Building a Synthetic Vascular Model: Evaluation in an Intracranial Aneurysms Detection Scenario

TL;DR

This work thoroughly describes the synthetic vasculature model, builds up a neural network designed for aneurysm segmentation and detection, and carries out an in-depth evaluation of the performance gap gained thanks to the synthetic model data augmentation.

Abstract

We hereby present a full synthetic model, able to mimic the various constituents of the cerebral vascular tree, including the cerebral arteries, bifurcations and intracranial aneurysms. This model intends to provide a substantial dataset of brain arteries which could be used by a 3D convolutional neural network to efficiently detect Intra-Cranial Aneurysms. The cerebral aneurysms most often occur on a particular structure of the vascular tree named the Circle of Willis. Various studies have been conducted to detect and monitor the aneurysms and those based on Deep Learning achieve the best performance. Specifically, in this work, we propose a full synthetic 3D model able to mimic the brain vasculature as acquired by Magnetic Resonance Angiography, Time Of Flight principle. Among the various MRI modalities, this latter allows for a good rendering of the blood vessels and is non-invasive. Our model has been designed to simultaneously mimic the arteries' geometry, the aneurysm shape, and the background noise. The vascular tree geometry is modeled thanks to an interpolation with 3D Spline functions, and the statistical properties of the background noise is collected from angiography acquisitions and reproduced within the model. In this work, we thoroughly describe the synthetic vasculature model, we build up a neural network designed for aneurysm segmentation and detection, finally, we carry out an in-depth evaluation of the performance gap gained thanks to the synthetic model data augmentation.

Figures (9)

• Figure 1: Schematic representation of the CoW. The yellow labels (from A to O) depict the particular bifurcations we are interested in for this study. The percentages within the gray discs represent the risk of aneurysm formation.
• Figure 2: Schematic representation of the whole bifurcation model. The upper part (yellow shaded block) represents the background noise modeling, whereas the lower part (light blue shaded block), shows the arterial geometry modeling. The green ellipses represent the different features the synthetic model can modify.
• Figure 3: Examples of modified bifurcation centerlines. The solid gray line represents the actual 3D branch, the black dashed line represents the best spline fit, and the dotted line represents an exaggerated modification of the spline coefficients.
• Figure 4: Computation of the distance separating the aneurysm and the bifurcation center. The ICA is located along the bisector, the distance separating points $\mathcal{A}$ (ICA center) and $\mathcal{C}$ (bifurcation node) needs to be estimated (see text).
• Figure 5: Comparison between the modeled bifurcations and the Ground Truth crop from a TOF (with/without ICA). We show both the gray level voxels patches (lower panels) and 3D bifurcation layout (upper panels).