Training of a deep learning based digital subtraction angiography method using synthetic data.

BACKGROUND: Digital subtraction angiography (DSA) is a fluoroscopy method primarily used for the diagnosis of cardiovascular diseases (CVDs). Deep learning-based DSA (DDSA) is developed to extract DSA-like images directly from fluoroscopic images, which helps in saving dose while improving image quality. It can also be applied where C-arm or patient motion is present and conventional DSA cannot be applied. However, due to the lack of clinical training data and unavoidable artifacts in DSA targets, current DDSA models still cannot satisfactorily display specific structures, nor can they predict noise-free images. PURPOSE: In this study, we propose a strategy for producing abundant synthetic DSA image pairs in which synthetic DSA targets are free of typical artifacts and noise commonly found in conventional DSA targets for DDSA model training. METHODS: More than 7,000 forward-projected computed tomography (CT) images and more than 25,000 synthetic vascular projection images were employed to create contrast-enhanced fluoroscopic images and corresponding DSA images, which were utilized as DSA image pairs for training of the DDSA networks. The CT projection images and vascular projection images were generated from eight whole-body CT scans and 1,584 3D vascular skeletons, respectively. All vessel skeletons were generated with stochastic Lindenmayer systems. We trained DDSA models on this synthetic dataset and compared them to the trainings on a clinical DSA dataset, which contains nearly 4,000 fluoroscopic x-ray images obtained from different models of C-arms. RESULTS: We evaluated DDSA models on clinical fluoroscopic data of different anatomies, including the leg, abdomen, and heart. The results on leg data showed for different methods that training on synthetic data performed similarly and sometimes outperformed training on clinical data. The results on abdomen and cardiac data demonstrated that models trained on synthetic data were able to extract clearer DSA-like images than conventional DSA and models trained on clinical data. The models trained on synthetic data consistently outperformed their clinical data counterparts, achieving higher scores in the quantitative evaluation of peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) metrics for DDSA images, as well as accuracy, precision, and Dice scores for segmentation of the DDSA images. CONCLUSIONS: We proposed an approach to train DDSA networks with synthetic DSA image pairs and extract DSA-like images from contrast-enhanced x-ray images directly. This is a potential tool to aid in diagnosis.

Real-time 3D reconstruction of guidewires and stents using two update X-ray projections in a rotating imaging setup.

BACKGROUND: Vascular diseases are often treated minimally invasively. The interventional material (stents, guidewires, etc.) used during such percutaneous interventions are visualized by some form of image guidance. Today, this image guidance is usually provided by 2D X-ray fluoroscopy, that is, a live 2D image. 3D X-ray fluoroscopy, that is, a live 3D image, could accelerate existing and enable new interventions. However, existing algorithms for the 3D reconstruction of interventional material require either too many X-ray projections and therefore dose, or are only capable of reconstructing single, curvilinear structures. PURPOSE: Using only two new X-ray projections per 3D reconstruction, we aim to reconstruct more complex arrangements of interventional material than was previously possible. METHODS: This is achieved by improving a previously presented deep learning-based reconstruction pipeline, which assumes that the X-ray images are acquired by a continuously rotating biplane system, in two ways: (a) separation of the reconstruction of different object types, (b) motion compensation using spatial transformer networks. RESULTS: Our pipeline achieves submillimeter accuracy on measured data of a stent and two guidewires inside an anthropomorphic phantom with respiratory motion. In an ablation study, we find that the aforementioned algorithmic changes improve our two figures of merit by 75 % (1.76 mm → 0.44 mm) and 59 % (1.15 mm → 0.47 mm) respectively. A comparison of our measured dose area product (DAP) rate to DAP rates of 2D fluoroscopy indicates a roughly similar dose burden. CONCLUSIONS: This dose efficiency combined with the ability to reconstruct complex arrangements of interventional material makes the presented algorithm a promising candidate to enable 3D fluoroscopy.

Real-time estimation of patient-specific dose distributions for medical CT using the deep dose estimation.

PURPOSE: With the rising number of computed tomography (CT) examinations and the trend toward personalized medicine, patient-specific dose estimates are becoming more and more important in CT imaging. However, current approaches are often too slow or too inaccurate to be applied routinely. Therefore, we propose the so-called deep dose estimation (DDE) to provide highly accurate patient dose distributions in real time METHODS: To combine accuracy and computational performance, the DDE algorithm uses a deep convolutional neural network to predict patient dose distributions. To do so, a U-net like architecture is trained to reproduce Monte Carlo simulations from a two-channel input consisting of a CT reconstruction and a first-order dose estimate. Here, the corresponding training data were generated using CT simulations based on 45 whole-body patient scans. For each patient, simulations were performed for different anatomies (pelvis, abdomen, thorax, head), different tube voltages (80 kV, 100 kV, 120 kV), different scan trajectories (circle, spiral), and with and without bowtie filtration and tube current modulation. Similar simulations were performed using a second set of eight whole-body CT scans from the Visual Concept Extraction Challenge in Radiology (Visceral) project to generate testing data. Finally, the DDE algorithm was evaluated with respect to the generalization to different scan parameters and the accuracy of organ dose and effective dose estimates based on an external organ segmentation. RESULTS: DDE dose distributions were quantified in terms of the mean absolute percentage error (MAPE) and a gamma analysis with respect to the ground truth Monte Carlo simulation. Both measures indicate that DDE generalizes well to different scan parameters and different anatomical regions with a maximum MAPE of 6.3% and a minimum gamma passing rate of 91%. Evaluating the organ dose values for all organs listed in the International Commission on Radiological Protection (ICRP) recommendation, shows an average error of 3.1% and maximum error of 7.2% (bone surface). CONCLUSIONS: The DDE algorithm provides an efficient approach to determine highly accurate dose distributions. Being able to process a whole-body CT scan in about 1.5 s, it provides a valuable alternative to Monte Carlo simulations on a graphics processing unit (GPU). Here, the main advantage of DDE is that it can be used on top of any existing Monte Carlo code such that real-time performance can be achieved without major adjustments. Thus, DDE opens up new options not only for dosimetry but also for scan and protocol optimization.

Deep learning-based reconstruction of interventional tools and devices from four X-ray projections for tomographic interventional guidance.

PURPOSE: Image guidance for minimally invasive interventions is usually performed by acquiring fluoroscopic images using a monoplanar or a biplanar C-arm system. However, the projective data provide only limited information about the spatial structure and position of interventional tools and devices such as stents, guide wires, or coils. In this work, we propose a deep learning-based pipeline for real-time tomographic (four-dimensional [4D]) interventional guidance at conventional dose levels. METHODS: Our pipeline is comprised of two steps. In the first one, interventional tools are extracted from four cone-beam CT projections using a deep convolutional neural network. These projections are then Feldkamp reconstructed and fed into a second network, which is trained to segment the interventional tools and devices in this highly undersampled reconstruction. Both networks are trained using simulated CT data and evaluated on both simulated data and C-arm cone-beam CT measurements of stents, coils, and guide wires. RESULTS: The pipeline is capable of reconstructing interventional tools from only four X-ray projections without the need for a patient prior. At an isotropic voxel size of 100 µ m , our methods achieve a precision/recall within a 100 µ m environment of the ground truth of 93%/98%, 90%/71%, and 93%/76% for guide wires, stents, and coils, respectively. CONCLUSIONS: A deep learning-based approach for 4D interventional guidance is able to overcome the drawbacks of today's interventional guidance by providing full spatiotemporal (4D) information about the interventional tools at dose levels comparable to conventional fluoroscopy.

Deep learning-based coronary artery motion estimation and compensation for short-scan cardiac CT.

PURPOSE: During a typical cardiac short scan, the heart can move several millimeters. As a result, the corresponding CT reconstructions may be corrupted by motion artifacts. Especially the assessment of small structures, such as the coronary arteries, is potentially impaired by the presence of these artifacts. In order to estimate and compensate for coronary artery motion, this manuscript proposes the deep partial angle-based motion compensation (Deep PAMoCo). METHODS: The basic principle of the Deep PAMoCo relies on the concept of partial angle reconstructions (PARs), that is, it divides the short scan data into several consecutive angular segments and reconstructs them separately. Subsequently, the PARs are deformed according to a motion vector field (MVF) such that they represent the same motion state and summed up to obtain the final motion-compensated reconstruction. However, in contrast to prior work that is based on the same principle, the Deep PAMoCo estimates and applies the MVF via a deep neural network to increase the computational performance as well as the quality of the motion compensated reconstructions. RESULTS: Using simulated data, it could be demonstrated that the Deep PAMoCo is able to remove almost all motion artifacts independent of the contrast, the radius and the motion amplitude of the coronary artery. In any case, the average error of the CT values along the coronary artery is about 25 HU while errors of up to 300 HU can be observed if no correction is applied. Similar results were obtained for clinical cardiac CT scans where the Deep PAMoCo clearly outperforms state-of-the-art coronary artery motion compensation approaches in terms of processing time as well as accuracy. CONCLUSIONS: The Deep PAMoCo provides an efficient approach to increase the diagnostic value of cardiac CT scans even if they are highly corrupted by motion.

Real-time scatter estimation for medical CT using the deep scatter estimation: Method and robustness analysis with respect to different anatomies, dose levels, tube voltages, and data truncation.

PURPOSE: X-ray scattering leads to CT images with a reduced contrast, inaccurate CT values as well as streak and cupping artifacts. Therefore, scatter correction is crucial to maintain the diagnostic value of CT and CBCT examinations. However, existing approaches are not able to combine both high accuracy and high computational performance. Therefore, we propose the deep scatter estimation (DSE): a deep convolutional neural network to derive highly accurate scatter estimates in real time. METHODS: Gold standard scatter estimation approaches rely on dedicated Monte Carlo (MC) photon transport codes. However, being computationally expensive, MC methods cannot be used routinely. To enable real-time scatter correction with similar accuracy, DSE uses a deep convolutional neural network that is trained to predict MC scatter estimates based on the acquired projection data. Here, the potential of DSE is demonstrated using simulations of CBCT head, thorax, and abdomen scans as well as measurements at an experimental table-top CBCT. Two conventional computationally efficient scatter estimation approaches were implemented as reference: a kernel-based scatter estimation (KSE) and the hybrid scatter estimation (HSE). RESULTS: The simulation study demonstrates that DSE generalizes well to varying tube voltages, varying noise levels as well as varying anatomical regions as long as they are appropriately represented within the training data. In any case the deviation of the scatter estimates from the ground truth MC scatter distribution is less than 1.8% while it is between 6.2% and 293.3% for HSE and between 11.2% and 20.5% for KSE. To evaluate the performance for real data, measurements of an anthropomorphic head phantom were performed. Errors were quantified by a comparison to a slit scan reconstruction. Here, the deviation is 278 HU (no correction), 123 HU (KSE), 65 HU (HSE), and 6 HU (DSE), respectively. CONCLUSIONS: The DSE clearly outperforms conventional scatter estimation approaches in terms of accuracy. DSE is nearly as accurate as Monte Carlo simulations but is superior in terms of speed (≈10 ms/projection) by orders of magnitude.