Michael Scheuering

Four-Chamber Heart Modeling and Automatic Segmentation for 3-D Cardiac CT Volumes Using Marginal Space Learning and Steerable Features

We propose an automatic four-chamber heart segmentation system for the quantitative functional analysis of the heart from cardiac computed tomography (CT) volumes. Two topics are discussed: heart modeling and automatic model fitting to an unseen volume. Heart modeling is a nontrivial task since the heart is a complex nonrigid organ. The model must be anatomically accurate, allow manual editing, and provide sufficient information to guide automatic detection and segmentation. Unlike previous work, we explicitly represent important landmarks (such as the valves and the ventricular septum cusps) among the control points of the model. The control points can be detected reliably to guide the automatic model fitting process. Using this model, we develop an efficient and robust approach for automatic heart chamber segmentation in 3D CT volumes. We formulate the segmentation as a two-step learning problem: anatomical structure localization and boundary delineation. In both steps, we exploit the recent advances in learning discriminative models. A novel algorithm, marginal space learning (MSL), is introduced to solve the 9-D similarity transformation search problem for localizing the heart chambers. After determining the pose of the heart chambers, we estimate the 3D shape through learning-based boundary delineation. The proposed method has been extensively tested on the largest dataset (with 323 volumes from 137 patients) ever reported in the literature. To the best of our knowledge, our system is the fastest with a speed of 4.0 s per volume (on a dual-core 3.2-GHz processor) for the automatic segmentation of all four chambers.

Fast Automatic Heart Chamber Segmentation from 3D CT Data Using Marginal Space Learning and Steerable Features

Multi-chamber heart segmentation is a prerequisite for global quantification of the cardiac function. The complexity of cardiac anatomy, poor contrast, noise or motion artifacts makes this segmentation problem a challenging task. In this paper, we present an efficient, robust, and fully automatic segmentation method for 3D cardiac computed tomography (CT) volumes. Our approach is based on recent advances in learning discriminative object models and we exploit a large database of annotated CT volumes. We formulate the segmentation as a two step learning problem: anatomical structure localization and boundary delineation. A novel algorithm, marginal space learning (MSL), is introduced to solve the 9-dimensional similarity search problem for localizing the heart chambers. MSL reduces the number of testing hypotheses by about six orders of magnitude. We also propose to use steerable image features, which incorporate the orientation and scale information into the distribution of sampling points, thus avoiding the time-consuming volume data rotation operations. After determining the similarity transformation of the heart chambers, we estimate the 3D shape through learning-based boundary delineation. Extensive experiments on multi-chamber heart segmentation demonstrate the efficiency and robustness of the proposed approach, comparing favorably to the state-of-the-art. This is the first study reporting stable results on a large cardiac CT dataset with 323 volumes. In addition, we achieve a speed of less than eight seconds for automatic segmentation of all four chambers.

Dual-source computed tomography: advances of improved temporal resolution in coronary plaque imaging.

Objectives:The aim of this study was to quantify image quality gains of a moving coronary plaque phantom using dual-source computed tomography (DSCT) providing 83 milliseconds temporal resolution in direct comparison to 64 slice single-source multidetector CT (MDCT) with a temporal resolution of 165 milliseconds. Materials and Methods:Three cardiac vessel phantoms with fixed 50% stenosis and changing plaque configurations were mounted on a moving device simulating cardiac motion. Scans were performed at a simulated heart frequency of 60 to 120 bpm. Image quality assessment was performed in different anatomic orientations inside a thoracic phantom. Results:A significant improvement of image quality using the DSCT could be found (P = 0.0002). Relevant factors influencing image quality aside from frequency (P = 0.0002) are plaque composition (P < 0.0001), as well as orientation (P < 0.0001). Conclusion:Scanning with 83 milliseconds temporal resolution improved image quality of coronary plaque at higher heart frequencies.

Fast volumetric deformation on general purpose hardware

High performance deformation of volumetric objects is a common problem in computer graphics that has not yet been handled sufficiently. As a supplement to 3D texture based volume rendering, a novel approach is presented, which adaptively subdivides the volume into piecewise linear patches. An appropriate mathematical model based on tri-linear interpolation and its approximations is proposed. New optimizations are introduced in this paper which are especially tailored to an efficient implementation using general purpose rasterization hardware, including new technologies, such as vertex programs and pixel shaders. Additionally, a high performance model for local illumination calculation is introduced, which meets the aesthetic requirements of visual arts and entertainment. The results demonstrate the significant performance benefit and allow for time-critical applications, such as computer assisted surgery.

Constrained marginal space learning for efficient 3D anatomical structure detection in medical images

Recently, we proposed marginal space learning (MSL) as a generic approach for automatic detection of 3D anatomical structures in many medical imaging modalities. To accurately localize a 3D object, we need to estimate nine parameters (three for position, three for orientation, and three for anisotropic scaling). Instead of uniformly searching the original nine-dimensional parameter space, only low-dimensional marginal spaces are uniformly searched in MSL, which significantly improves the speed. In many real applications, a strong correlation may exist among parameters in the same marginal spaces. For example, a large object may have large scaling values along all directions. In this paper, we propose constrained MSL to exploit this correlation for further speed-up. As another major contribution, we propose to use quaternions for 3D orientation representation and distance measurement to overcome the inherent drawbacks of Euler angles in the original MSL. The proposed method has been tested on three 3D anatomical structure detection problems in medical images, including liver detection in computed tomography (CT) volumes, and left ventricle detection in both CT and ultrasound volumes. Experiments on the largest datasets ever reported show that constrained MSL can improve the detection speed up to 14 times, while achieving comparable or better detection accuracy. It takes less than half a second to detect a 3D anatomical structure in a volume.

Dynamic Model-Driven Quantitative and Visual Evaluation of the Aortic Valve from 4D CT

Aortic valve disease is an important cardio-vascular disorder, which affects 2.5% of the global population and often requires elaborate clinical management. Experts agree that visual and quantitative evaluation of the valve, crucial throughout the clinical workflow, is currently limited to 2D imaging which can potentially yield inaccurate measurements. In this paper, we propose a novel approach for morphological and functional quantification of the aortic valve based on a 4D model estimated from computed tomography data. A physiological model of the aortic valve, capable to express large shape variations, is generated using parametric splines together with anatomically-driven topological and geometrical constraints. Recent advances in discriminative learning and incremental searching methods allow rapid estimation of the model parameters from 4D Cardiac CT specifically for each patient. The proposed approach enables precise valve evaluation with model-based dynamic measurements and advanced visualization. Extensive experiments and initial clinical validation demonstrate the efficiency and accuracy of the proposed approach. To the best of our knowledge this is the first time such a patient specific 4D aortic valve model is proposed.

Automatic segmentation of the left ventricle and computation of diagnostic parameters using regiongrowing and a statistical model

The manual segmentation and analysis of high-resolution multi-slice cardiac CT datasets is both labor intensive and time consuming. Therefore it is necessary to supply the cardiologist with powerful software tools to segment the myocardium and compute the relevant diagnostic parameters. In this work we present a semi-automatic cardiac segmentation approach with minimal user interaction. It is based on a combination of an adaptive slice-based regiongrowing and a modified Active Shape Model (ASM). Starting with a single manual click point in the ascending aorta, the aorta, the left atrium and the left ventricle get segmented with the slice-based adaptive regiongrowing. The approximate position of the aortic and mitral valve as well as the principal axes of the left ventricle (LV) are determined. To prevent the regiongrowing from draining into neighboring anatomical structures via CT artifacts, we implemented a draining control by examining a cubic region around the currently processed voxel. Additionally, we use moment-based parameters to integrate simple anatomical knowledge into the regiongrowing process. Using the results of the preceding regiongrowing process, a ventricle-centric and normalized coordinate system is established which is used to adapt a previously trained ASM to the image, using an iterative multi-resolution approach. After fitting the ASM to the image, we can use the generated model-points to create an exact surface model of the left ventricular myocardium for visualization and for computing the diagnostically relevant parameters, like the ventricular blood volume and the myocardial wall thickness.

Comparison of Fractional Flow Reserve Based on Computational Fluid Dynamics Modeling Using Coronary Angiographic Vessel Morphology Versus Invasively Measured Fractional Flow Reserve

Invasive fractional flow reserve (FFRinvasive), although gold standard to identify hemodynamically relevant coronary stenoses, is time consuming and potentially associated with complications. We developed and evaluated a new approach to determine lesion-specific FFR on the basis of coronary anatomy as visualized by invasive coronary angiography (FFRangio): 100 coronary lesions (50% to 90% diameter stenosis) in 73 patients (48 men, 25 women; mean age 67 ± 9 years) were studied. On the basis of coronary angiograms acquired at rest from 2 views at angulations at least 30° apart, a PC-based computational fluid dynamics modeling software used personalized boundary conditions determined from 3-dimensional reconstructed angiography, heart rate, and blood pressure to derive FFRangio. The results were compared with FFRinvasive. Interobserver variability was determined in a subset of 25 narrowings. Twenty-nine of 100 coronary lesions were hemodynamically significant (FFRinvasive ≤ 0.80). FFRangio identified these with an accuracy of 90%, sensitivity of 79%, specificity of 94%, positive predictive value of 85%, and negative predictive value of 92%. The area under the receiver operating characteristic curve was 0.93. Correlation between FFRinvasive (mean: 0.84 ± 0.11) and FFRangio (mean: 0.85 ± 0.12) was r = 0.85. Interobserver variability of FFRangio was low, with a correlation of r = 0.88. In conclusion, estimation of coronary FFR with PC-based computational fluid dynamics modeling on the basis of lesion morphology as determined by invasive angiography is possible with high diagnostic accuracy compared to invasive measurements.

Multi-Scale Feature Extraction for Learning-Based Classification of Coronary Artery Stenosis

Assessment of computed tomography coronary angiograms for diagnostic purposes is a mostly manual, timeconsuming task demanding a high degree of clinical experience. In order to support diagnosis, a method for reliable automatic detection of stenotic lesions in computed tomography angiograms is presented. Thereby, lesions are detected by boosting-based classification. Hence, a strong classifier is trained using the AdaBoost algorithm on annotated data. Subsequently, the resulting strong classification function is used in order to detect different types of coronary lesions in previously unseen data. As pattern recognition algorithms require a description of the objects to be classified, a novel approach for feature extraction in computed tomography angiograms is introduced. By generation of cylinder segments that approximate the vessel shape at multiple scales, feature values can be extracted that adequately describe the properties of stenotic lesions. As a result of the multi-scale approach, the algorithm is capable of dealing with the variability of stenotic lesion configuration. Evaluation of the algorithm was performed on a large database containing unseen segmented centerlines from cardiac computed tomography images. Results showed that the method was able to detect stenotic cardiovascular diseases with high sensitivity and specificity. Moreover, lesion based evaluation revealed that the majority of stenosis can be reliable identified in terms of position, type and extent.

Intraoperative augmented reality for minimally invasive liver interventions

Minimally invasive liver interventions demand a lot of experience due to the limited access to the field of operation. In particular, the correct placement of the trocar and the navigation within the patients body are hampered. In this work, we present an intraoperative augmented reality system (IARS) that directly projects preoperatively planned information and structures extracted from CT data, onto the real laparoscopic video images. Our system consists of a preoperative planning tool for liver surgery and an intraoperative real time visualization component. The planning software takes into account the individual anatomy of the intrahepatic vessels and determines the vascular territories. Methods for fast segmentation of the liver parenchyma, of the intrahepatic vessels and of liver lesions are provided. In addition, very efficient algorithms for skeletonization and vascular analysis allowing the approximation of patient-individual liver vascular territories are included. The intraoperative visualization is based on a standard graphics adapter for hardware accelerated high performance direct volume rendering. The preoperative CT data is rigidly registered to the patient position by the use of fiducials that are attached to the patients body, and anatomical landmarks in combination with an electro-magnetic navigation system. Our system was evaluated in vivo during a minimally invasive intervention simulation in a swine under anesthesia.