Perry Radau

Evaluation of current algorithms for segmentation of scar tissue from late Gadolinium enhancement cardiovascular magnetic resonance of the left atrium: an open-access grand challenge

BackgroundLate Gadolinium enhancement (LGE) cardiovascular magnetic resonance (CMR) imaging can be used to visualise regions of fibrosis and scarring in the left atrium (LA) myocardium. This can be important for treatment stratification of patients with atrial fibrillation (AF) and for assessment of treatment after radio frequency catheter ablation (RFCA). In this paper we present a standardised evaluation benchmarking framework for algorithms segmenting fibrosis and scar from LGE CMR images. The algorithms reported are the response to an open challenge that was put to the medical imaging community through an ISBI (IEEE International Symposium on Biomedical Imaging) workshop.MethodsThe image database consisted of 60 multicenter, multivendor LGE CMR image datasets from patients with AF, with 30 images taken before and 30 after RFCA for the treatment of AF. A reference standard for scar and fibrosis was established by merging manual segmentations from three observers. Furthermore, scar was also quantified using 2, 3 and 4 standard deviations (SD) and full-width-at-half-maximum (FWHM) methods. Seven institutions responded to the challenge: Imperial College (IC), Mevis Fraunhofer (MV), Sunnybrook Health Sciences (SY), Harvard/Boston University (HB), Yale School of Medicine (YL), King’s College London (KCL) and Utah CARMA (UTA, UTB). There were 8 different algorithms evaluated in this study.ResultsSome algorithms were able to perform significantly better than SD and FWHM methods in both pre- and post-ablation imaging. Segmentation in pre-ablation images was challenging and good correlation with the reference standard was found in post-ablation images. Overlap scores (out of 100) with the reference standard were as follows: Pre: IC = 37, MV = 22, SY = 17, YL = 48, KCL = 30, UTA = 42, UTB = 45; Post: IC = 76, MV = 85, SY = 73, HB = 76, YL = 84, KCL = 78, UTA = 78, UTB = 72.ConclusionsThe study concludes that currently no algorithm is deemed clearly better than others. There is scope for further algorithmic developments in LA fibrosis and scar quantification from LGE CMR images. Benchmarking of future scar segmentation algorithms is thus important. The proposed benchmarking framework is made available as open-source and new participants can evaluate their algorithms via a web-based interface.

Innovations in imaging for chronic total occlusions: a glimpse into the future of angiography’s blind-spot

Chronic total occlusions (CTOs) are a subset of lesions that present a considerable burden to cardiovascular patients. There exists a strong clinical desire to improve non-surgical options for CTO revascularization. While several techniques, devices, and guide wires have been developed and refined for use in CTOs, the inability of angiography to adequately visualize occluded arterial segments makes interventions in this setting technically challenging. This review describes the current status of several invasive and non-invasive imaging techniques that may facilitate improved image guidance during CTO revascularization, with the goals of improving procedure safety and efficacy while reducing the time required to complete these interventions. Cardiac imaging also has important potential roles in selecting patients most likely to benefit from revascularization as well as pre-procedural planning, post-procedural assessment of revascularized segments and long-term outcomes studies. Modalities discussed include non-invasive techniques, such as CT(computed tomography) angiography and cardiac magnetic resonance imaging (MRI), as well as invasive techniques, such as intravascular ultrasound, optical coherence tomography, intravascular MRI, and conventional angiography. While some of these techniques have some evidence to support their use at present, others are at earlier stages of development. Strategies that combine imaging techniques with the use of interventional therapies may provide significant opportunities to improve results in CTO interventions and represent an active area of investigation.

Pattern Recognition of Abnormal Left Ventricle Wall Motion in Cardiac MR

There are four main problems that limit application of pattern recognition techniques for recognition of abnormal cardiac left ventricle (LV) wall motion: (1) Normalization of the LVs size, shape, intensity level and position; (2) defining a spatial correspondence between phases and subjects; (3) extracting features; (4) and discriminating abnormal from normal wall motion. Solving these four problems is required for application of pattern recognition techniques to classify the normal and abnormal LV wall motion. In this work, we introduce a normalization scheme to solve the first and second problems. With this scheme, LVs are normalized to the same position, size, and intensity level. Using the normalized images, we proposed an intra-segment classification criterion based on a correlation measure to solve the third and fourth problems. Application of the method to recognition of abnormal cardiac MR LV wall motion showed promising results.

VURTIGO: visualization platform for real-time, MRI-Guided cardiac electroanatomic mapping

Guidance of electrophysiological (EP) procedures by magnetic resonance imaging (MRI) has significant advantages over x-ray fluoroscopy. Display of electroanatomic mapping (EAM) during an intervention fused with a prior MR volume and DE-MRI derived tissue classification should improve the accuracy of cardiac resynchronization therapy (CRT) for ventricular arrhythmias. Improved accuracy in the spatial localization of recorded EP points will produce an EAM to constrain and customize patient-specific cardiac electroanatomic models being developed for understanding the patterns of arrhythmogenic slow conduction zones causing reentry circuits and treatment planning. The Vurtigo software presented here is a four dimensional (3D+time) real-time visualization application for guiding interventions capable of displaying prior volumes, real-time MRI scan planes, EAM (voltage or activation times), segmented models, and tracked catheters. This paper will describe the architecture and features of Vurtigo followed by the application example of guiding percutaneous cardiac electroanatomic mapping in porcine models.

Multiscale properties of weighted total variation flow with applications to denoising and registration

Images consist of structures of varying scales: large scale structures such as flat regions, and small scale structures such as noise, textures, and rapidly oscillatory patterns. In the hierarchical (BV, L(2)) image decomposition, Tadmor, et al. (2004) start with extracting coarse scale structures from a given image, and successively extract finer structures from the residuals in each step of the iterative decomposition. We propose to begin instead by extracting the finest structures from the given image and then proceed to extract increasingly coarser structures. In most images, noise could be considered as a fine scale structure. Thus, starting the image decomposition with finer scales, rather than large scales, leads to fast denoising. We note that our approach turns out to be equivalent to the nonstationary regularization in Scherzer and Weickert (2000). The continuous limit of this procedure leads to a time-scaled version of total variation flow. Motivated by specific clinical applications, we introduce an image depending weight in the regularization functional, and study the corresponding weighted TV flow. We show that the edge-preserving property of the multiscale representation of an input image obtained with the weighted TV flow can be enhanced and localized by appropriate choice of the weight. We use this in developing an efficient and edge-preserving denoising algorithm with control on speed and localization properties. We examine analytical properties of the weighted TV flow that give precise information about the denoising speed and the rate of change of energy of the images. An additional contribution of the paper is to use the images obtained at different scales for robust multiscale registration. We show that the inherently multiscale nature of the weighted TV flow improved performance for registration of noisy cardiac MRI images, compared to other methods such as bilateral or Gaussian filtering. A clinical application of the multiscale registration algorithm is also demonstrated for aligning viability assessment magnetic resonance (MR) images from 8 patients with previous myocardial infarctions.

Myocardial segmentation in late-enhancement MR images via registration and propagation of cine contours

Segmentation of myocardium in Late Gadolinium Enhanced (LGE) MR images is often difficult due to accumulation of contrast agent in the infarct areas, leading to poor delineation from adjacent blood pools. Thus, manual determination of the endo-and epicardial contours is challenging, time consuming, and subject to significant intra-and inter-observer variability. In this paper, we propose to use prior information from cine images of the same patient to achieve accurate segmentation in the corresponding LGE images. The proposed method first delineates the endo-and epicardial borders in the higher quality cine images of the patients heart. Then, a robust multiscale registration framework incorporating multiscale total variation (TV) flow as a preprocessing procedure is used to align the 3D cine and 2D LGE data for the same patient. The contours from the cine images are then propagated to the LGE dataset using the same transformation. Promising results were achieved through experimental validation.

Multiscale TV flow with applications to fast denoising and registration

Medical images consist of image structures of varying scales, with different scales representing different components. For example, in cardiac images, left ventricle, myocardium and blood pool are the large scale structures, whereas infarct and noise are represented by relatively small scale structures. Thus, extracting different scales in an image i.e. multiscale image representation, is a valuable tool in medical image processing. There are various multiscale representation techniques based on different image decomposition algorithms and denoising methods. Gaussian blurring with varying standard deviation can be considered as a multiscale representation, but it diffuses the image isotropically, thereby diffusing main edges. On the other hand, inverse scale representations based on variational formulations preserve edges; but they tend to be time consuming and thus unsuitable for real-time applications. In the present work, we propose a fast multiscale representation technique, motivated by successive decomposition of smooth parts based on total variation (TV ) minimization. Thus, we smooth a given image at an increasing scale, producing a multiscale TV representation. As noise is a small scale component of an image, we can effectively use the proposed method for denoising . We also prove that the denoising speed, up to the time-step, is determined by the user, making the algorithm well-suited for real-time applications. The proposed method inherits edge preserving property from total variation flow. Using this property, we propose a novel multiscale image registration algorithm, where we register corresponding scales in images, thereby registering images efficiently and accurately.

X-ray and magnetic resonance imaging fusion for cardiac resynchronization therapy.

Cardiac Resynchronization Therapy (CRT) can effectively treat left ventricle (LV) driven Heart Failure (HF). However, 30% of the CRT recipients do not experience symptomatic benefit. Recent studies show that the CRT response rate can reach 95% when the LV pacing lead is placed at an optimal site at a region of maximal LV dyssynchrony and away from myocardial scars. Cardiac Magnetic Resonance (CMR) can identify the optimal site in three dimensions (3D). 3D CMR data can be registered to clinical standard x-ray fluoroscopy to achieve an optimal pacing of the LV. We have developed a 3D CMR to 2D x-ray image registration method for CRT procedures. We have employed the LV pacing lead on x-ray images and coronary sinus on MR data as landmarks. The registration method makes use of a guidewire simulation algorithm, edge based image registration technique and x-ray C-arm tracking to register the coronary sinus and pacing lead landmarks.

In-vivo MRI and in-vivo electro-anatomical voltage map characteristics of infarct heterogeneity in a swine model

The arrhythmogenic substrate in patients with prior myocardial infarct (MI) is located at the border zone, BZ. In this study we correlated the BZ identified by two methods: electro-anatomical voltage mapping (EAVM) and a novel MRI method, multi-contrast late enhancement (MCLE). A pre-clinical porcine model with chronic MI was used to characterize BZ via MRI and EAVM. Results focus on the comparison between scar percentage and BZ percentage identified by each method. The correlation coefficient for BZ percentage between the two methods was 0.74 with a p-value of less the 0.0001. Bland-Altman plots were also used to compare between the two methods (slope of 0.83 ±0.045). For a case of subtle infarct, there was only 1.3% infarct identified on EAVM compared to 22.2% on the corresponding slice on MCLE. The percentage of infarct on MCLE in subtle infarct does not relate to percentage of infarct in EAVM. Future registration between T1 maps and EAVM will permit a quantitative comparison of MRI and EAVM measures.

Watershed segmentation of basal left ventricle for quantitation of cine cardiac MRI function.

To quantitatively analyze global and regional cardiac function from MR, clinical parameters such as ejection fraction (EF) and volumes are required. These depend upon accurate delineation of endo- and epicardial contours of the left ventricle (LV). Previous work [1] has demonstrated the difficulty of accurate LV segmentation, especially in basal slices where the LV outflow tract (LVOT) interrupts continuous myocardial contours. A novel method for the robust, accurate and fully automatic LV segmentation from short axis (SA) cine MR images is presented in this study that applies watershed technique to solve basal slice segmentation.