networks using an adversarial training approach. Finally, we transfer the parameters learned from MRI to CT. A series of experiments were conducted in which we first optimized the performance of our framework via ten-fold cross-validation of over 2900 cardiac MRI images. Then, the fine-tuned network was run on an independent data set with 2360 cardiac CT images. The results of all the experiments conducted on the proposed adversarial reverse mapping show excellent performance in estimating multitype cardiac indices.

Automatic estimation of aortic and mitral valve displacements in dynamic CTA with 4D graph-cuts

The location of the mitral and aortic valves in dynamic cardiac imaging is useful for extracting functional derived parameters such as ejection fraction, valve excursions, and global longitudinal strain, and when performing anatomical structures tracking using slice following or valve intervention's planning. Completely automatic segmentation methods are still challenging tasks because of their fast movements and the different positions that prevent good visibility of the leaflets along the full cardiac cycle. In this article, we propose a processing pipeline to track the displacement of the aortic and mitral valve annuli from high-resolution cardiac four-dimensional computed tomographic angiography (4D-CTA). The proposed method is based on the dynamic separation of left ventricle, left atrium and aorta using statistical shape modeling and an energy minimization algorithm based on graph-cuts and has been evaluated on a set of 15 electrocardiography-gated 4D-CTAs. We report a mean agreement distance between manual annotations and our proposed method of 2.52±1.06 mm for the mitral annulus and 2.00±0.69 mm for the aortic valve annulus based on valve locations detected from manual anatomical landmarks. In addition, we show the effect of detecting the valvular planes on derived functional parameters (ejection fraction, global longitudinal strain, and excursions of the mitral and aortic valves).

Left ventricle segmentation using a Bayesian approach with distance dependent shape priors

We propose a method for segmentation of the left ventricle in magnetic resonance cardiac images. The framework consists of an initial Bayesian segmentation of the central slice of the volume. This segmentation is used to locate a shape prior for the LV myocardial tissue. This shape prior is determined using the fact that the myocardium is approximately annular as seen in the short-axis. Then a second Bayesian segmentation is performed to obtain the final result. This procedure is repeated for the rest of the slices. An extrapolation of the area of the LV is used to determine a stopping criterion. The method was evaluated on the databases of the Cardiac Atlas project. Our results demonstrate a suitable accuracy for myocardial segmentation (≈0.8 Dice's coefficient). For the endocardium and the epicardium the Dice's coefficients are 0.94 and 0.9 respectively. The accuracy was also evaluated in terms of the Hausdorff distance and the average distance. For the myocardium we obtain 8 mm and 2 mm respectively. Our results demonstrate the capability and merits of the proposed method to estimate the structure of the LV. The method requires minimal user input and generates results with quality comparable to more complex approaches. This paper suggests a new efficient approach for automatic LV quantification based on a Bayesian technique with shape priors with errors comparable to state-of-the-art techniques.

Pyramid feature adaptation for semi-supervised cardiac bi-ventricle segmentation

Cardiac bi-ventricle segmentation (BVS) is an essential task for assessing cardiac indices, such as the ejection fraction and volume of the left ventricle (LV) and right ventricle (RV). However, BVS is extremely challenging due to the high variability of the bi-ventricle structure and lack of labeled data. In this paper, we propose a pyramid feature adaptation based semi-supervised method (PABVS) for cardiac bi-ventricle segmentation. The PABVS first extracts the multiscale pyramid features of bi-ventricle structure to cope with the high variability of bi-ventricle structure. Then, a weighted pyramid feature adaptation strategy is proposed to ensure a smooth feature space among labeled data and unlabeled data. In particular, the PABVS performs weighted feature adaptation at each level of a multiscale pyramid feature based on adversarial learning. It gives less importance to outlier feature layers of labeled data and more importance to representative layers. The experimental results on magnetic resonance images show that our proposed PABVS can achieve Dice values 0.915 for EpiLV with 40% labeled data and the Dice values 0.976 for EpiLV with all labeled data, which outperforms mainstream semi-supervised methods. This endows our PABVS with great potential for the effective clinical application of BVS.

Topographic analysis of eyelid position using digital image processing software

PURPOSE

To propose a novel analysis technique for objective quantification of topographic eyelid position with an algorithmatically calculated scheme and to determine its feasibility.

METHODS

One hundred normal eyelids from 100 patients were segmented using a graph cut algorithm, and 11 shape features of eyelids were semi-automatically quantified using in-house software. To evaluate the intra- and inter-examiner reliability of this software, intra-class correlation coefficients (ICCs) were used. To evaluate the diagnostic value of this scheme, the correlations between semi-automatic and manual measurements of margin reflex distance 1 (MRD1) and margin reflex distance 2 (MRD2) were analysed using a Bland-Altman analysis. To determine the degree of agreement according to manual MRD length, the relationship between the variance of semi-automatic measurements and the manual measurements was evaluated using linear regression.

RESULTS

Intra- and inter-examiner reliability were excellent, with ICCs ranging from 0.913 to 0.980 in 11 shape features including MRD1, MRD2, palpebral fissure, lid perimeter, upper and lower lid lengths, roundness, total area, and medial, central, and lateral areas. The correlations between semi-automatic and manual MRDs were also excellent, with better correlation in MRD1 than in MRD2 (R = 0.893 and 0.823, respectively). In addition, significant positive relationships were observed between the variance and the length of MRD1 and 2; the longer the MRD length, the more the variance.

CONCLUSION

The proposed novel optimized integrative scheme, which is shown to have high repeatability and reproducibility, is useful for topographic analysis of eyelid position.

Sources of variability in quantification of cardiovascular magnetic resonance infarct size - reproducibility among three core laboratories

BACKGROUND

Acute myocardial infarct (AMI) size depicted by late gadolinium enhancement cardiovascular magnetic resonance (CMR) is increasingly used as an efficacy endpoint in randomized trials comparing AMI therapies. Infarct size is quantified using manual planimetry (MANUAL), visual scoring (VISUAL), or automated techniques using signal-intensity thresholding (AUTO). Although AUTO is considered the most reproducible, prior studies did not account for the subjective determination of endocardial/epicardial borders, which all methods require. For MANUAL and VISUAL, prior studies did not address how to treat intermediate signal intensities due to partial volume.

METHODS

To assess sources of variability, AMI size was measured in 30 patients and 12 controls by 3 core-laboratories using 8 methods, each separated by more than 2 months time (n = 720 evaluations). The methods were: (1,2) AUTOSegment, AUTOFWHM (using Segment software or the full-width-at-half-maximum algorithm, respectively); (3,4) AUTO-UCSegment, AUTO-UCFWHM (user correction for endocardial border pixels, no-reflow, etc.); (5) MANUAL; (6) MANUAL-ISI (adjustment for intermediate signal-intensities); (7) VISUAL; (8) VISUAL-ISI.

RESULTS

Mean infarct size varied between 16.8% and 27.2% of LV mass depending on method. Even automated techniques with no user interaction for infarct borders resulted in significant within-patient variability given the need to subjectively trace endocardial/epicardial contours. The coefficient-of-variation (CV) was 10.6% and 14.6% for AUTOSegment and AUTOFWHM, respectively. For manual and visual categories, reproducibility was improved when intermediate signal-intensities were considered (MANUAL-ISI vs MANUAL: CV = 8.3% vs 14.4%; p = 0.03; VISUAL-ISI vs VISUAL: CV = 8.4% vs 10.9%; p = 0.01). For AUTO-UCSegment, MANUAL-ISI, and VISUAL-ISI (best technique in each category) within-patient variability due to the quantification method was less than 10% of total variability, and the required sample sizes for detecting a 5% absolute difference in infarct size were 62, 63, and 62 patients, respectively.

CONCLUSION

Among CMR core-laboratories, an important source of variability in infarct size quantification is the subjective delineation of endocardial/epicardial borders. When intermediate signal intensities are considered in manual planimetry and visual scoring, reproducibility and impact on sample size are similar to automated techniques.

Segmentation of medical images using mean value guided contour

Partial differential equation-based (PDE-based) methods are extensively used in image segmentation, especially in contour models. Difficulties associated with the boundaries, namely troubles with developing initialization, inadequate convergence to boundary concavities, and difficulties connected to saddle points and stationary points of active contours make the contour models suffer from a feeble performance of referring to complex geometries. The present paper is designed to take advantage of mean value theorem rather than minimizing energy function for contours. It is efficiently capable of resolving above-mentioned problems by applying this theorem to the edge map gradient vectors, which is calculated from the image. Since the contour is computed in a straightforward manner from an edge map instead of force balance equation, it varies from other contour-based image segmentation methods. To illustrate the ability of the proposed model in complex geometries and ruptures, several experiments were also provided to validate the model. The experiments' results demonstrated that the proposed method, which is called mean value guided contour (MVGC), is capable of repositioning contours into boundary concavities and has suitable forcefulness in complex geometries.

An SPCNN-GVF-based approach for the automatic segmentation of left ventricle in cardiac cine MR images

PURPOSE

Accurate segmentation of left ventricle (LV) is essential for the cardiac function analysis. However, it is labor intensive and time consuming for radiologists to delineate LV boundary manually. In this paper, we present a novel self-correcting framework for the fully automatic LV segmentation.

METHODS

Firstly, a time-domain method is designed to extract a rectangular region of interest around the heart. Then, the simplified pulse-coupled neural network (SPCNN) is employed to locate the LV cavity. Different from the existing approaches, SPCNN can realize the self-correcting segmentation due to its parameter controllability. Subsequently, the post-processing based on the maximum gradient searching is proposed to obtain the accurate endocardium. Finally, a new external force based on the shape similarity is defined and integrated into the gradient vector flow (GVF) snake with the balloon force to segment the epicardium.

RESULTS

We obtain encouraging segmentation results tested on the database provided by MICCAI 2009. The average percentage of good contours is 92.26 %, the average perpendicular distance is 2.38 mm, and the overlapping dice metric is 0.89. Besides, the experiment results show good correlations between the automatic segmentation and the manual delineation (for the LV ejection fraction and the LV myocardial mass, the correlation coefficients R are 0.9683 and 0.9278, respectively).

CONCLUSION

We propose an effective and fast method combing the SPCNN and the improved GVF for the automatic segmentation of LV.

Feasibility of automated 3-dimensional magnetic resonance imaging pancreas segmentation

Purpose

With the advent of magnetic resonance imaging (MRI) guided radiation therapy, internal organ motion can be imaged simultaneously during treatment. In this study, we evaluate the feasibility of pancreas MRI segmentation using state-of-the-art segmentation methods.

Methods and materials

T2-weighted half-Fourier acquisition single-shot turbo spin-echo and T1 weighted volumetric interpolated breath-hold examination images were acquired on 3 patients and 2 healthy volunteers for a total of 12 imaging volumes. A novel dictionary learning (DL) method was used to segment the pancreas and compared to t mean-shift merging, distance regularized level set, and graph cuts, and the segmentation results were compared with manual contours using Dice's index, Hausdorff distance, and shift of the center of the organ (SHIFT).

Results

All volumetric interpolated breath-hold examination images were successfully segmented by at least 1 of the autosegmentation method with Dice's index >0.83 and SHIFT ≤2 mm using the best automated segmentation method. The automated segmentation error of half-Fourier acquisition single-shot turbo spin-echo images was significantly greater. DL is statistically superior to the other methods in Dice’s overlapping index. For the Hausdorff distance and SHIFT measurement, distance regularized level set and DL performed slightly superior to the graph cuts method, and substantially superior to mean-shift merging. DL required least human supervision and was faster to compute.

Conclusions

Our study demonstrated potential feasibility of automated segmentation of the pancreas on MRI scans with minimal human supervision at the beginning of imaging acquisition. The achieved accuracy is promising for organ localization.

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.

Segmentation of the left ventricular endocardium from magnetic resonance images by using different statistical shape models

We evaluate in this paper different strategies for the construction of a statistical shape model (SSM) of the left ventricle (LV) to be used for segmentation in cardiac magnetic resonance (CMR) images. From a large database of LV surfaces obtained throughout the cardiac cycle from 3D echocardiographic (3DE) LV images, different LV shape models were built by varying the considered phase in the cardiac cycle and the registration procedure employed for surface alignment. Principal component analysis was computed to describe the statistical variability of the SSMs, which were then deformed by applying an active shape model (ASM) approach to segment the LV endocardium in CMR images of 45 patients. Segmentation performance was evaluated by comparing LV volumes derived by ASM segmentation with different SSMs and those obtained by manual tracing, considered as a reference. A high correlation (r(2)>0.92) was found in all cases, with better results when using the SSM models comprising more than one frame of the cardiac cycle.

The Localization and Characterization of Ischemic Scars in relation to the Infarct Related Coronary Artery Assessed by Cardiac Magnetic Resonance and a Novel Automatic Postprocessing Method

Aims. The correspondence between the localization and morphology of ischemic scars and the infarct related artery (IRA) by use of cardiac magnetic resonance imaging and a novel automatic postprocessing method. Methods and Results. Thirty-four patients with one-year-old single IRA myocardial infarction were examined. Endocardium, epicardium, and the point where right and left ventricles are coinciding were manually marked. All measurements were automatically assessed by the method. The following are results with manual assessments of scar properties in parenthesis: mean scar size (FWHM criterion): 7.8 ± 5.5 as %LV (17.4 ± 8.6%); mean endocardial extent of infarction: 44 ± 26° (124 ± 47°); mean endocardial extent of infarction as %LV circumference: 9.7 ± 7.0% (34.6 ± 13.0%); and mean transmurality: 52 ± 20% of LV wall thickness (77 ± 23%). Scars located in segments 1, 2, 7, 8, 13, and 14 by use of the automatic method were 96-100% specific for LAD occlusion. The highest specificities of RCA and LCX occlusions were segment 4 with 93% and segment 6 with 64%, respectively. The scar localization assessed automatically or manually was without major differences. Conclusion. The automatic method is applicable and able to assess localization, size, transmurality, and endocardial extent of ischemic scars.

Contrast-optimized composite image derived from multigradient echo cardiac magnetic resonance imaging improves reproducibility of myocardial contours and T2* measurement

Objectives

Reproducibility of myocardial contour determination in cardiac magnetic resonance imaging is important, especially when determining T2* values per myocardial segment as a prognostic factor of heart failure or thalassemia. A method creating a composite image with contrasts optimized for drawing myocardial contours is introduced and compared with the standard method on a single image.

Materials and methods

A total of 36 short-axis slices from bright-blood multigradient echo (MGE) T2* scans of 21 patients were acquired at eight echo times. Four observers drew free-hand myocardial contours on one manually selected T2* image (method 1) and on one image composed by blending three images acquired at TEs providing optimum contrast-to-noise ratio between the myocardium and its surrounding regions (method 2).

Results

Myocardial contouring by method 2 met higher interobserver reproducibility than method 1 (P < 0.001) with smaller Coefficient of variance (CoV) of T2* values in the presence of myocardial iron accumulation (9.79 vs. 15.91 %) and in both global myocardial and mid-ventricular septum regions (12.29 vs. 16.88 and 5.76 vs. 8.16 %, respectively).

Conclusion

The use of contrast-optimized composite images in MGE data analysis improves reproducibility of myocardial contour determination, leading to increased consistency in the calculated T2* values enhancing the diagnostic impact of this measure of iron overload.