[show abstract][hide abstract] ABSTRACT: The purpose of this study was to develop and validate a method for automated segmentation of the carotid artery lumen from volumetric MR Angiographic (MRA) images using a deformable tubular 3D Non-Uniform Rational B-Splines (NURBS) model. A flexible 3D tubular NURBS model was designed to delineate the carotid arterial lumen. User interaction was allowed to guide the model by placement of forbidden areas. Contrast-enhanced MRA (CE-MRA) from 21 patients with carotid atherosclerotic disease were included in this study. The validation was performed against expert drawn contours on multi-planar reformatted image slices perpendicular to the artery. Excellent linear correlations were found on cross-sectional area measurement (r = 0.98, P < 0.05) and on luminal diameter (r = 0.98, P < 0.05). Strong match in terms of the Dice similarity indices were achieved: 0.95 ± 0.02 (common carotid artery), 0.90 ± 0.07 (internal carotid artery), 0.87 ± 0.07 (external carotid artery), 0.88 ± 0.09 (carotid bifurcation) and 0.75 ± 0.20 (stenosed segments). Slight overestimation of stenosis grading by the automated method was observed. The mean differences was 7.20% (SD = 21.00%) and 5.2% (SD = 21.96%) when validated against two observers. Reproducibility in stenosis grade calculation by the automated method was high; the mean difference between two repeated analyses was 1.9 ± 7.3%. In conclusion, the automated method shows high potential for clinical application in the analysis of CE-MRA of carotid arteries.
The international journal of cardiovascular imaging 12/2011; 28(6):1513-24. · 2.15 Impact Factor
[show abstract][hide abstract] ABSTRACT: To correlate an automated regional wall motion abnormality (RWMA) detection method based on combined rest and dobutamine-stress cardiac MRI with the assessment of myocardial infarction from contrast-enhanced MRI (CE-MRI), and to demonstrate that adding stress data improves the detection of scar segments compared with rest data alone.
An automated RWMA detection method was built based on a statistical model of normokinetic myocardium from 41 healthy volunteers. The method was adapted to detect changes in RWMA from rest to stress. Twelve patients with myocardial infarction were included in the experiment. The correlation with CE-MRI was performed on two measurements: infarct transmurality and scar detection.
Compared with infarct transmurality, the probability of normokinetic motion decreased progressively as infarct transmurality increased. These probability values were 0.59 for non-scar segments, for <25% transmurality was 0.4 (SE=0.04), for 25-50% was 0.33 (SE=0.03), for 50-75% was 0.21 (SE=0.03) and for ≥75% was 0.10 (SE=0.03). For scar tissue detection, adding stress data significantly improved the performance (P<0.001, confidence interval=99.9%). The sensitivity, specificity, and accuracy increased by 34%, 30%, and 32%, respectively. The area under the receiver operating characteristics curve was 0.63 when rest-only data was used, but it was improved to 0.87 when stress data was added.
The presented automated RWMA assessment was capable of detecting wall motion improvements from rest to stress. The method correlated well with infarct transmurality from CE-MRI. Detection of scar regions was more accurate when rest and stress data were combined compared with rest data alone.
Journal of Magnetic Resonance Imaging 08/2011; 34(2):270-8. · 2.57 Impact Factor
[show abstract][hide abstract] ABSTRACT: This paper presents an automated method for regional wall motion abnormality detection (RWMA) from rest and stress cardiac
MRI. The automated RWMA detection is based on a statistical shape model of myocardial contraction trained on slice-based myocardial
contours from in ED and ES. A combination of rigid and non-rigid registrations is introduced to align a patient shape to the
normokinetic myocardium model, where pure contractility information is kept. The automated RWMA method is applied to identify
potentially infarcted myocardial segments from rest–stress MRI alone.
In this study, 41 cardiac MRI studies of healthy subjects were used to build the statistical normokinetic model, while 12
myocardial infarct patients were included for validation. The rest–stress data produced a better separation between scar and
normal segments compared to the rest–only data. The sensitivity, specificity and accuracy were increased by 34%, 30%, and
32%, respectively. The area under the ROC curve for the rest–stress data was improved to 0.87 compared to 0.63 for the rest–only
Functional Imaging and Modeling of the Heart - 6th International Conference, FIMH 2011, New York City, NY, USA, May 25-27, 2011. Proceedings; 01/2011
[show abstract][hide abstract] ABSTRACT: Interactive visualization is required to inspect and monitor the automatic segmentation of vessels derived from contrast-enhanced magnetic resonance angiography (CE-MRA). A dual-view visualization scheme consisting of curved planar reformation (CPR) and direct volume rendering (DVR) was developed for this purpose and tested.
A dual view visualization scheme was developed using the vessel pathline for both camera position and rotation in 3D, greatly reducing the degrees of freedom (DOF) required for navigation. Pathline-based navigation facilitates coupling of the CPR and DVR views, as local position and orientation can be matched precisely. The new technique was compared to traditional techniques in a user study. Layperson users were required to perform a visual search task that involves checking for (minor) errors in segmentations of MRA data from a software phantom. The task requires the user to examine both views.
Pathline-based navigation and coupling of CPR and DVR provide user speed performance improvements in a vessel inspection task. Interactive MRA visualization with this method, where rotational degrees of freedom were reduced, had no negative effect.
The DOF reduction achieved by the new navigation technique is beneficial to user performance. The technique is promising and merits comprehensive evaluation in a realistic clinical setting.
International Journal of Computer Assisted Radiology and Surgery 09/2010; 6(5):591-9. · 1.36 Impact Factor
[show abstract][hide abstract] ABSTRACT: We describe a series of experiments that compared 2D/3D input methods for selection and positioning tasks related to medical image analysis. For our study, we chose a switchable P5 Glove Controller, which can be used to provide both 2DOF and 6DOF input control. Our results suggest that for both tasks the overall performance and accuracy can be improved when the input device with more degrees of freedom (DOF) is used for manipulation of the visualized medical data. 3D input turned out to be more beneficial for the positioning task than for the selection task. In order to determine a potential source of the difference in the task completion time between 2D and 3D input, we also investigated whether there was a significant difference between 2DOF and 6DOF input methods with regard to the time spent on task-specific basic manipulations.
International Journal of Human-Computer Studies 01/2010; 68(6):355-369. · 1.42 Impact Factor
[show abstract][hide abstract] ABSTRACT: User-centered design is often performed with- out regard to individual user differences. In this paper, we report results of an empirical study aimed to evaluate whether computer experience and demographic user char- acteristics would have an effect on the way people interact with the visualized medical data in a 3D virtual environ- ment using 2D and 3D input devices. We analyzed the interaction through performance data, questionnaires and observations. The results suggest that differences in gender, age and game experience have an effect on people's behavior and task performance, as well as on subjective user preferences.
[show abstract][hide abstract] ABSTRACT: We study the effectiveness of stereoscopy and smooth motion as 3D cues for medical interpretation of vascular structures as obtained by 3D medical imaging techniques. We designed a user study where the user has to follow a path in a mazelike solid shaded 3D structure. The user controls rotation of the model. We measure user performance in terms of time taken and error rate. The experiment was executed with 32 (medical and non-medical) users. The results show that motion cue is more important than stereoscopy, and that stereoscopy has no added value when motion is already present, which is not consistent with previous experiments.
Proceedings of the International Conference on Advanced Visual Interfaces, AVI 2010, Roma, Italy, May 26-28, 2010; 01/2010
[show abstract][hide abstract] ABSTRACT: Vascular disease diagnosis often requires a precise segmentation of the vessel lumen. When 3D (Magnetic Resonance Angiography, MRA, or Computed Tomography Angiography, CTA) imaging is available, this can be done automatically, but occasional errors are inevitable. So, the segmentation has to be checked by clinicians. This requires appropriate visualisation techniques. A number of visualisation techniques exist, but there has been little in the way of user studies that compare the different alternatives. In this study we examine how users interact with several basic visualisations, when performing a visual search task, checking vascular segmentation correctness of segmented MRA data. These visualisations are: direct volume rendering (DVR), isosurface rendering, and curved planar reformatting (CPR). Additionally, we examine if visual highlighting of potential errors can help the user find errors, so a fourth visualisation we examine is DVR with visual highlighting. Our main findings are that CPR performs fastest but has higher error rate, and there are no significant differences between the other three visualisations. We did find that visual highlighting actually has slower performance in early trials, suggesting that users learned to ignore them.
IMAGAPP 2010 - Proceedings of the International Conference on Imaging Theory and Applications and IVAPP 2010 - Proceedings of the International Conference on Information Visualization Theory and Applications, Angers, France, May 17 - 21, 2010; 01/2010
[show abstract][hide abstract] ABSTRACT: We describe a series of experiments that compared the 2D and 3D input methods for selection and positioning tasks related to medical image analysis. For this study, we chose a switchable P5 glove controller, which can be used to provide both 2DOF and 6DOF input control. Our results suggest that for both tasks the overall completion time and accuracy can be improved when the input device with more degrees of freedom is used for manipulation of the visualized medical data. However, 3D input appeared to be more beneficial for the positioning task than for the selection task.
Visualisation, 2009. VIZ '09. Second International Conference in; 08/2009
[show abstract][hide abstract] ABSTRACT: 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.
IEEE transactions on medical imaging. 03/2009; 28(4):595-607.
[show abstract][hide abstract] ABSTRACT: In this chapter we focus on the quantitative extraction of small differences in an image sequence caused by motion, and in
an image pair by differences in depth. We like to extract the local motion parameters as a small local shift over time or
space. We call the resulting vectorfield the optic flow from the image sequence, a spatio-temporal feature, and we call the resulting vectorfield the disparity map for the stereo pair. As the application of the method described in this chapter is virtually the same for stereo disparity
extraction, we will focus in the treatment on spatio-temporal optic flow.
[show abstract][hide abstract] ABSTRACT: This work investigates knowledge driven segmentation of cardiac MR perfusion sequences. We build upon previous work on multi-band AAMs to integrate into the segmentation both spatial priors about myocardial shape as well as temporal priors about characteristic perfusion patterns. Different temporal and spatial features are developed without a strict need for temporal correspondence across the image sequences. We also investigate which combination of spatial and temporal features yields the best segmentation performance. Our evaluation criteria were boundary errors wrt manual segmentations, area overlap, and convergence envelope. From a quantitative evaluation on 19 perfusion studies, we conclude that a combination of the maximum intensity projection feature and gradient orientation map yields the best segmentation performance, with an average point-to-curve error of 0.9-1 pixel wrt manual contours. We also conclude that addition of different temporal features does not necessarily increase performance.
[show abstract][hide abstract] ABSTRACT: We propose an operational method to extract the left ventricle (LV) systole dynamics using harmonic phase (HARP) images extracted from tagged cardiac MR sequences. Established techniques to generate HARP sequences provide independent evidence for motion extraction, in the sense that the combined linear system for scalar brightness conservation, applied to the HARP images, can be uniquely solved for a dense field of motion parameters without the need for regularization. In contrast to some of the previously proposed popular methods, no segmentation or tracking of tags over time, nor interpolation of a sparse motion field explicitly coupled to the tag pattern is required, and the problem of tag fading is bypassed. An important novelty is the incorporation of automatic local scale selection so as to obtain a robust solution, which not only yields a stable, but also a smoothly varying motion field of the (healthy) LV myocardial wall. The scheme relies on an integer parameter representing order of approximation, and allows one to simultaneously obtain a dense field of differential tensors capturing the low order differential structure of the motion field, which is useful for the computation of relevant local quantities such as strain rates and material acceleration fields. The methodology is generic and straightforward to implement, and can be generalized to 3D and, in principle, to account for higher order differential structure.
[show abstract][hide abstract] ABSTRACT: Heart disease can negatively influence cardiac pump func- tion. To assess cardiac tissue function, a method based on classical opti- cal flow theory applied in the spectral domain is presented. Assumption of pixel intensity conservation is replaced by assumption of spatial phase conservation. Simultaneous application to two independent observations of the same optical flow field removes the necessity of additional con- straints (i.e. flow field smoothness, normal flow) to solve the optical flow constraint equation (OFCE). Using the 1st order Taylor expansion of the OFCE, our system yields not only pixel displacements, but also the 1st order differential structure of the displacements (i.e. strains), which otherwise should be calculated as a post-processing step. Operation at pixel level obviates the need for interpolation of tag lines or sparse flow field representation. Experiments show coherent flow fields of a human cardiac systole. Comparison with velocity encoded MRI shows a good resemblance.
Journal of Process Control 01/2007; · 1.81 Impact Factor
[show abstract][hide abstract] ABSTRACT: An important assessment in patients with ischemic heart disease is whether myocardial contractility may improve after treatment.
The prediction of myocardial contractility improvement is generally performed under physical or pharmalogical stress conditions.
In this paper, we present a technique to build a statistical model of healthy myocardial contraction using independent component
analysis. The model is used to detect regions with abnormal contraction in patients both during rest and stress.