Time Continuous Tracking and Segmentation of Cardiovascular Magnetic Resonance Images Using Multidimensional Dynamic Programming

CyCoSeg: A Cyclic Collaborative Framework for Automated Medical Image Segmentation

Deep neural networks have been tremendously successful at segmenting objects in images. However, it has been shown they still have limitations on challenging problems such as the segmentation of medical images. The main reason behind this lower success resides in the reduced size of the object in the image. In this paper we overcome this limitation through a cyclic collaborative framework, CyCoSeg. The proposed framework is based on a deep active shape model (D-ASM), which provides prior information about the shape of the object, and a semantic segmentation network (SSN). These two models collaborate to reach the desired segmentation by influencing each other: SSN helps D-ASM identify relevant keypoints in the image through an Expectation Maximization formulation, while D-ASM provides a segmentation proposal that guides the SSN. This cycle is repeated until both models converge. Extensive experimental evaluation shows CyCoSeg boosts the performance of the baseline models, including several popular SSNs, while avoiding major architectural modifications. The effectiveness of our method is demonstrated on the left ventricle segmentation on two benchmark datasets, where our approach achieves one of the most competitive results in segmentation accuracy. Furthermore, its generalization is demonstrated for lungs and kidneys segmentation in CT scans.

Fine-grained calibrated double-attention convolutional network for left ventricular segmentation

Objective. Left ventricular (LV) segmentation of cardiac magnetic resonance imaging (MRI) is essential for diagnosing and treating the early stage of heart diseases. In convolutional neural networks, the target information of the LV in feature maps may be lost with convolution and max-pooling, particularly at the end of systolic. Fine segmentation of ventricular contour is still a challenge, and it may cause problems with inaccurate calculation of clinical parameters (e.g. ventricular volume). In order to improve the similarity of the neural network output and the target segmentation region, in this paper, a fine-grained calibrated double-attention convolutional network (FCDA-Net) is proposed to finely segment the endocardium and epicardium from ventricular MRI. Approach. FCDA-Net takes the U-net as the backbone network, and the encoder-decoder structure incorporates a double grouped-attention module that is constructed by a fine calibration spatial attention module (fcSAM) and a fine calibration channel attention module (fcCAM). The double grouped-attention mechanism enhances the expression of information in both spatial and channelwise feature maps to achieve fine calibration. Main Results. The proposed approach is evaluated on the public MICCAI 2009 challenge dataset, and ablation experiments are conducted to demonstrate the effect of each grouped-attention module. Compared with other advanced segmentation methods, FCDA-Net can obtain better LV segmentation performance. Significance. The LV segmentation results of MRI can be used to perform more accurate quantitative analysis of many essential clinical parameters and it can play an important role in image-guided clinical surgery.

Ultrafast Computation of Left Ventricular Ejection Fraction by Using Temporal Intensity Variation in Cine Cardiac Magnetic Resonance

Cardiac magnetic resonance enables comprehensive cardiac evaluation; however, intense time and labor requirements for data acquisition and processing have discouraged many clinicians from using it. We have developed an alternative image-processing algorithm that requires minimal user interaction: an ultrafast algorithm that computes left ventricular ejection fraction (LVEF) by using temporal intensity variation in cine balanced steady-state free precession (bSSFP) short-axis images, with or without contrast medium. We evaluated the algorithm's performance against an expert observer's analysis for segmenting the LV cavity in 65 study participants (LVEF range, 12%-70%). In 12 instances, contrast medium was administered before cine imaging. Bland-Altman analysis revealed quantitative effects of LV basal, midcavity, and apical morphologic variation on the algorithm's accuracy. Total computation time for the LV stack was <2.5 seconds. The algorithm accurately delineated endocardial boundaries in 1,132 of 1,216 slices (93%). When contours in the extreme basal and apical slices were not adequate, they were replaced with manually drawn contours. The Bland-Altman mean differences were <1.2 mL (0.8%) for end-diastolic volume, <5 mL (6%) for end-systolic volume, and <3% for LVEF. Standard deviation of the difference was ≤4.1% of LV volume for all sections except the midcavity in end-systole (8.3% of end-systolic volume). We conclude that temporal intensity variation-based ultrafast LVEF computation is clinically accurate across a range of LV shapes and wall motions and is suitable for postcontrast cine SSFP imaging. Our algorithm enables real-time processing of cine bSSFP images on a commercial scanner console within 3 seconds in an unobtrusive automated process.

Artificial intelligence in pediatric and adult congenital cardiac MRI: an unmet clinical need

Cardiac MRI (CMR) allows non-invasive, non-ionizing assessment of cardiac function and anatomy in patients with congenital heart disease (CHD). The utility of CMR as a non-invasive imaging tool for evaluation of CHD have been growing exponentially over the past decade. The algorithms based on artificial intelligence (AI), and in particular, deep learning, have rapidly become a methodology of choice for analyzing CMR. A wide range of applications for AI have been developed to tackle challenges in various aspects of CMR, and significant advances have also been made from image acquisition to image analysis and diagnosis. We include an overview of AI definitions, different architectures, and details on well-known methods. This paper reviews the major deep learning concepts used for analyses of patients with CHD. In the end, we have summarized a list of open challenges and concerns to be considered for future studies.

Quantification of aortic pulse wave velocity from a population based cohort: a fully automatic method

BACKGROUND

Aortic pulse wave velocity (PWV) is an indicator of aortic stiffness and is used as a predictor of adverse cardiovascular events. PWV can be non-invasively assessed using magnetic resonance imaging (MRI). PWV computation requires two components, the length of the aortic arch and the time taken for the systolic pressure wave to travel through the aortic arch. The aortic length is calculated using a multi-slice 3D scan and the transit time is computed using a 2D velocity encoded MRI (VE) scan. In this study we present and evaluate an automatic method to quantify the aortic pulse wave velocity using a large population-based cohort.

METHODS

For this study 212 subjects were retrospectively selected from a large multi-center heart-brain connection cohort. For each subject a multi-slice 3D scan of the aorta was acquired in an oblique-sagittal plane and a 2D VE scan acquired in a transverse plane cutting through the proximal ascending and descending aorta. PWV was calculated in three stages: (i) a multi-atlas-based segmentation method was developed to segment the aortic arch from the multi-slice 3D scan and subsequently estimate the length of the proximal aorta, (ii) an algorithm that delineates the proximal ascending and descending aorta from the time-resolved 2D VE scan and subsequently obtains the velocity-time flow curves was also developed, and (iii) automatic methods that can compute the transit time from the velocity-time flow curves were implemented and investigated. Finally the PWV was obtained by combining the aortic length and the transit time.

RESULTS

Quantitative evaluation with respect to the length of the aortic arch as well as the computed PWV were performend by comparing the results of the novel automatic method to those obtained manually. The mean absolute difference in aortic length obtained automatically as compared to those obtained manually was 3.3 ± 2.8 mm (p < 0.05), the manual inter-observer variability on a subset of 45 scans was 3.4 ± 3.4 mm (p = 0.49). Bland-Altman analysis between the automataic method and the manual methods showed a bias of 0.0 (-5.0,5.0) m/s for the foot-to-foot approach, -0.1 (-1.2, 1.1) and -0.2 (-2.6, 2.1) m/s for the half-max and the cross-correlation methods, respectively.

CONCLUSION

We proposed and evaluated a fully automatic method to calculate the PWV on a large set of multi-center MRI scans. It was observed that the overall results obtained had very good agreement with manual analysis. Our proposed automatic method would be very beneficial for large population based studies, where manual analysis requires a lot of manpower.

Objective selection of short-axis slices for automated quantification of left ventricular size and function by cardiovascular magnetic resonance

BACKGROUND

Quantification of left ventricular (LV) volume from cardiovascular magnetic resonance images relies on subjective and often challenging selection of short-axis (SAX) slices. We hypothesized that this could be solved by defining mitral annular (MA) plane and apex in long-axis (LAX) views, which could be combined with automated LV volume analysis that does not rely on manual tracing of the endocardial border.

METHODS

SAX images from 50 subjects were analyzed using custom software. LV apex and insertion points of the mitral leaflets were marked on LAX views and used to approximate MA plane. End-systolic and end-diastolic LV volumes (ESV, EDV) were measured while including only slices or their parts located between MA plane and LV apex. Endocardial borders were automatically detected using our previously validated algorithm and also manually traced to obtain reference values.

RESULTS

Selection of anatomic landmarks in LAX views allowed automated measurement of LV volumes without the need for subjective slice selection. Intertechnique comparisons resulted in high correlations (EDV: r=0.95; ESV: r=0.96) and small biases (1 and 9ml). Combined three-dimensional displays of LAX and SAX views with the MA plane showed that in 7/10 worst cases, intertechnique discordance was due to incorrect manual tracing at LV base that erroneously included part of atrial cavity in LV volume or excluded part of LV cavity, i.e., incorrect reference values.

CONCLUSION

Defining the MA plane and apex in the LAX views obviates the need for subjective slice selection and eliminates errors in LV volume measurements.

Quantification of bound microbubbles in ultrasound molecular imaging

Molecular markers associated with diseases can be visualized and quantified noninvasively with targeted ultrasound contrast agent (t-UCA) consisting of microbubbles (MBs) that can bind to specific molecular targets. Techniques used for quantifying t-UCA assume that all unbound MBs are taken out of the blood pool few minutes after injection and only MBs bound to the molecular markers remain. However, differences in physiology, diseases, and experimental conditions can increase the longevity of unbound MBs. In such conditions, unbound MBs will falsely be quantified as bound MBs. We have developed a novel technique to distinguish and classify bound from unbound MBs. In the post-processing steps, first, tissue motion was compensated using block-matching (BM) techniques. To preserve only stationary contrast signals, a minimum intensity projection (MinIP) or 20th-percentile intensity projection (PerIP) was applied. The after-flash MinIP or PerIP was subtracted from the before-flash MinIP or PerIP. In this way, tissue artifacts in contrast images were suppressed. In the next step, bound MB candidates were detected. Finally, detected objects were tracked to classify the candidates as unbound or bound MBs based on their displacement. This technique was validated in vitro, followed by two in vivo experiments in mice. Tumors (n = 2) and salivary glands of hypercholesterolemic mice (n = 8) were imaged using a commercially available scanner. Boluses of 100 μL of a commercially available t-UCA targeted to angiogenesis markers and untargeted control UCA were injected separately. Our results show considerable reduction in misclassification of unbound MBs as bound ones. Using our method, the ratio of bound MBs in salivary gland for images with targeted UCA versus control UCA was improved by up to two times compared with unprocessed images.

Automatic segmentation of the left ventricle in cardiac MRI using local binary fitting model and dynamic programming techniques

Segmentation of the left ventricle is very important to quantitatively analyze global and regional cardiac function from magnetic resonance. The aim of this study is to develop a novel algorithm for segmenting left ventricle on short-axis cardiac magnetic resonance images (MRI) to improve the performance of computer-aided diagnosis (CAD) systems. In this research, an automatic segmentation method for left ventricle is proposed on the basis of local binary fitting (LBF) model and dynamic programming techniques. The validation experiments are performed on a pool of data sets of 45 cases. For both endo- and epi-cardial contours of our results, percentage of good contours is about 93.5%, the average perpendicular distance are about 2 mm. The overlapping dice metric is about 0.91. The regression and determination coefficient between the experts and our proposed method on the LV mass is 1.038 and 0.9033, respectively; they are 1.076 and 0.9386 for ejection fraction (EF). The proposed segmentation method shows the better performance and has great potential in improving the accuracy of computer-aided diagnosis systems in cardiovascular diseases.

Fully automatic detection of salient features in 3-d transesophageal images

Most automated segmentation approaches to the mitral valve and left ventricle in 3-D echocardiography require a manual initialization. In this article, we propose a fully automatic scheme to initialize a multicavity segmentation approach in 3-D transesophageal echocardiography by detecting the left ventricle long axis, the mitral valve and the aortic valve location. Our approach uses a probabilistic and structural tissue classification to find structures such as the mitral and aortic valves; the Hough transform for circles to find the center of the left ventricle; and multidimensional dynamic programming to find the best position for the left ventricle long axis. For accuracy and agreement assessment, the proposed method was evaluated in 19 patients with respect to manual landmarks and as initialization of a multicavity segmentation approach for the left ventricle, the right ventricle, the left atrium, the right atrium and the aorta. The segmentation results revealed no statistically significant differences between manual and automated initialization in a paired t-test (p > 0.05). Additionally, small biases between manual and automated initialization were detected in the Bland-Altman analysis (bias, variance) for the left ventricle (-0.04, 0.10); right ventricle (-0.07, 0.18); left atrium (-0.01, 0.03); right atrium (-0.04, 0.13); and aorta (-0.05, 0.14). These results indicate that the proposed approach provides robust and accurate detection to initialize a multicavity segmentation approach without any user interaction.

Method to create regional mechanical dyssynchrony maps from short-axis cine steady-state free-precession images

PURPOSE

To develop a robust method to assess regional mechanical dyssynchrony from cine short-axis MR images. Cardiac resynchronization therapy (CRT) is an effective treatment for patients with heart failure and evidence of left-ventricular (LV) dyssynchrony. Patient response to CRT is greatest when the LV pacing lead is placed in the most dyssynchronous segment. Existing techniques for assessing regional dyssynchrony require difficult acquisition and/or postprocessing. Our goal was to develop a widely applicable and robust method to assess regional mechanical dyssynchrony.

MATERIALS AND METHODS

Using the endocardial boundary, radial displacement curves (RDCs) were generated throughout the LV. Cross-correlation was used to determine the delay time between each RDC and a patient-specific reference. Delay times were projected onto the American Heart Association 17-segment model creating a regional dyssynchrony map. Our method was tested in 10 normal individuals and 10 patients enrolled for CRT (QRS > 120 ms, NYHA III-IV, EF < 35%).

RESULTS

Delay times over the LV were 23.9 ± 33.8 ms and 93.1 ± 99.9 ms (P < 0.001) in normal subjects and patients, respectively. Interobserver reproducibility for segment averages was 6.8 ± 39.3 ms and there was 70% agreement in identifying the latest contracting segment.

CONCLUSION

We have developed a method that can reliably calculate regional delay times from cine steady-state free-precession (SSFP) images. Maps of regional dyssynchrony could be used to identify the latest-contracting segment to assist in CRT lead implantation.

Automatic functional analysis of left ventricle in cardiac cine MRI

RATIONALE AND OBJECTIVES

A fully automated left ventricle segmentation method for the functional analysis of cine short axis (SAX) magnetic resonance (MR) images was developed, and its performance evaluated with 133 studies of subjects with diverse pathology: ischemic heart failure (n=34), non-ischemic heart failure (n=30), hypertrophy (n=32), and healthy (n=37).

MATERIALS AND METHODS

The proposed automatic method locates the left ventricle (LV), then for each image detects the contours of the endocardium, epicardium, papillary muscles and trabeculations. Manually and automatically determined contours and functional parameters were compared quantitatively.

RESULTS

There was no significant difference between automatically and manually determined end systolic volume (ESV), end diastolic volume (EDV), ejection fraction (EF) and left ventricular mass (LVM) for each of the four groups (paired sample t-test, α=0.05). The automatically determined functional parameters showed high correlations with those derived from manual contours, and the Bland-Altman analysis biases were small (1.51 mL, 1.69 mL, -0.02%, -0.66 g for ESV, EDV, EF and LVM, respectively).

CONCLUSIONS

The proposed technique automatically and rapidly detects endocardial, epicardial, papillary muscles' and trabeculations' contours providing accurate and reproducible quantitative MRI parameters, including LV mass and EF.

Automatic left ventricle segmentation in cardiac MRI using topological stable-state thresholding and region restricted dynamic programming

RATIONALE AND OBJECTIVES

Segmentation of the left ventricle (LV) is very important in the assessment of cardiac functional parameters. The aim of this study is to develop a novel and robust algorithm which can improve the accuracy of automatic LV segmentation on short-axis cardiac magnetic resonance images (MRI).

MATERIALS AND METHODS

The database used in this study consists of 45 cases obtained from the Sunnybrook Health Sciences Centre. The 45 cases contain 12 ischemic heart failures, 12 non-ischemic heart failures, 12 LV hypertrophies, and 9 normal cases. Three key techniques are developed in this segmentation algorithm: 1) topological stable-state thresholding method is proposed to refine the endocardial contour, 2) an edge map with non-maxima gradient suppression approach, and 3) a region-restricted technique that is proposed to improve the dynamic programming to derive the epicardial boundary.

RESULTS

The validation experiments were performed on a pool of data sets of 45 cases. For both endo- and epicardial contours of our results, percentage of good contours is about 91%, the average perpendicular distance is about 2 mm, and the overlapping dice metric is about 0.91. The regression and determination coefficient for the experts and our proposed method on the ejection fraction is 1.05 and 0.9048, respectively; they are 0.98 and 0.8221 for LV mass.

CONCLUSIONS

An automatic method using topological stable-state thresholding and region restricted dynamic programming has been proposed to segment left ventricle in short-axis cardiac MRI. Evaluation results indicate that the proposed segmentation method can improve the accuracy and robust of left ventricle segmentation. The proposed segmentation approach shows the better performance and has great potential in improving the accuracy of computer-aided diagnosis systems in cardiovascular diseases.

Abnormal right atrial and right ventricular diastolic function relate to impaired clinical condition in patients operated for tetralogy of Fallot

BACKGROUND

Atrial enlargement may reflect ventricular diastolic dysfunction. Although patients with tetralogy of Fallot (TOF) have been studied extensively, little is known about atrial size and function. We assessed bi-atrial size and function in patients after TOF repair, and related them to biventricular systolic and diastolic function, and clinical parameters.

METHODS

51 Patients (21 ± 8 years) and 30 healthy controls (31 ± 7 years) were included and underwent magnetic resonance imaging to assess bi-atrial and biventricular size, systolic and diastolic function. Patients also underwent exercise testing, and N-terminal prohormone brain natriuretic peptide (NT-proBNP) assessment.

RESULTS

In patients, right atrial (RA) minimal volume (34 ± 8 ml/m(2) vs. 28 ± 8 ml/m(2), p=0.001) and late emptying fraction were increased; RA early emptying fraction was decreased. Patients had longer right ventricular (RV) deceleration time (0.24 ± 0.10 vs. 0.13 ± 0.04, p<0.001), reflecting impaired RV relaxation, and larger RV volumes. Patients with end-diastolic forward flow (EDFF) had larger RA and RV size, abnormal RA emptying, higher NT-proBNP levels, higher VE/VCO2 slope (ventilatory response to carbon dioxide production), and the most abnormal LV diastolic function (impaired compliance). Patients with abnormal RA emptying (reservoir function <30% and pump function >24%) had higher NT-proBNP levels and worse exercise capacity. RA minimal volume was associated with RV end-diastolic volume (r=0.35, p=0.013).

CONCLUSIONS

In TOF patients with moderate RV dilatation, abnormal bi-atrial function and biventricular diastolic dysfunction are common. Abnormal RA emptying was associated with signs of impaired clinical condition, as was the presence of EDFF. These parameters, together with RA enlargement, could serve as useful markers for clinically relevant RV diastolic dysfunction.

An image-based comprehensive approach for automatic segmentation of left ventricle from cardiac short axis cine MR images

Segmentation of the left ventricle is important in the assessment of cardiac functional parameters. Manual segmentation of cardiac cine MR images for acquiring these parameters is time-consuming. Accuracy and automation are the two important criteria in improving cardiac image segmentation methods. In this paper, we present a comprehensive approach to segment the left ventricle from short axis cine cardiac MR images automatically. Our method incorporates a number of image processing and analysis techniques including thresholding, edge detection, mathematical morphology, and image filtering to build an efficient process flow. This process flow makes use of various features in cardiac MR images to achieve high accurate segmentation results. Our method was tested on 45 clinical short axis cine cardiac images and the results are compared with manual delineated ground truth (average perpendicular distance of contours near 2 mm and mean myocardium mass overlapping over 90%). This approach provides cardiac radiologists a practical method for an accurate segmentation of the left ventricle.

Left ventricular modelling: a quantitative functional assessment tool based on cardiac magnetic resonance imaging

We present the development and testing of a semi-automated tool to support the diagnosis of left ventricle (LV) dysfunctions from cardiac magnetic resonance (CMR). CMR short-axis images of the LVs were obtained in 15 patients and processed to detect endocardial and epicardial contours and compute volume, mass and regional wall motion (WM). Results were compared with those obtained from manual tracing by an expert cardiologist. Nearest neighbour tracking and finite-element theory were merged to calculate local myocardial strains and torsion. The method was tested on a virtual phantom, on a healthy LV and on two ischaemic LVs with different severity of the pathology. Automated analysis of CMR data was feasible in 13/15 patients: computed LV volumes and wall mass correlated well with manually extracted data. The detection of regional WM abnormalities showed good sensitivity (77.8%), specificity (85.1%) and accuracy (82%). On the virtual phantom, computed local strains differed by less than 14 per cent from the results of commercial finite-element solver. Strain calculation on the healthy LV showed uniform and synchronized circumferential strains, with peak shortening of about 20 per cent at end systole, progressively higher systolic wall thickening going from base to apex, and a 10° torsion. In the two pathological LVs, synchronicity and homogeneity were partially lost, anomalies being more evident for the more severely injured LV. Moreover, LV torsion was dramatically reduced. Preliminary testing confirmed the validity of our approach, which allowed for the fast analysis of LV function, even though future improvements are possible.

Registration of 4D cardiac CT sequences under trajectory constraints with multichannel diffeomorphic demons

We propose a framework for the nonlinear spatiotemporal registration of 4D time-series of images based on the Diffeomorphic Demons (DD) algorithm. In this framework, the 4D spatiotemporal registration is decoupled into a 4D temporal registration, defined as mapping physiological states, and a 4D spatial registration, defined as mapping trajectories of physical points. Our contribution focuses more specifically on the 4D spatial registration that should be consistent over time as opposed to 3D registration that solely aims at mapping homologous points at a given time-point. First, we estimate in each sequence the motion displacement field, which is a dense representation of the point trajectories we want to register. Then, we perform simultaneously 3D registrations of corresponding time-points with the constraints to map the same physical points over time called the trajectory constraints. Under these constraints, we show that the 4D spatial registration can be formulated as a multichannel registration of 3D images. To solve it, we propose a novel version of the Diffeomorphic Demons (DD) algorithm extended to vector-valued 3D images, the Multichannel Diffeomorphic Demons (MDD). For evaluation, this framework is applied to the registration of 4D cardiac computed tomography (CT) sequences and compared to other standard methods with real patient data and synthetic data simulated from a physiologically realistic electromechanical cardiac model. Results show that the trajectory constraints act as a temporal regularization consistent with motion whereas the multichannel registration acts as a spatial regularization. Finally, using these trajectory constraints with multichannel registration yields the best compromise between registration accuracy, temporal and spatial smoothness, and computation times. A prospective example of application is also presented with the spatiotemporal registration of 4D cardiac CT sequences of the same patient before and after radiofrequency ablation (RFA) in case of atrial fibrillation (AF). The intersequence spatial transformations over a cardiac cycle allow to analyze and quantify the regression of left ventricular hypertrophy and its impact on the cardiac function.

Unsupervised fully automated inline analysis of global left ventricular function in CINE MR imaging

OBJECTIVES

To implement and evaluate the accuracy of unsupervised fully automated inline analysis of global ventricular function and myocardial mass (MM). To compare automated with manual segmentation in patients with cardiac disorders.

MATERIALS AND METHODS

In 50 patients, cine imaging of the left ventricle was performed with an accelerated retrogated steady state free precession sequence (GRAPPA; R = 2) on a 1.5 Tesla whole body scanner (MAGNETOM Avanto, Siemens Healthcare, Germany). A spatial resolution of 1.4 x 1.9 mm was achieved with a slice thickness of 8 mm and a temporal resolution of 42 milliseconds. Ventricular coverage was based on 9 to 12 short axis slices extending from the annulus of the mitral valve to the apex with 2 mm gaps. Fully automated segmentation and contouring was performed instantaneously after image acquisition. In addition to automated processing, cine data sets were also manually segmented using a semi-automated postprocessing software. Results of both methods were compared with regard to end-diastolic volume (EDV), end-systolic volume (ESV), ejection fraction (EF), and MM. A subgroup analysis was performed in patients with normal (> or =55%) and reduced EF (<55%) based on the results of the manual analysis.

RESULTS

Thirty-two percent of patients had a reduced left ventricular EF of <55%. Volumetric results of the automated inline analysis for EDV (r = 0.96), ESV (r = 0.95), EF (r = 0.89), and MM (r = 0.96) showed high correlation with the results of manual segmentation (all P < 0.001). Head-to-head comparison did not show significant differences between automated and manual evaluation for EDV (153.6 +/- 52.7 mL vs. 149.1 +/- 48.3 mL; P = 0.05), ESV (61.6 +/- 31.0 mL vs. 64.1 +/- 31.7 mL; P = 0.08), and EF (58.0 +/- 11.6% vs. 58.6 +/- 11.6%; P = 0.5). However, differences were significant for MM (150.0 +/- 61.3 g vs. 142.4 +/- 59.0 g; P < 0.01). The standard error was 15.6 (EDV), 9.7 (ESV), 5.0 (EF), and 17.1 (mass). The mean time for manual analysis was 15 minutes.

CONCLUSIONS

Unsupervised fully automated segmentation and contouring during image reconstruction enables an accurate evaluation of global systolic cardiac function.

A dual propagation contours technique for semi-automated assessment of systolic and diastolic cardiac function by CMR

BACKGROUND

Although cardiovascular magnetic resonance (CMR) is frequently performed to measure accurate LV volumes and ejection fractions, LV volume-time curves (VTC) derived ejection and filling rates are not routinely calculated due to lack of robust LV segmentation techniques. VTC derived peak filling rates can be used to accurately assess LV diastolic function, an important clinical parameter. We developed a novel geometry-independent dual-contour propagation technique, making use of LV endocardial contours manually drawn at end systole and end diastole, to compute VTC and measured LV ejection and filling rates in hypertensive patients and normal volunteers.

METHODS

39 normal volunteers and 49 hypertensive patients underwent CMR. LV contours were manually drawn on all time frames in 18 normal volunteers. The dual-contour propagation algorithm was used to propagate contours throughout the cardiac cycle. The results were compared to those obtained with single-contour propagation (using either end-diastolic or end-systolic contours) and commercially available software. We then used the dual-contour propagation technique to measure peak ejection rate (PER) and peak early diastolic and late diastolic filling rates (ePFR and aPFR) in all normal volunteers and hypertensive patients.

RESULTS

Compared to single-contour propagation methods and the commercial method, VTC by dual-contour propagation showed significantly better agreement with manually-derived VTC. Ejection and filling rates by dual-contour propagation agreed with manual (dual-contour - manual PER: -0.12 +/- 0.08; ePFR: -0.07 +/- 0.07; aPFR: 0.06 +/- 0.03 EDV/s, all P = NS). However, the time for the manual method was approximately 4 hours per study versus approximately 7 minutes for dual-contour propagation. LV systolic function measured by LVEF and PER did not differ between normal volunteers and hypertensive patients. However, ePFR was lower in hypertensive patients vs. normal volunteers, while aPFR was higher, indicative of altered diastolic filling rates in hypertensive patients.

CONCLUSION

Dual-propagated contours can accurately measure both systolic and diastolic volumetric indices that can be applied in a routine clinical CMR environment. With dual-contour propagation, the user interaction that is routinely performed to measure LVEF is leveraged to obtain additional clinically relevant parameters.

Automated frame-by-frame endocardial border detection from cardiac magnetic resonance images for quantitative assessment of left ventricular function: validation and clinical feasibility

PURPOSE

To develop a technique based on image noise distribution for automated endocardial border detection from cardiac magnetic resonance (CMR) images throughout the cardiac cycle, validate it, and test its clinical utility.

MATERIALS AND METHODS

Images obtained in 36 patients were analyzed using custom software to obtain left ventricular (LV) volume throughout the cardiac cycle, end-systolic and end-diastolic LV volumes, and ejection fraction (EF). Validation against manually-traced endocardial boundaries included intertechnique comparisons of LV volumes, slice areas, and border positions. Then, the clinical feasibility of the dynamic automated analysis of LV function was tested in 14 patients with normal LV function, 12 patients with systolic dysfunction, and 10 patients with diastolic dysfunction.

RESULTS

Analysis time for one cardiac cycle was <15 minutes. Intertechnique comparisons resulted in high correlation (r>0.96), small biases (volumes: -6 mL; EF: 4.6%) and narrow limits of agreement (volumes: 17.6 mL; EF: 9.2%). We found significant intergroup differences in multiple quantitative indices of systolic and diastolic function.

CONCLUSION

Fast, automated, dynamic detection of LV endocardial boundaries is feasible and allows accurate quantification of LV size and function, which is potentially clinically useful for objective assessment of systolic and diastolic dysfunction.

Automated detection of regional wall motion abnormalities based on a statistical model applied to multislice short-axis cardiac MR images

In this paper, a statistical shape analysis method for myocardial contraction is presented that was built to detect and locate regional wall motion abnormalities (RWMA). For each slice level (base, middle, and apex), 44 short-axis magnetic resonance images were selected from healthy volunteers to train a statistical model of normal myocardial contraction using independent component analysis (ICA). A classification algorithm was constructed from the ICA components to automatically detect and localize abnormally contracting regions of the myocardium. The algorithm was validated on 45 patients suffering from ischemic heart disease. Two validations were performed; one with visual wall motion scores (VWMS) and the other with wall thickening (WT) used as references. Accuracy of the ICA-based method on each slice level was 69.93% (base), 89.63% (middle), and 72.78% (apex) when WT was used as reference, and 63.70% (base), 67.41% (middle), and 66.67% (apex) when VWMS was used as reference. From this we conclude that the proposed method is a promising diagnostic support tool to assist clinicians in reducing the subjectivity in VWMS.

Automatic image-driven segmentation of the ventricles in cardiac cine MRI

PURPOSE

To propose and to evaluate a novel method for the automatic segmentation of the heart's two ventricles from dynamic ("cine") short-axis "steady state free precession" (SSFP) MR images. This segmentation task is of significant clinical importance. Previously published automated methods have various disadvantages for routine clinical use.

MATERIALS AND METHODS

The proposed method is primarily image-driven: it exploits the spatiotemporal information provided by modern 3D+time SSFP cardiac MRI, and makes only few and plausible assumptions about the image acquisition and about the imaged heart. Specifically, the method does not require previously trained statistical shape models or gray-level appearance models, as often used by other methods.

RESULTS

The performance of the segmentation method was demonstrated through a qualitative visual validation on 32 clinical exams: no gross failures for the left-ventricle (right-ventricle) on 31 (29) of the exams were found. A validation of resulting quantitative cardiac functional parameters showed good agreement with a manual quantification of 19 clinical exams.

CONCLUSION

The proposed method is feasible, fast, and robust against anatomical variability and image contrast variations.

Addition of the long-axis information to short-axis contours reduces interstudy variability of left-ventricular analysis in cardiac magnetic resonance studies

OBJECTIVES

To reduce interstudy variability using long-axis information for correcting short-axis (SA) contours at basal and apical level for left-ventricular analysis by magnetic resonance imaging.

MATERIALS AND METHODS

A total of 20 patients with documented heart failure and 20 volunteers underwent magnetic resonance imaging examination twice for measuring endocardial end-diastolic volume, endocardial end-systolic volume, mass, and ejection fraction. The boundary of the left ventricle, the mitral valve plane, and apex were marked manually on the 2- and 4-chamber long-axis images. Automatic epicardial and endocardial contour detection was performed on the SA images using the intersection of the outlines from the long axis as starting positions. The same observer compared the interstudy variability of this method with analysis that was based on the SA images only.

RESULTS

The interstudy variability decreased when information from the long axis was included; for end-systolic volume, 9.6% versus 4.7% (P = 0.00014); for end-diastolic volume, 4.9% versus 2.5% (P = 0.0011); for mass, 7.4% versus 5.0% (P = 0.11); and for ejection fraction 12.2% versus 5.6% (P = 0.0017), respectively.

CONCLUSIONS

Identification of the mitral valve plane and apex on long-axis images to limit the extent of volume at the base and the apex of the heart reduces interstudy variability for left-ventricular functional assessment.

Time continuous detection of the left ventricular long axis and the mitral valve plane in 3-D echocardiography

Automated segmentation approaches for the left ventricle (LV) in 3-D echocardiography (3DE) often rely on manual initialization. So far, little effort has been put into automating the initialization procedure to get to a fully automatic segmentation approach. We propose a fully automatic method for the detection of the LV long axis (LAX) and the mitral valve plane (MVP) over the full cardiac cycle, for the initialization of segmentation algorithms in 3DE. Our method exploits the cyclic motion of the LV and therefore detects salient structures in a time-continuous way. Probabilities to candidate LV center points are assigned through a Hough transform for circles. The LV LAX is detected by combining dynamic programming detections on these probabilities in 3-D and 2D + time to obtain a time continuous solution. Subsequently, the mitral valve plane is detected in a projection of the data on a plane through the previously detected LAX. The method easily adjusts to different acquisition routines and combines robustness with good accuracy and low computational costs. Automatic detection was evaluated using patient data acquired with the fast rotating ultrasound (FRU) transducer (n=11 patients) and with the Philips Sonos 7500 ultrasound system (Philips Medical Systems, Andover, MA, USA), with the X4 matrix transducer (n=14 patients). For the FRU-transducer data, the LAX was estimated with a distance error of 2.85+/-1.70 mm (mean+/-SD) and an angle of 5.25+/-3.17 degrees; the mitral valve plane was estimated with a distance of -1.54+/-4.31 mm. For the matrix data, these distances were 1.96+/-1.30 mm with an angle error of 5.95+/-2.11 and -1.66+/-5.27 mm for the mitral valve plane. These results confirm that the method is very suitable for automatic detection of the LV LAX and MVP. It provides a basis for further automatic exploration of the LV and could therefore serve as a replacement of manual initialization of 3-D segmentation approaches.

Advances in magnetic resonance (2007)

Advances in clinical magnetic resonance (MR) are discussed in this review in the context of publications from Investigative Radiology during 2006 and 2007. The articles relevant to this topic, published during this 2 year time period, are considered as organized by anatomic region. An additional final focus of discussion is in regards to those studies involving MR contrast media.

Automated tracking of the mitral valve annulus motion in apical echocardiographic images using multidimensional dynamic programming

We developed a semiautomatic method for tracking the mitral valve annulus (MVA) in echocardiographic images, in particular, tracking the septal and the lateral mitral valve hinge points. The algorithm is based on multidimensional dynamic programming combined with apodized block matching. The method was tested on single-beat apical four chamber image sequences of 20 patients with acute myocardial infarction. The automated tracking results were evaluated by comparing them with the average manual tracking results of two experts. The mitral valve hinge point displacements and the total mitral excursions obtained by the automatic technique agreed well with those obtained manually and outperformed two commonly used tracking methods (forward tracking and minimum tracking). In conclusion, this novel semiautomatic tracking method is clinically valuable and capable of tracking the MVA motion within the limits of interobserver variability. The technique is robust, even in low frame rate, redigitized VCR images of clinical quality.

Advances in Magnetic Resonance (2006)

Advances in the field of magnetic resonance (MR) as it pertains to clinical diagnostic radiology are examined in this review on the basis of publications in Investigative Radiology over the past 2 years (2005-2006). The articles published during that timeframe are discussed, organizationally wise, by anatomic region with an additional focus on studies involving MR contrast media.