Liran Goshen

An iodine-calcium separation analysis and virtually non-contrasted image generation obtained with single source dual energy MDCT

Dual energy CT enables the differentiation between various materials by analyzing their unique attenuation spectral response. These responses are represented as vectors on an image-based energy map. Practically, this spectral information can be very noisy requiring further analysis. Furthermore, the material response vectors may be affected by beam-hardening and varied between images.

Learning Patient-Specific Lumped Models for Interactive Coronary Blood Flow Simulations

We propose a parametric lumped model (LM) for fast patient-specific computational fluid dynamic simulations of blood flow in elongated vessel networks to alleviate the computational burden of 3D finite element (FE) simulations. We learn the coefficients balancing the local nonlinear hydraulic effects from a training set of precomputed FE simulations. Our LM yields pressure predictions accurate up to 2.76mmHg on 35 coronary trees obtained from 32 coronary computed tomography angiograms. We also observe a very good predictive performance on a validation set of 59 physiological measurements suggesting that FE simulations can be replaced by our LM. As LM predictions can be computed extremely fast, our approach paves the way to use a personalised interactive biophysical model with realtime feedback in clinical practice.

A unique noncathartic CT colonography approach by using two-layer dual-energy MDCT and a special algorithmic colon cleansing method

Cathartic bowel preparation as part of CT colonography examination (CTC) can be uncomfortable or even dangerous for certain patient groups. Noncathartic CT colonography (i.e. without cathartic cleansing) including contrast-material fecal tagging can offer significant clinical advantages and can increase the screening compliance for colorectal cancer. Current techniques of conventional CTC with fecal tagging use “electronic cleansing” algorithms to remove the remaining tagged colonic contents from the images. In such cases these methods can give satisfactory results, but they face serious problems with the inferior tagging quality of noncathartic protocols. To overcome this fundamental problem, a completely new approach is proposed using both a two-layer dual-energy MDCT and a dedicated algorithmic cleansing method that utilizes the spectral information. Feasibility study was performed with a two-layer dual-energy MDCT which was utilized in clinical studies with oral intake of both iodine and barium contrast agents. The new method was compared to a conventional electronic cleansing technique. We show that the new approach is better at detecting highly dilute contrast agents in the colon, particularly where the tagged colonic content is mixed or adjacent to air regions. Therefore, the new technique may surmount the need for cathartic bowel preparation in CTC.

Arterial double-contrast dual-energy MDCT: in-vivo rabbit atherosclerosis with iodinated nanoparticles and gadolinium agents

An in-vivo feasibility study of potentially improved atherosclerosis CT imaging is presented. By administration of two different contrast agents to rabbits with induced atherosclerotic plaques we aim at identifying both soft plaque and vessel lumen simultaneously. Initial injection of iodinated nanoparticle (INP) contrast agent (N1177 - Nanoscan Imaging), two to four hours before scan, leads to its later accumulation in macrophage-rich soft plaque, while a second gadolinium contrast agent (Magnevist) injected immediately prior to the scan blends with the aortic blood. The distinction between the two agents in a single scan is achieved with a double-layer dual-energy MDCT (Philips Healthcare) following material separation analysis using the reconstructed images of the different x-ray spectra. A single contrast agent injection scan, where only INP was injected two hours prior to the scan, was compared to a double-contrast scan taken four hours after INP injection and immediately after gadolinium injection. On the single contrast agent scan we observed along the aorta walls, localized iodine accumulation which can point on INP uptake by atherosclerotic plaque. In the double-contrast scan the gadolinium contributes a clearer depiction of the vessel lumen in addition to the lasting INP presence. The material separation shows a good correlation to the pathologies inferred from the conventional CT images of the two different scans while performing only a single scan prevents miss-registration problems and reduces radiation dose. These results suggest that a double-contrast dual-energy CT may be used for advanced clinical diagnostic applications.

Automatic coronary lumen segmentation with partial volume modeling improves lesions' hemodynamic significance assessment.

The determination of hemodynamic significance of coronary artery lesions from cardiac computed tomography angiography (CCTA) based on blood flow simulations has the potential to improve CCTA’s specificity, thus resulting in improved clinical decision making. Accurate coronary lumen segmentation required for flow simulation is challenging due to several factors. Specifically, the partial-volume effect (PVE) in small-diameter lumina may result in overestimation of the lumen diameter that can lead to an erroneous hemodynamic significance assessment. In this work, we present a coronary artery segmentation algorithm tailored specifically for flow simulations by accounting for the PVE. Our algorithm detects lumen regions that may be subject to the PVE by analyzing the intensity values along the coronary centerline and integrates this information into a machine-learning based graph min-cut segmentation framework to obtain accurate coronary lumen segmentations. We demonstrate the improvement in hemodynamic significance assessment achieved by accounting for the PVE in the automatic segmentation of 91 coronary artery lesions from 85 patients. We compare hemodynamic significance assessments by means of fractional flow reserve (FFR) resulting from simulations on 3D models generated by our segmentation algorithm with and without accounting for the PVE. By accounting for the PVE we improved the area under the ROC curve for detecting hemodynamically significant CAD by 29% (N=91, 0.85 vs. 0.66, p<0.05, Delong’s test) with invasive FFR threshold of 0.8 as the reference standard. Our algorithm has the potential to facilitate non-invasive hemodynamic significance assessment of coronary lesions.

Learning a Sparse Database for Patch-Based Medical Image Segmentation

We introduce a functional for the learning of an optimal database for patch-based image segmentation with application to coronary lumen segmentation from coronary computed tomography angiography (CCTA) data. The proposed functional consists of fidelity, sparseness and robustness to small-variations terms and their associated weights. Existing work address database optimization by prototype selection aiming to optimize the database by either adding or removing prototypes according to a set of predefined rules. In contrast, we formulate the database optimization task as an energy minimization problem that can be solved using standard numerical tools. We apply the proposed database optimization functional to the task of optimizing a database for patch-base coronary lumen segmentation. Our experiments using the publicly available MICCAI 2012 coronary lumen segmentation challenge data show that optimizing the database using the proposed approach reduced database size by 96% while maintaining the same level of lumen segmentation accuracy. Moreover, we show that the optimized database yields an improved specificity of CCTA based fractional flow reserve (0.73 vs 0.7 for all lesions and 0.68 vs 0.65 for obstructive lesions) using a training set of 132 (76 obstructive) coronary lesions with invasively measured FFR as the reference.

Simultaneous Multi-phase Coronary CT Angiography Analysis for Coronary Artery Disease Evaluation

Multi-Detector Computed Tomography (MDCT) is becoming increasingly important in the diagnosis of Coronary Artery Disease (CAD). Cardiac MDCT scan generally allows for reconstruction of several frames/phases in the cardiac cycle. The reconstructed images are then used to create curved multi-planar reformation (MPR) views wherein coronary lesions are best diagnosed. However, the generation of such MPR views for all potentially reconstructed phases is tedious and time consuming. Therefore, only a single phase is commonly used for diagnosis which may reduce the overall diagnostic accuracy. In the current work, we propose a new method that enable diagnosis of lesions from all reconstructed phases on a common MPR view simultaneously. Our method extracts the coronary centerline in one phase only. Next, it performs a fast registration of a region of interest between the multiple phases. Finally, the multiple phases are aligned to the MPR view and the clinician is able to review the multiple phases simultaneously. Our experiments indicate that the analysis time of multi-phase coronary CTA data can be reduced to less than 30 % of the currently required time using our method.

Virtual nonenhanced abdominal dual-energy MDCT: Analysis of image characteristics.

AIM To evaluate abdominal and pelvic image characteristics and artifacts on virtual nonenhanced (VNE) images generated from contrast-enhanced dual-energy multidetector computed tomography (MDCT) studies. METHODS Hadassah-Hebrew University Medical Institutional Review Board approval was obtained; 22 patients underwent clinically-indicated abdominal and pelvic single-source dual-energy MDCT (Philips Healthcare, Cleveland, OH, USA), pre- and post-IV administration of Omnipaque 300 contrast (100 cc). Various solid and vascular structures were evaluated. VNE images were generated from the portal contrast-enhanced phase using probabilistic separation. Contrast-enhanced-, regular nonenhanced (RNE)-, and VNE images were evaluated with a total of 1494 density measurements. The ratio of iodine contrast deletion was calculated. Visualization of calcifications, urinary tract stones, and image artifacts in VNE images were assessed. RESULTS VNE images were successfully generated in all patients. Significant portal-phase iodine contrast deletion was seen in the kidney (61.7%), adrenal gland (55.3%), iliac artery (55.0%), aorta (51.6%), and spleen (34.5%). Contrast deletion was also significant in the right atrium (RA) (51.5%) and portal vein (39.3%), but insignificant in the iliac vein and inferior vena cava (IVC). Average post contrast-to-VNE HU differences were significant (P < 0.05) in the: RA -135.3 (SD 121.8), aorta -114.1 (SD 48.5), iliac artery -104.6 (SD 53.7), kidney -30.3 (SD 34.9), spleen -9.2 (SD 8.8), and portal vein -7.7 (SD 13.2). Average VNE-to-RNE HU differences were significant in all organs but the prostate and subcutaneous fat: aorta 38.0 (SD 9.3), RA 37.8 (SD 16.1), portal vein 21.8 (SD 12.0), IVC 12.2 (SD 11.6), muscle 3.3 (SD 4.9), liver 5.7 (SD 6.4), spleen 22.3 (SD 9.8), kidney 40.5 (SD 6.8), and adrenal 20.7 (SD 13.5). On VNE images, 196/213 calcifications (92%) and 5/6 renal stones (84%) were visualized. Lytic-like artifacts in the vertebral bodies were seen in all studies. CONCLUSION Iodine deletion in VNE images is most significant in arteries, and less significant in solid organs and veins. Most vascular and intra-abdominal organ calcifications are preserved.