This work explores a Generative Adversarial Network (GAN) based approach for hemorrhage detection on color Doppler ultrasound images of blood vessels. Given the challenges of collecting hemorrhage data and the inherent pathology variability, we investigate an unsupervised anomaly detection network which learns a manifold of normal blood flow variability and subsequently identifies anomalous flow patterns that fall outside the learned manifold. As an initial, feasibility study, we collected ultrasound color Doppler images of brachial arteries from 11 healthy volunteers. The images were pre-processed to mask out velocities in surrounding tissues and were subsequently cropped, resized, augmented and normalized. The network was trained on 1530 images from 8 healthy volunteers and tested on 70 images from 2 healthy volunteers. In addition, the network was tested on 6 synthetic images generated to simulate blood flow velocity patterns at the site of hemorrhage. Results show significant (p<0.05) differences in anomaly scores of normal arteries and simulated injured arteries. The residual images, or the reconstruction error maps, show promise in localizing anomalies at pixel level.
Fractional Flow Reserve (FFR), the ratio of arterial pressure distal to a coronary lesion to the proximal pressure, is indicative of its hemodynamic significance. This quantity can be determined from invasive measurements made with a catheter, or by using computational methods incorporating models of the the coronary vasculature. One of the inputs needed by a model-based approach for estimating FFR from Computed Tomography Angiography (CTA) images (denoted FFR-CT) is the geometry of the coronary arteries, which requires segmentation of the coronary lumen. Several algorithms have been proposed for coronary lumen segmentation, including the recent application of machine learning techniques. For evaluating these algorithms or for training machine learning algorithms, manual segmentation of the lumen has been considered as ground truth. However, since there is inter-subject variability in manual segmentation, it would be useful to first assess the extent to which this variability affects the predicted FFR values. In the current study, we evaluated the impact of inter-subject variability in manual segmentation on computed FFR, using datasets with three different manual segmentations provided as part of the Rotterdam Coronary Artery Evaluation Framework. FFR was computed using a coronary blood flow model. Our results indicate that variability in manual segmentations on FFR estimates depend on the FFR value. For FFR ≥ 0.97, variability in manual segmentations does not impact FFR estimates, while, for lower FFR values, the variability in manual segmentations leads to significant variability in FFR. The results of this study indicate that researchers should exercise caution when treating manual segmentations as ground truth for estimating FFR from CTA images.
One of the factors affecting the accuracy of patient-specific, imaging-based computational hemodynamic studies is the accuracy of geometric models created from medical images. In the present study we have investigated as to how accurate the geometric models should be in the context of cerebral aneurysms in order to obtain an accurate reproduction of intra-aneurysmal hemodynamics in individual patients using numerical simulations. Computed tomography angiography (CTA) images obtained for a patient-specific anterior communicating artery (ACoA) aneurysm and a patient-specific middle cerebral artery (MCA) aneurysm were used to construct the geometric models. For each aneurysm, two models were created, one using a different threshold value for image segmentation than the other. The average distance between the models was about the size of one in-plane pixel. It was found that for the MCA aneurysm, the simulated pressure and shear stress distributions for the two models were entirely different while for the ACoA aneurysm the mean pressure distribution obtained for the two models were similar, but the shear stress distributions were completely different. These results indicate that accurate reproduction of intra-aneurysmal hemodynamics would require the geometric reconstruction from medical images to be highly accurate.
The accuracy and reproducibility of hemodynamic simulation for a brain aneurysm system was determined by comparison of physical measurements made in a curved duct with the corresponding simulations produced by three different solvers, and by inter-solver comparison of blood flow in a patient-specific, imaging-based model of an aneurysm. The simulations were in close agreement with measurements made in the square duct. This suggests that hemodynamic simulation is accurate for models with strong curvature flow. The simulation results produced by solvers using the model of the brain aneurysm were consistent with each other, suggesting that hemodynamic simulations of patient-specific imaging-based aneurysm models are consistent and reproducible by different solvers. These results support the validity of patient-specific imaging-based simulations.
Numerical simulations of pulsatile blood flow were conducted in a patient-specific model of an anterior communicating artery aneurysm that was found to grow over time. The effect of changes in inflow parameters on the numerical simulation results was also investigated since patient-specific velocity measurements were not available to be used as inflow conditions. It was found that shear stress distribution in the region where aneurysm growth is observed is sensitive to how flow rates are distributed in the A1 segments of the anterior cerebral arteries. The impingement location for the blood stream coming from the left anterior cerebral artery was also sensitive to the flow rate distribution in the A1 segments. Further, it was found that changing the inflow Reynolds number without altering the flow rate distribution in the A1 segments also affects the shear stress distribution on the aneurysm surface. These results suggest that at least for the anterior communicating artery aneurysm considered in this study, knowledge of the actual velocities in the A1 segments is necessary to make judgments on the hemodynamic parameter responsible for the aneurysm's growth.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.