3D Reconstruction of Patient-Specific Carotid Artery Geometry Using Clinical Ultrasound Imaging

Abstract

One of the techniques used to diagnose carotid artery disease is the ultrasound (US) examination. The initiation and development of vascular diseases depends also on the flow conditions in the artery. Additional parameters that cannot be directly measured can be obtained by performing numerical simulations using patient-specific geometry. In this study, the Finite Element Method (FEM) is used to analyze the distribution of relevant blood flow characteristics. Images obtained from the US examinations are used to adapt the generalized carotid bifurcation model to the specific patient. The approach presented in this study combines the deep learning approach for the image segmentation and automated 3D reconstruction method to create a semi-generic geometrical model of the carotid artery that is adapted to the specific patient, using data obtained from only several US images. The presented methodology enables efficient segmentation, extraction of the morphological parameters and creation of 3D meshed volume models that can be also used for the further computational simulations of blood flow.

Improved Three-Dimensional Reconstruction of Patient-Specific Carotid Bifurcation Using Deep Learning Based Segmentation of Ultrasound Images

“…The methodology for 3D reconstruction using US images segmented with deep learning techniques is described extensively in literature [5,7]. In this paper, only the basic idea will be described.…”

“…Similarly to the previous studies [5,7], a clinical data set is used to train the CNNs. The clinical data set consisted of both longitudinal and transversal US images.…”

“…Within cited literature [5,7], the generalised model of the carotid artery is adapted to the particular patient, by using the longitudinal US image only for the ICA branch. As already mentioned in Section 2.1, within this study the longitudinal US images contained the whole carotid bifurcation.…”

“…A -reconstructed geometry using the generic model as the basis, as shown in [5]; B -reconstructed geometry using the improved methodology presented in this study.…”

“…In order to perform a more detailed analysis of the state of the patient's carotid bifurcation, it would be very useful if a three-dimensional (3D) model were available. For this purpose, the 3D reconstruction was performed in literature by using the available 2D cross-sections [5]. Within this approach, the US images were segmented by using U-Net [6] based deep convolutional neural networks (CNN).…”