The accurate segmentation of blood vessels and centerline extraction are critical in vascular imaging applications, ranging from preoperative planning to hemodynamic modeling. This study introduces a novel one-stage method for simultaneous vessel segmentation and centerline extraction using a multitask neural network. We designed a hybrid architecture that integrates convolutional and graph layers, along with a task-specific loss function, to effectively capture the topological relationships between segmentation and centerline extraction, leveraging their complementary features. The proposed end-to-end framework directly predicts the centerline as a polyline with real-valued coordinates, thereby eliminating the need for post-processing steps commonly required by previous methods that infer centerlines either implicitly or without ensuring point connectivity. We evaluated our approach on a combined dataset of 142 computed tomography angiography images of the thoracic and abdominal regions from LIDC-IDRI and AMOS datasets. The results demonstrate that our method achieves superior centerline extraction performance (Surface Dice with threshold of 3 mm: 97.65%2.07%) compared to state-of-the-art techniques, and attains the highest subvoxel resolution (Surface Dice with threshold of 1 mm: 72.52%8.96%). In addition, we conducted a robustness analysis to evaluate the model stability under small rigid and deformable transformations of the input data, and benchmarked its robustness against the widely used VMTK toolkit.
The analysis was performed for input condition sensitivity of hemodynamic simulation performed for a zone of abdominal aortic aneurysm. We took three versions of patient-specific configurations of aneurysm decease and computed the haemodynamics with different spatial and velocity profiles at the aorta inlet. Their impact on hemodynamic characteristics was evaluated. At total, the study was performed for three spatial variants (uniform, parabolic, and parabolic-and-secondary-flow velocity profiles) and three versions of time behavior of velocity profiles: this produces nine cases for every of three chosen geometries. The study demonstrated that one can neglect the impact of the spatial profile for inlet velocity (including the non-coaxial velocity vector components). Meanwhile, the value of reverse diastolic flow is significant for the solution. However, simulation in the zone of abdominal aortic aneurysm does not demonstrate great differences in simulation results for the values of wall shear stress and velocity for the data averaged over a cardiac cycle. For the distribution of the oscillation index of shear, the maximum deviation from the basic solution is about ∼ 10 %, which is quite acceptable for clinical applications.
An abdominal aortic aneurysm (AAA) is a dangerous pathology that needs regular monitoring based on medical images. Currently, only visual estimates of the growth rate and methods based on the assessment of changes in the maximum diameter of the aneurysm in clinical practice are used. However, the quantitative assessment of vessel wall growth rate based on deformable image registration is gaining popularity in research. This paper presents a study of the applicability of the neural network approach of image registration for the quantitative growth assessment problem. In this study, we analyzed classical and neural network methods of image registration and used VoxelMorph and HyperMorph neural network architectures to evaluate local AAA growth based on the available dataset. Also, we compared the results of the obtained maximum local deformations of the AAA with the method of estimating the change in the maximum diameter.
1 June 2024 in Journal of Applied Mechanics and Technical Physics
A method is proposed for three-dimensional quantification of the shagginess of a pathologically altered aorta with account for a three-dimensional lumen morphology. An algorithm for assessing local shagginess is developed and implemented, and an integral criterion for shagginess in the anatomical region is proposed. The algorithm is tested on data from two patients with a shaggy and a smooth lumen. The results obtained in this study are in good agreement with visual assessment performed by expert surgeons and are also confirmed by clinical outcomes during surgery.