Search

OBJECTIVE To provide an automatic method for segmentation and diameter measurement of type B aortic dissection (TBAD). MATERIALS AND METHODS Aortic computed tomography angiographic images from 139 patients with TBAD were consecutively collected. We implemented a deep learning method based on a three-dimensional (3D) deep convolutional neural (CNN) network, which realizes automatic segmentation and measurement of the entire aorta (EA), true lumen (TL), and false lumen (FL). The accuracy, stability, and measurement time were compared between deep learning and manual methods. The intra- and inter-observer reproducibility of the manual method was also evaluated. RESULTS The mean dice coefficient scores were 0.958, 0.961, and 0.932 for EA, TL, and FL, respectively. There was a linear relationship between the reference standard and measurement by the manual and deep learning method (r = 0.964 and 0.991, respectively). The average measurement error of the deep learning method was less than that of the manual method (EA, 1.64% vs. 4.13%; TL, 2.46% vs. 11.67%; FL, 2.50% vs. 8.02%). Bland-Altman plots revealed that the deviations of the diameters between the deep learning method and the reference standard were -0.042 mm (-3.412 to 3.330 mm), -0.376 mm (-3.328 to 2.577 mm), and 0.026 mm (-3.040 to 3.092 mm) for EA, TL, and FL, respectively. For the manual method, the corresponding deviations were -0.166 mm (-1.419 to 1.086 mm), -0.050 mm (-0.970 to 1.070 mm), and -0.085 mm (-1.010 to 0.084 mm). Intra- and inter-observer differences were found in measurements with the manual method, but not with the deep learning method. The measurement time with the deep learning method was markedly shorter than with the manual method (21.7 ± 1.1 vs. 82.5 ± 16.1 minutes, p < 0.001). CONCLUSION The performance of efficient segmentation and diameter measurement of TBADs based on the 3D deep CNN was both accurate and stable. This method is promising for evaluating aortic morphology automatically and alleviating the workload of radiologists in the near future.

The computed tomography angiography (CTA) postprocessing manually recognized by technologists is extremely labor intensive and error prone. We propose an artificial intelligence reconstruction system supported by an optimized physiological anatomical-based 3D convolutional neural network that can automatically achieve CTA reconstruction in healthcare services. This system is trained and tested with 18,766 head and neck CTA scans from 5 tertiary hospitals in China collected between June 2017 and November 2018. The overall reconstruction accuracy of the independent testing dataset is 0.931. It is clinically applicable due to its consistency with manually processed images, which achieves a qualification rate of 92.1%. This system reduces the time consumed from 14.22 ± 3.64 min to 4.94 ± 0.36 min, the number of clicks from 115.87 ± 25.9 to 4 and the labor force from 3 to 1 technologist after five months application. Thus, the system facilitates clinical workflows and provides an opportunity for clinical technologists to improve humanistic patient care., Manual postprocessing of computed tomography angiography (CTA) images is extremely labor intensive and error prone. Here, the authors propose an artificial intelligence reconstruction system that can automatically achieve CTA reconstruction in healthcare services.

The cerebral artery network is a complex system. Each cerebral hemisphere has three main arteries: anterior/middle/posterior cerebral artery (ACA/MCA/PCA). In the middle of the brain, these arteries form the Willis circle. The vessel diseases around the Willis circle are generally a warning signal of stroke. Computed tomography angiography (CTA) is an important tool to visualize the cerebral artery system, but due to the complexity of the artery network, it is difficult to generate straightened lumen of the arteries. Hence, radiologists need to spend a long time inspecting every corner on the original CTA image and detecting the stenosis. In this paper, we propose an automatic pipeline that can identify the main arteries, dissect them into segments, and generate their straightened lumen. The main idea is to utilize the registration to identify key points of the cerebral artery system and apply the shortest path finding algorithm to extract the main routes of the artery system. Our approach is robust to all kinds of brain/neck scanning protocols and most topological variations of the artery system. Furthermore, our approach is very fast (15.7 s per case on average). Consequently, our method is more useful in real scenarios compared with previous methods.

In aortic endovascular repair, the prediction of stented vessel remodeling informs treatment plans and risk evaluation; however, there are no highly accurate and efficient methods to quantitatively simulate stented vessels. This study developed a fast virtual stenting algorithm to simulate stent-induced aortic remodeling to assist in real-time thoracic endovascular aortic repair planning. Methods: The virtual stenting algorithm was established based on simplex deformable mesh and mechanical contact analysis. The key parameters of the mechanical contact analysis were derived from mechanical tests on aortic tissue (n=40) and commonly used stent-grafts (n=6). Genetic algorithm was applied to select weighting parameters. Testing and validation of the algorithm were performed using pre- and post-treatment computed tomography angiography datasets of type-B aortic dissection cases (n=66). Results: The algorithm was efficient in simulating stent-induced aortic deformation (mean computing time on a single processor: 13.78±2.80s) and accurate at the morphological (curvature difference: 1.57±0.57%; cross-sectional area difference: 4.11±0.85%) and hemodynamic (similarity of wall shear stress-derived parameters: 90.16-90.94%) levels. Stent-induced wall deformation was higher (p

Endovascular repair with parallel stent-grafts (SG) is a challenging technique that reconstructs the luminal flow pathways by implanting parallel-placed SGs into the vessel. After treatment, occlusion and shifting of the parallel SGs are sometimes reported, which could be fatal and difficult to be re-operated. These issues are highly related to the local hemodynamic conditions in the stented region. In this study, a patient case treated by the octopus endograft technique (a head-SG with three limb-SGs) and experienced limb-SG occlusion is studied. 3-D models are established based on computed tomography (CT) angiography datasets pretreatment and posttreatment as well as during follow-ups. Hemodynamic quantities such as pressure drop, wall shear stress-related parameters, and flow division in limb-SGs and visceral arteries are quantitatively investigated. Optimizations on the length of the head-SG and diameter of the limb-SGs are analyzed based on various scenarios. The results indicate that when reconstructing the flow pathways via octopus stenting, it is important to ensure the flow distribution as physiologically required with this new morphology. Position (or length) of the head-SG and diameter of the limb-SGs play an important role in controlling flow division, and high time average wall shear stress (TAWSS) around the head-SG acts as a main factor for graft immigration. This study, by proposing optimization suggestions with hemodynamic analyses for a specific case, implicates that pretreatment SG scenarios may assist in wise selection and placement of the device and thus may improve long-term effectiveness of this kind of challenging endovascular repair techniques.

Thoracic endovascular aortic repair (TEVAR) is commonly applied in type-B aortic dissection. For patients with dissection affects descending aorta and extends downward to involve abdominal aorta and possibly iliac arteries, false lumen (FL) expansion might occur post-TEVAR. Predictions of dissection development may assist in medical decision on re-intervention or surgery. In this study, two patients are selected with similar morphological features at initial presentation but with different long-term FL development post-TEVAR (stable and enlarged FL). Patient-specific models are established for each of the follow-ups. Flow boundaries and computational validations are obtained from Doppler ultrasound velocimetry. By analyzing the hemodynamic parameters, the false-to-true luminal pressure difference (PDiff) and particle relative residence time (RRT) are found related to FL remodeling. It is found that (i) the position of the first FL flow entry is the watershed of negative-and-positive PDiff and, in long-term follow-ups, and the position of largest PDiff is consistent with that of the greatest increase of FL width; (ii) high RRT occurs at the FL proximal tip and similar magnitude of RRT is found in both stable and enlarged cases; (iii) comparing to the RRT at 7days post-TEVAR, an increase of RRT afterwards in short-term is found in the stable case while a slight decrease of this parameter is found in the enlarged case, indicating that the variation of RRT in short-term post-TEVAR might be potential to predict long-term FL remodeling.