Deep Reinforcement Learning with Explicit Spatio-Sequential Encoding Network for Coronary Ostia Identification in CT Images.

Accurate identification of the coronary ostia from 3D coronary computed tomography angiography (CCTA) is a essential prerequisite step for automatically tracking and segmenting three main coronary arteries. In this paper, we propose a novel deep reinforcement learning (DRL) framework to localize the two coronary ostia from 3D CCTA. An optimal action policy is determined using a fully explicit spatial-sequential encoding policy network applying 2.5D Markovian states with three past histories. The proposed network is trained using a dueling DRL framework on the CAT08 dataset. The experiment results show that our method is more efficient and accurate than the other methods. blueFloating-point operations (FLOPs) are calculated to measure computational efficiency. The result shows that there are 2.5M FLOPs on the proposed method, which is about 10 times smaller value than 3D box-based methods. In terms of accuracy, the proposed method shows that 2.22 ± 1.12 mm and 1.94 ± 0.83 errors on the left and right coronary ostia, respectively. The proposed method can be applied to the tasks to identify other target objects by changing the target locations in the ground truth data. Further, the proposed method can be utilized as a pre-processing method for coronary artery tracking methods.

Clinical feasibility of catheter-directed selective intracoronary computed tomography angiography using an extremely low dose of iodine in patients with coronary artery disease.

OBJECTIVE: This study aimed to evaluate the clinical feasibility of catheter-directed selective computed tomography angiography (S-CTA) in patients with coronary artery disease (CAD). METHODS: We prospectively enrolled 65 patients diagnosed with CAD who underwent conventional computed tomography angiography (C-CTA). C-CTA was performed with 60-90 mL of contrast medium (370 mg iodine/mL), whereas S-CTA was performed with 15 mL of contrast medium and 17.19 mg iodine/mL. Luminal enhancement range, homogeneity of luminal enhancement, image quality, plaque volume (PV), and percent aggregate plaque volume (%APV) were measured. Paired Student's t test, Wilcoxon rank-sum test, and Pearson's correlation coefficient were used to compare two methods. RESULTS: Luminal enhancement was significantly higher on S-CTA than on C-CTA (324.4 ± 8.0 Hounsfield unit (HU) vs. 312.0 ± 8.0 HU, p < 0.0001 in the per-vessel analysis). Transluminal attenuation gradient showed a significantly slower reduction pattern on S-CTA than on C-CTA (-0.65 HU/10 mm vs. -0.89 HU/10 mm, p < 0.0001 in the per-vessel analysis). Image noise was significantly lower on S-CTA than on C-CTA (39.6 ± 10.0 HU vs. 43.9 ± 9.4 HU, p < 0.0001). There was excellent correlation between S-CTA and C-CTA with respect to PV and %APV (r = 0.99, r = 0.98, respectively). CONCLUSIONS: S-CTA might be useful in facilitating atherosclerotic plaque analysis and providing guidance for complex lesions such as chronic total occlusion, particularly in cases in which on-site procedure planning is required. KEY POINTS: • Selective computed tomography angiography (S-CTA) can serve as an intraprocedural computed tomography angiography protocol. • S-CTA was performed with low dose of iodine compared with conventional computed tomography angiography. • S-CTA enables on-site atherosclerotic plaque analysis.

Quantitative Assessment of Foot Blood Flow by Using Dynamic Volume Perfusion CT Technique: A Feasibility Study.

PURPOSE: To demonstrate the feasibility of foot blood flow measurement by using dynamic volume perfusion computed tomographic (CT) technique with the upslope method in an animal experiment and a human study. MATERIALS AND METHODS: The human study was approved by the institutional review board, and written informed consent was obtained from all patients. The animal study was approved by the research animal care and use committee. A perfusion CT experiment was first performed by using rabbits. A color-coded perfusion map was reconstructed by using in-house perfusion analysis software based on the upslope method, and the measured blood flow on the map was compared with the reference standard microsphere method by using correlation analysis. A total of 17 perfusion CT sessions were then performed (a) once in five human patients and (b) twice (before and after endovascular revascularization) in six human patients. Perfusion maps of blood flow were reconstructed and analyzed. The Wilcoxon signed rank test was used to prove significant differences in blood flow before and after treatment. RESULTS: The animal experiment demonstrated a strong correlation (R(2) = 0.965) in blood flow between perfusion CT and the microsphere method. Perfusion maps were obtained successfully in 16 human clinical sessions (94%) with the use of 32 mL of contrast medium and an effective radiation dose of 0.31 mSv (k factor for the ankle, 0.0002). The plantar dermis showed the highest blood flow among all anatomic structures of the foot, including muscle, subcutaneous tissue, tendon, and bone. After a successful revascularization procedure, the blood flow of the plantar dermis increased by 153% (P = .031). The interpretations of the color-coded perfusion map correlated well with the clinical and angiographic findings. CONCLUSION: Perfusion CT could be used to measure foot blood flow in both animals and humans. It can be a useful modality for the diagnosis of peripheral arterial disease by providing quantitative information on foot perfusion status.

Simultaneous Viability Assessment and Invasive Coronary Angiography Using a Therapeutic CT System in Chronic Myocardial Infarction Patients.

PURPOSE: In a preclinical study using a swine myocardial infarction (MI) model, a delayed enhancement (DE)-multi-detector computed tomography (MDCT) scan was performed using a hybrid system alongside diagnostic invasive coronary angiography (ICA) without the additional use of a contrast agent, and demonstrated an excellent correlation in the infarct area compared with histopathologic specimens. In the present investigation, we evaluated the feasibility and diagnostic accuracy of a myocardial viability assessment by DE-MDCT using a hybrid system comprising ICA and MDCT alongside diagnostic ICA without the additional use of a contrast agent. MATERIALS AND METHODS: We prospectively enrolled 13 patients (median age: 67 years) with a previous MI (>6 months) scheduled to undergo ICA. All patients underwent cardiac magnetic resonance (CMR) imaging before diagnostic ICA. MDCT viability scans were performed concurrently with diagnostic ICA without the use of additional contrast. The total myocardial scar volume per patient and average transmurality per myocardial segment measured by DE-MDCT were compared with those from DE-CMR. RESULTS: The DE volume measured by MDCT showed an excellent correlation with the volume measured by CMR (r=0.986, p<0.0001). The transmurality per segment by MDCT was well-correlated with CMR (r=0.900, p<0.0001); the diagnostic performance of MDCT in differentiating non-viable from viable myocardium using a 50% transmurality criterion was good with a sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of 87.5%, 99.5%, 87.5%, 99.5%, and 99.1%, respectively. CONCLUSION: The feasibility of the DE-MDCT viability assessment acquired simultaneously with conventional ICA was proven in patients with chronic MI using DE-CMR as the reference standard.

An Artificial Intelligence-Based Automated Echocardiographic Analysis: Enhancing Efficiency and Prognostic Evaluation in Patients With Revascularized STEMI.

BACKGROUND AND OBJECTIVES: Although various cardiac parameters on echocardiography have clinical importance, their measurement by conventional manual methods is time-consuming and subject to variability. We evaluated the feasibility, accuracy, and predictive value of an artificial intelligence (AI)-based automated system for echocardiographic analysis in patients with ST-segment elevation myocardial infarction (STEMI). METHODS: The AI-based system was developed using a nationwide echocardiographic dataset from five tertiary hospitals, and automatically identified views, then segmented and tracked the left ventricle (LV) and left atrium (LA) to produce volume and strain values. Both conventional manual measurements and AI-based fully automated measurements of the LV ejection fraction and global longitudinal strain, and LA volume index and reservoir strain were performed in 632 patients with STEMI. RESULTS: The AI-based system accurately identified necessary views (overall accuracy, 98.5%) and successfully measured LV and LA volumes and strains in all cases in which conventional methods were applicable. Inter-method analysis showed strong correlations between measurement methods, with Pearson coefficients ranging 0.81-0.92 and intraclass correlation coefficients ranging 0.74-0.90. For the prediction of clinical outcomes (composite of all-cause death, re-hospitalization due to heart failure, ventricular arrhythmia, and recurrent myocardial infarction), AI-derived measurements showed predictive value independent of clinical risk factors, comparable to those from conventional manual measurements. CONCLUSIONS: Our fully automated AI-based approach for LV and LA analysis on echocardiography is feasible and provides accurate measurements, comparable to conventional methods, in patients with STEMI, offering a promising solution for comprehensive echocardiographic analysis, reduced workloads, and improved patient care.

Artificial intelligence-enhanced automation for M-mode echocardiographic analysis: ensuring fully automated, reliable, and reproducible measurements.

To enhance M-mode echocardiography's utility for measuring cardiac structures, we developed and evaluated an artificial intelligence (AI)-based automated analysis system for M-mode images through the aorta and left atrium [M-mode (Ao-LA)], and through the left ventricle [M-mode (LV)]. Our system, integrating two deep neural networks (DNN) for view classification and image segmentation, alongside an auto-measurement algorithm, was developed using 5,958 M-mode images [3,258 M-mode (LA-Ao), and 2,700 M-mode (LV)] drawn from a nationwide echocardiographic dataset collated from five tertiary hospitals. The performance of view classification and segmentation DNNs were evaluated on 594 M-mode images, while automatic measurement accuracy was tested on separate internal test set with 100 M-mode images as well as external test set with 280 images (140 sinus rhythm and 140 atrial fibrillation). Performance evaluation showed the view classification DNN's overall accuracy of 99.8% and segmentation DNN's Dice similarity coefficient of 94.3%. Within the internal test set, all automated measurements, including LA, Ao, and LV wall and cavity, resonated strongly with expert evaluations, exhibiting Pearson's correlation coefficients (PCCs) of 0.81-0.99. This performance persisted in the external test set for both sinus rhythm (PCC, 0.84-0.98) and atrial fibrillation (PCC, 0.70-0.97). Notably, automatic measurements, consistently offering multi-cardiac cycle readings, showcased a stronger correlation with the averaged multi-cycle manual measurements than with those of a single representative cycle. Our AI-based system for automatic M-mode echocardiographic analysis demonstrated excellent accuracy, reproducibility, and speed. This automated approach has the potential to improve efficiency and reduce variability in clinical practice.

Artificial intelligence-enhanced automation of left ventricular diastolic assessment: a pilot study for feasibility, diagnostic validation, and outcome prediction.

Background: Evaluating left ventricular diastolic function (LVDF) is crucial in echocardiography; however, the complexity and time demands of current guidelines challenge clinical use. This study aimed to develop an artificial intelligence (AI)-based framework for automatic LVDF assessment to reduce subjectivity and improve accuracy and outcome prediction. Methods: We developed an AI-based LVDF assessment framework using a nationwide echocardiographic dataset from five tertiary hospitals. This framework automatically identifies views, calculates diastolic parameters, including mitral inflow and annular velocities (E/A ratio, e' velocity, and E/e' ratio), maximal tricuspid regurgitation velocity, left atrial (LA) volume index, and left atrial reservoir strain (LARS). Subsequently, it grades LVDF according to guidelines. The AI-framework was validated on an external dataset composed of randomly screened 173 outpatients who underwent transthoracic echocardiography with suspicion for diastolic dysfunction and 33 individuals from medical check-ups with normal echocardiograms at Seoul National University Bundang Hospital, tertiary medical center in Korea, between May 2012 and June 2022. Additionally, we assessed the predictive value of AI-derived diastolic parameters and LVDF grades for a clinical endpoint, defined as a composite of all-cause death and hospitalization for heart failure, using Cox-regression risk modelling. Results: In an evaluation with 200 echocardiographic examinations (167 suspected diastolic dysfunction patients, 33 controls), it achieves an overall accuracy of 99.1% in identifying necessary views. Strong correlations (Pearson coefficient 0.901-0.959) were observed between AI-derived and manually-derived measurements of diastolic parameters, including LARS as well as conventional parameters. When following the guidelines, whether utilizing AI-derived or manually-derived parameters, the evaluation of LVDF consistently showed high concordance rates (94%). However, both methods exhibited lower concordance rates with the clinician's prior assessments (77.5% and 78.5%, respectively). Importantly, both AI-derived and manually-derived LVDF grades independently demonstrated significant prognostic value [adjusted hazard ratio (HR) =3.03; P=0.03 and adjusted HR =2.75; P=0.04, respectively] for predicting clinical outcome. In contrast, the clinician's prior grading lost its significance as a prognostic indicator after adjusting for clinical risk factors (adjusted HR =1.63; P=0.36). AI-derived LARS values significantly decreased with worsening LVDF (P for trend <0.001), and low LARS (<17%) was associated with increased risk for the clinical outcome (Log-rank P=0.04) relative to that for preserved LARS (≥17%). Conclusions: Our AI-based approach for automatic LVDF assessment on echocardiography is feasible, potentially enhancing clinical diagnosis and outcome prediction.

Deep Learning on Multiphysical Features and Hemodynamic Modeling for Abdominal Aortic Aneurysm Growth Prediction.

Prediction of abdominal aortic aneurysm (AAA) growth is of essential importance for the early treatment and surgical intervention of AAA. Capturing key features of vascular growth, such as blood flow and intraluminal thrombus (ILT) accumulation play a crucial role in uncovering the intricated mechanism of vascular adaptation, which can ultimately enhance AAA growth prediction capabilities. However, local correlations between hemodynamic metrics, biological and morphological characteristics, and AAA growth rates present high inter-patient variability that results in that the temporal-spatial biochemical and mechanical processes are still not fully understood. Hence, this study aims to integrate the physics-based knowledge with deep learning with a patch-based convolutional neural network (CNN) approach by incorporating important multiphysical features relating to its pathogenesis for validating its impact on AAA growth prediction. For this task, we observe that the unstructured multiphysical features cannot be directly employed in the kernel-based CNN. To tackle this issue, we propose a parameterization of features to leverage the spatio-temporal relations between multiphysical features. The proposed architecture was tested on different combinations of four features including radius, intraluminal thrombus thickness, time-average wall shear stress, and growth rate from 54 patients with 5-fold cross-validation with two metrics, a root mean squared error (RMSE) and relative error (RE). We conduct extensive experiments on AAA patients, the results show the effect of leveraging multiphysical features and demonstrate the superiority of the presented architecture to previous state-of-the-art methods in AAA growth prediction.

Generative adversarial network with radiomic feature reproducibility analysis for computed tomography denoising.

BACKGROUND: Most computed tomography (CT) denoising algorithms have been evaluated using image quality analysis (IQA) methods developed for natural image, which do not adequately capture the texture details in medical imaging. Radiomics is an emerging image analysis technique that extracts texture information to provide a more objective basis for medical imaging diagnostics, overcoming the subjective nature of traditional methods. By utilizing the difficulty of reproducing radiomics features under different imaging protocols, we can more accurately evaluate the performance of CT denoising algorithms. METHOD: We introduced radiomic feature reproducibility analysis as an evaluation metric for a denoising algorithm. Also, we proposed a low-dose CT denoising method based on a generative adversarial network (GAN), which outperformed well-known CT denoising methods. RESULTS: Although the proposed model produced excellent results visually, the traditional image assessment metrics such as peak signal-to-noise ratio and structural similarity failed to show distinctive performance differences between the proposed method and the conventional ones. However, radiomic feature reproducibility analysis provided a distinctive assessment of the CT denoising performance. Furthermore, radiomic feature reproducibility analysis allowed fine-tuning of the hyper-parameters of the GAN. CONCLUSION: We demonstrated that the well-tuned GAN architecture outperforms the well-known CT denoising methods. Our study is the first to introduce radiomics reproducibility analysis as an evaluation metric for CT denoising. We look forward that the study may bridge the gap between traditional objective and subjective evaluations in the clinical medical imaging field.

Reconnection of fragmented parts of coronary arteries using local geometric features in X-ray angiography images.

The segmentation of coronary arteries in X-ray images is essential for image-based guiding procedures and the diagnosis of cardiovascular disease. However, owing to the complex and thin structures of the coronary arteries, it is challenging to accurately segment arteries in X-ray images using only a single neural network model. Consequently, coronary artery images obtained by segmentation with a single model are often fragmented, with parts of the arteries missing. Sophisticated post-processing is then required to identify and reconnect the fragmented regions. In this paper, we propose a method to reconstruct the missing regions of coronary arteries using X-ray angiography images. METHOD: We apply an independent convolutional neural network model considering local details, as well as a local geometric prior, for reconnecting the disconnected fragments. We implemented and compared the proposed method with several convolutional neural networks with customized encoding backbones as baseline models. RESULTS: When integrated with our method, existing models improved considerably in terms of similarity with ground truth, with a mean increase of 0.330 of the Dice similarity coefficient in local regions of disconnected arteries. The method is efficient and is able to recover missing fragments in a short number of iterations. CONCLUSION AND SIGNIFICANCE: Owing to the restoration of missing fragments of coronary arteries, the proposed method enables a significant enhancement of clinical impact. The method is general and can simply be integrated into other existing methods for coronary artery segmentation.

Assessment of Image Quality for Selective Intracoronary Contrast-Injected CT Angiography in a Hybrid Angio-CT System: A Feasibility Study in Swine.

PURPOSE: To compare image quality in selective intracoronary contrast-injected computed tomography angiography (Selective-CTA) with that in conventional intravenous contrast-injected CTA (IV-CTA). MATERIALS AND METHODS: Six pigs (35 to 40 kg) underwent both IV-CTA using an intravenous injection (60 mL) and Selective-CTA using an intracoronary injection (20 mL) through a guide-wire during/after percutaneous coronary intervention. Images of the common coronary artery were acquired. Scans were performed using a combined machine comprising an invasive coronary angiography suite and a 320-channel multi-slice CT scanner. Quantitative image quality parameters of CT attenuation, image noise, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), mean lumen diameter (MLD), and mean lumen area (MLA) were measured and compared. Qualitative analysis was performed using intraclass correlation coefficient (ICC), which was calculated for analysis of interobserver agreement. RESULTS: Quantitative image quality, determined by assessing the uniformity of CT attenuation (399.06 vs. 330.21, p<0.001), image noise (24.93 vs. 18.43, p<0.001), SNR (16.43 vs. 18.52, p=0.005), and CNR (11.56 vs. 13.46, p=0.002), differed significantly between IV-CTA and Selective-CTA. MLD and MLA showed no significant difference overall (2.38 vs. 2.44, p=0.068, 4.72 vs. 4.95, p=0.078). The density of contrast agent was significantly lower for selective-CTA (13.13 mg/mL) than for IV-CTA (400 mg/mL). Agreement between observers was acceptable (ICC=0.79±0.08). CONCLUSION: Our feasibility study in swine showed that compared to IV-CTA, Selective-CTA provides better image quality and requires less iodine contrast medium.

Left Ventricle Quantification Challenge: A Comprehensive Comparison and Evaluation of Segmentation and Regression for Mid-Ventricular Short-Axis Cardiac MR Data.

Automatic quantification of the left ventricle (LV) from cardiac magnetic resonance (CMR) images plays an important role in making the diagnosis procedure efficient, reliable, and alleviating the laborious reading work for physicians. Considerable efforts have been devoted to LV quantification using different strategies that include segmentation-based (SG) methods and the recent direct regression (DR) methods. Although both SG and DR methods have obtained great success for the task, a systematic platform to benchmark them remains absent because of differences in label information during model learning. In this paper, we conducted an unbiased evaluation and comparison of cardiac LV quantification methods that were submitted to the Left Ventricle Quantification (LVQuan) challenge, which was held in conjunction with the Statistical Atlases and Computational Modeling of the Heart (STACOM) workshop at the MICCAI 2018. The challenge was targeted at the quantification of 1) areas of LV cavity and myocardium, 2) dimensions of the LV cavity, 3) regional wall thicknesses (RWT), and 4) the cardiac phase, from mid-ventricle short-axis CMR images. First, we constructed a public quantification dataset Cardiac-DIG with ground truth labels for both the myocardium mask and these quantification targets across the entire cardiac cycle. Then, the key techniques employed by each submission were described. Next, quantitative validation of these submissions were conducted with the constructed dataset. The evaluation results revealed that both SG and DR methods can offer good LV quantification performance, even though DR methods do not require densely labeled masks for supervision. Among the 12 submissions, the DR method LDAMT offered the best performance, with a mean estimation error of 301 mm 2 for the two areas, 2.15 mm for the cavity dimensions, 2.03 mm for RWTs, and a 9.5% error rate for the cardiac phase classification. Three of the SG methods also delivered comparable performances. Finally, we discussed the advantages and disadvantages of SG and DR methods, as well as the unsolved problems in automatic cardiac quantification for clinical practice applications.

Diagnostic Accuracy of a Novel On-site Virtual Fractional Flow Reserve Parallel Computing System.

PURPOSE: To evaluate the diagnostic accuracy of a novel on-site virtual fractional flow reserve (vFFR) derived from coronary computed tomography angiography (CTA). MATERIALS AND METHODS: We analyzed 100 vessels from 57 patients who had undergone CTA followed by invasive FFR during coronary angiography. Coronary lumen segmentation and three-dimensional reconstruction were conducted using a completely automated algorithm, and parallel computing based vFFR prediction was performed. Lesion-specific ischemia based on FFR was defined as significant at ≤0.8, as well as ≤0.75, and obstructive CTA stenosis was defined that ≥50%. The diagnostic performance of vFFR was compared to invasive FFR at both ≤0.8 and ≤0.75. RESULTS: The average computation time was 12 minutes per patient. The correlation coefficient (r) between vFFR and invasive FFR was 0.75 [95% confidence interval (CI) 0.65 to 0.83], and Bland-Altman analysis showed a mean bias of 0.005 (95% CI -0.011 to 0.021) with 95% limits of agreement of -0.16 to 0.17 between vFFR and FFR. The accuracy, sensitivity, specificity, positive predictive value, and negative predictive value were 78.0%, 87.1%, 72.5%, 58.7%, and 92.6%, respectively, using the FFR cutoff of 0.80. They were 87.0%, 95.0%, 80.0%, 54.3%, and 98.5%, respectively, with the FFR cutoff of 0.75. The area under the receiver-operating characteristics curve of vFFR versus obstructive CTA stenosis was 0.88 versus 0.61 for the FFR cutoff of 0.80, respectively; it was 0.94 versus 0.62 for the FFR cutoff of 0.75. CONCLUSION: Our novel, fully automated, on-site vFFR technology showed excellent diagnostic performance for the detection of lesion-specific ischemia.

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.

Deep Learning Techniques for Automatic MRI Cardiac Multi-Structures Segmentation and Diagnosis: Is the Problem Solved?

Delineation of the left ventricular cavity, myocardium, and right ventricle from cardiac magnetic resonance images (multi-slice 2-D cine MRI) is a common clinical task to establish diagnosis. The automation of the corresponding tasks has thus been the subject of intense research over the past decades. In this paper, we introduce the "Automatic Cardiac Diagnosis Challenge" dataset (ACDC), the largest publicly available and fully annotated dataset for the purpose of cardiac MRI (CMR) assessment. The dataset contains data from 150 multi-equipments CMRI recordings with reference measurements and classification from two medical experts. The overarching objective of this paper is to measure how far state-of-the-art deep learning methods can go at assessing CMRI, i.e., segmenting the myocardium and the two ventricles as well as classifying pathologies. In the wake of the 2017 MICCAI-ACDC challenge, we report results from deep learning methods provided by nine research groups for the segmentation task and four groups for the classification task. Results show that the best methods faithfully reproduce the expert analysis, leading to a mean value of 0.97 correlation score for the automatic extraction of clinical indices and an accuracy of 0.96 for automatic diagnosis. These results clearly open the door to highly accurate and fully automatic analysis of cardiac CMRI. We also identify scenarios for which deep learning methods are still failing. Both the dataset and detailed results are publicly available online, while the platform will remain open for new submissions.

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 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.

Geodesic Distance Algorithm for Extracting the Ascending Aorta from 3D CT Images.

This paper presents a method for the automatic 3D segmentation of the ascending aorta from coronary computed tomography angiography (CCTA). The segmentation is performed in three steps. First, the initial seed points are selected by minimizing a newly proposed energy function across the Hough circles. Second, the ascending aorta is segmented by geodesic distance transformation. Third, the seed points are effectively transferred through the next axial slice by a novel transfer function. Experiments are performed using a database composed of 10 patients' CCTA images. For the experiment, the ground truths are annotated manually on the axial image slices by a medical expert. A comparative evaluation with state-of-the-art commercial aorta segmentation algorithms shows that our approach is computationally more efficient and accurate under the DSC (Dice Similarity Coefficient) measurements.

Spinal Bone Bruise: Can Computed Tomography (CT) Enable Accurate Diagnosis?

RATIONALE AND OBJECTIVES: The study aimed to evaluate the usefulness of dual-energy computed tomography (DECT) before and after calcium subtraction in the diagnosis of spinal bone bruise. MATERIALS AND METHODS: Among the patients who visited our emergency department between January 2013 and July 2014 who underwent both spinal DECT and magnetic resonance imaging, 38 patients (men:women = 25:13; mean age: 55.6 years, range: 28-82) were included. The patients were divided into two groups, those with and without acute spinal compression fracture, based on magnetic resonance imaging findings. In the fracture group (n = 22), the ratio of Hounsfield unit (HU) values was calculated between the fracture level and the next normal inferior vertebra in the DECT before and after calcium subtraction. In the non-fracture group (n = 16), the ratios of HU values were calculated between two normal adjacent vertebrae. The mean HU ratios were compared between the two groups. RESULTS: The mean HU ratio was higher in the fracture group (calcium subtraction: before: 1.57 and 1.59; after: 1.74 and 1.76) than the non-fracture group (before: 1.07 and 1.08; after: 1.07 and 1.07) (P < 0.001). The mean HU ratio between before and after calcium subtraction images was different only in the fracture group (P < 0.05). There was no significant difference in the area under the curve, sensitivity, specificity, positive and negative predictive values, and accuracy (before: 0.846, 87.5%, 81.2%, 87.5%, 81.2%, 85%; after: 0.865, 91.7%, 81.2%, 88%, 86.7%, 87.5% in high energy) between the images before and after calcium subtraction. CONCLUSION: The HU ratio between the fractured and next normal vertebra was diagnostic for spinal bone bruise on DECT images both before and after calcium subtraction.

Viability assessment after conventional coronary angiography using a novel cardiovascular interventional therapeutic CT system: Comparison with gross morphology in a subacute infarct swine model.

BACKGROUND: Given the lack of promptness and inevitable use of additional contrast agents, the myocardial viability imaging procedures have not been used widely for determining the need to performing revascularization. OBJECTIVE: This study is aimed to evaluate the feasibility of myocardial viability assessment, consecutively with diagnostic invasive coronary angiography (ICA) without use of additional contrast agent, using a novel hybrid system comprising ICA and multislice CT (MSCT). METHODS: In all, 14 Yucatan miniature swine models (female; age, 3 months; weight, 28-30 kg) were subjected to ICA followed by balloon occlusion (90 minutes) and reperfusion of the left anterior descending coronary artery. Two weeks after induction of myocardial infarction, delayed hyperenhancement (DHE) images were obtained, using a novel combined machine comprising ICA and 320-channel MSCT scanner (Aquilion ONE, Toshiba), after 2, 5, 7, 10, 15, and 20 minutes after conventional ICA. The heart was sliced in 10-mm consecutive sections in the short-axis plane and was embedded in a solution of 1% triphenyltetrazolium chloride (TTC). Infarct size was determined as TTC-negative areas as a percentage of total left ventricular area. On MSCT images, infarct size per slice was calculated by dividing the DHE area by the total slice area (%) and compared with histochemical analyses. RESULTS: Serial MSCT scans revealed a peak CT attenuation of the infarct area (222.5 ± 36.5 Hounsfield units) with a maximum mean difference in CT attenuation between the infarct areas and normal myocardium of at 2 minutes after contrast injection (106.4; P for difference = 0.002). Furthermore, the percentage difference of infarct size by MSCT vs histopathologic specimen was significantly lower at 2 (8.5% ± 1.8%) and 5 minutes (9.5% ± 1.9%) than those after 7 minutes. Direct comparisons of slice-matched DHE area by MSCT demonstrated excellent correlation with TTC-derived infarct size (r = 0.952; P < .001). Bland-Altman plots of the differences between DHE by MSCT and TTC-derived infarct measurements plotted against their means showed good agreement between the 2 methods. CONCLUSION: The feasibility of myocardial viability assessment by DHE using MSCT after conventional ICA was proven in experimental models, and the optimal viability images were obtained after 2 to 5 minutes after the final intracoronary injection of contrast agent for conventional ICA.