Yeonggul Jang

Automatic Coronary Artery Segmentation Using Active Search for Branches and Seemingly Disconnected Vessel Segments from Coronary CT Angiography

We propose a Bayesian tracking and segmentation method of coronary arteries on coronary computed tomographic angiography (CCTA). The geometry of coronary arteries including lumen boundary is estimated in Maximum A Posteriori (MAP) framework. Three consecutive sphere based filtering is combined with a stochastic process that is based on the similarity of the consecutive local neighborhood voxels and the geometric constraint of a vessel. It is also founded on the prior knowledge that an artery can be seen locally disconnected and consist of branches which may be seemingly disconnected due to plaque build up. For such problem, an active search method is proposed to find branches and seemingly disconnected but actually connected vessel segments. Several new measures have been developed for branch detection, disconnection check and planar vesselness measure. Using public domain Rotterdam CT dataset, the accuracy of extracted centerline is demonstrated and automatic reconstruction of coronary artery mesh is shown.

Feasibility of Selective Catheter-Directed Coronary Computed Tomography Angiography Using Ultralow-Dose Intracoronary Contrast Injection in a Swine Model

ObjectiveSelective catheter-directed intracoronary contrast injected coronary computed tomography angiography (selective CCTA) has recently been introduced for on-site evaluation of coronary artery disease during coronary artery catheterization. In this study, we aimed to develop a feasible protocol for selective CCTA using ultralow-dose contrast medium as compared with conventional intravenous CCTA (IV CCTA). Materials and MethodsA novel combined system incorporating coronary angiography and a 320-detector row computed tomographic scanner was used to study 4 swine (35–40 kg) under animal institutional review board approval. A selective CCTA scan was simultaneously performed with an injection of 13.13 mgI/mL of modulated contrast medium at multiple different injection rates including 2, 3, and 4 mL/s and different total injection volumes of either 20 or 30 mL. Intravenous CCTA was performed with 60 mL of contrast medium, followed by 30 mL of saline chaser at 5 mL/s. Coronary mean and peak intensity, transluminal attenuation gradient, as well as 3-dimensional maximum intensity projections were obtained. ResultsAttenuation values (mean ± standard error, in Hounsfield units [HUs]) of selective CCTA for the left anterior descending (LAD) and right coronary artery (RCA) using the various combinations of injection rates and total injection volumes were as follows: 20 mL at 2 mL/s (LAD, 270.3 ± 20.4 HU; RCA, 322.6 ± 7.4 HU), 20 mL at 3 mL/s (LAD, 262.9 ± 20.4 HU; RCA, 264.7 ± 7.4 HU), 30 mL at 3 mL/s (LAD, 276.8 ± 20.4 HU; RCA, 274.0 ± 7.4 HU), 20 mL at 4 mL/s (LAD, 268.0 ± 20.4 HU; RCA, 277.7 ± 7.4 HU), and 30 mL at 4 mL/s (LAD, 251.3 ± 20.4 HU; RCA, 334.7 ± 7.4 HU). The representative protocol of the selective CCTA studies produced results within the optimal enhancement range (approximately 250-350 HU) for all segments, and comparison of transluminal attenuation gradient data with selective CCTA and IV CCTA studies demonstrated that the former method was more homogenous (−1.5245 and −1.7558 for LAD as well as 0.0459 and 0.0799 for RCA, respectively). Notably, the volume of iodine contrast medium used for selective CCTA was reported to be 1.09% (0.2 g) of IV CCTA (24 g). ConclusionsThe current findings demonstrate the feasibility of selective CCTA using ultralow-dose intracoronary contrast injection. This technique may provide additional means of coronary evaluation in patients who may require strategic planning before a procedure using a combined modality system.

Deep Learning Based Coronary Artery Motion Artifact Compensation Using Style-Transfer Synthesis in CT Images.

Motion artifact compensation of the coronary artery in computed tomography (CT) is required to quantify the risk of coronary artery disease more accurately. We present a novel method based on deep learning for motion artifact compensation in coronary CT angiography (CCTA). The ground-truth, i.e., coronary artery without motion, was synthesized using full-phase four-dimensional (4D) CT by applying style-transfer method because it is medically impossible to obtain in practice. The network for motion artifact compensation based on very deep convolutional neural network (CNN) is trained using the synthesized ground-truth. An observer study was performed for the evaluation of the proposed method. The motion artifacts were markedly reduced and boundaries of the coronary artery were much sharper than before applying the proposed method, with a strong inter-observer agreement (kappa = 0.78).

Automatic aortic valve landmark localization in coronary CT angiography using colonial walk

The minimally invasive transcatheter aortic valve implantation (TAVI) is the most prevalent method to treat aortic valve stenosis. For pre-operative surgical planning, contrast-enhanced coronary CT angiography (CCTA) is used as the imaging technique to acquire 3-D measurements of the valve. Accurate localization of the eight aortic valve landmarks in CT images plays a vital role in the TAVI workflow because a small error risks blocking the coronary circulation. In order to examine the valve and mark the landmarks, physicians prefer a view parallel to the hinge plane, instead of using the conventional axial, coronal or sagittal view. However, customizing the view is a difficult and time-consuming task because of unclear aorta pose and different artifacts of CCTA. Therefore, automatic localization of landmarks can serve as a useful guide to the physicians customizing the viewpoint. In this paper, we present an automatic method to localize the aortic valve landmarks using colonial walk, a regression tree-based machine-learning algorithm. For efficient learning from the training set, we propose a two-phase optimized search space learning model in which a representative point inside the valvular area is first learned from the whole CT volume. All eight landmarks are then learned from a smaller area around that point. Experiment with preprocedural CCTA images of TAVI undergoing patients showed that our method is robust under high stenotic variation and notably efficient, as it requires only 12 milliseconds to localize all eight landmarks, as tested on a 3.60 GHz single-core CPU.

Fully Automatic Segmentation of Coronary Arteries Based on Deep Neural Network in Intravascular Ultrasound Images

Accurate segmentation of coronary arteries is important for the diagnosis of cardiovascular diseases. In this paper, we propose a fully convolutional neural network to efficiently delineate the boundaries of the wall and lumen of the coronary arteries using intravascular ultrasound (IVUS) images. Our network addresses multi-label segmentation of the wall and lumen areas at the same time. The primary body of the proposed network is U-shaped which contains the encoding and decoding paths to learn rich hierarchical representations. The multi-scale input layer is adapted to take a multi-scale input. We deploy a multi-label loss function with weighted pixel-wise cross-entropy to alleviate imbalance of the rate of background, wall, and lumen. The proposed method is compared with three existing methods and the segmentation results are measured on four metrics, dice similarity coefficient, Jaccard index, percentage of area difference, and Hausdorff distance on totally 38,478 IVUS images from 35 subjects.

Coronary luminal and wall mask prediction using convolutional neural network

A significant amount of research has been done on the segmentation of coronary arteries. However, the resulting automated boundary delineation is still not suitable for clinical utilization. The convolutional neural network was driving advances in the medical image processing. We propose the brief convolutional network (BCN) that automatically produces the labeled mask with the luminal and wall boundaries of the coronary artery. We utilized 50 patients of CCTA - intravascular ultrasound matched image data sets. Training and testing were performed on 40 and 10 patient data sets, respectively. The prediction of luminal and wall mask was performed using stacked BCN on the each image view: axial, coronal, and sagittal of straightened curved planar reformation. We defined the vector that includes probability from BCN result on each image view and proposed amplified probability. We used an Adaptive Boost regressor with an extremely randomized tree regressor to determine the label for unknown probability vector.

Maximum a posteriori estimation method for aorta localization and coronary seed identification

A robust method is proposed for the automatic identification of seed points (coronary ostia) for the segmentation of coronary arteries from CT image.Our method provides both aorta and ostia localization.Anatomical and geometrical priors are statistically obtained and used in MAP estimation.Two components are jointly found in MAP estimation (e.g. ascending and descending aortas, left and right coronary ostia). We propose a robust method for the automatic identification of seed points for the segmentation of coronary arteries from coronary computed tomography angiography (CCTA). The detection of the aorta and the two ostia for use as seed points is required for the automatic segmentation of coronary arteries. Our method is based on a Bayesian framework combining anatomical and geometrical features. We demonstrate the robustness and accuracy of our method by comparison with two conventional methods on 130 CT cases.

Assessment of myocardial viability based on dual-energy computed tomography in patients with chronic myocardial infarction: comparison with magnetic resonance imaging

PURPOSE To evaluate the diagnostic performance of dual-energy computed tomography (DECT) for the assessment of myocardial viability compared with magnetic resonance imaging (MRI) in patients with chronic myocardial infarction (CMI). METHODS AND MATERIAL Twenty-six patients were prospectively enrolled, who underwent DECT and MRI at delayed phase. The infarct volumes for DECT and MRI were measured. RESULTS In per-segment and per-vessel analysis, DECT showed excellent diagnostic performance compared with MRI (diagnostic accuracy: 86.2%, 81.2% respectively). In volume analysis, DECT correlated well with MRI (r=0.966, p<0.0001). CONCLUSIONS DECT has excellent diagnostic performance for detecting CMI.

Automatic Segmentation of LV and RV in Cardiac MRI

Automatic and accurate segmentation of Left Ventricle (LV) and Right Ventricle (RV) in cine-MRI is required to analyze cardiac function and viability. We present a fully convolutional neural network to efficiently segment LV and RV as well as myocardium. The network is trained end-to-end from scratch. Average dice scores from five-fold cross-validation on the ACDC training dataset were 0.94, 0.89, and 0.88 for LV, RV, and myocardium. Experimental results show the robustness of the proposed architecture.

Generation of Triangular Mesh of Coronary Artery Using Mesh Merging

Generating a 3D surface model from coronary artery segmentation helps to not only improve the rendering efficiency but also the diagnostic accuracy by providing physiological informations such as fractional flow reserve using computational fluid dynamics (CFD). This paper proposes a method to generate a triangular surface mesh using vessel structure information acquired with coronary artery segmentation. The marching cube algorithm is a typical method for generating a triangular surface mesh from a segmentation result as bit mask. But it is difficult for methods based on marching cube algorithm to express the lumen of thin, small and winding vessels because the algorithm only works in a three-dimensional (3D) discrete space. The proposed method generates a more accurate triangular surface mesh for each singular vessel using vessel centerlines, normal vectors and lumen diameters estimated during the process of coronary artery segmentation as the input. Then, the meshes that are overlapped due to branching are processed by mesh merging and merged into a coronary mesh.