Ultrasound imaging is commonly used for patients with atheroscelerotic plaques in the carotid artery. While B-mode ultrasound can be used for detection and measurement of these plaques, interpreting these images can be a subjective and time-consuming task. Deep learning algorithms have been proven to be an effective tool for interpreting medical images, especially for classification and segmentation tasks. Here, we propose a deep learning model to automatically detect and measure plaques in transverse B-mode images of the carotid artery.

Computational Fluid Dynamics (CFD) modeling of blood flow is a promising technique to obtain hemodynamic biomarkers and predict recurrent strokes. Few studies, however, have validated CFD against other methods for measuring hemodynamics, e.g. phase contrast MRI (PC-MRI). To resolve the uncertainties about the accuracy of CFD, this study validates the results of an automated software that combines deep learning and CFD for simulating blood flow given an MR angiogram against PC-MRI.

The accurate segmentation of medical images has important consequences in clinical applications. Noisy and artefact-heavy images can result in erroneous image segmentation and often require expert understanding of the target anatomy by clinicians to interpret and compensate for missing and obfuscated data. This is especially true in ultrasound imaging where shadowing and speckle artefacts are common.

Ultrasound imaging is commonly used for detection and measurement of carotid plaques, an important cause of ischemic stroke. Despite its importance, accurate interpretation of ultrasound can be difficult and subjective. Here, we evaluated the accuracy of our deep learning model for automatic detection of carotid plaques in b-mode ultrasound against expert interpretation of the images.

Computational models of blood flow in cerebral arteries present an opportunity to improve stroke prediction using hemodynamic biomarkers. Such models, however, are rarely validated against real clinical data. We evaluated the accuracy of a computational fluid dynamics (CFD) model which uses allometric scaling laws against population-based Phase Contrast MRI (PC-MRI) measurements of blood flow in the brain.

Analysis of cerebral arteries in MRA TOF is challenging and time consuming. Established image processing methods suffer from artefacts, especially in the presence of pathology. We propose and assess the performance of a 3D convolutional neural network for automatic segmentation of cerebral arteries in MRA TOF which is robust to common MRI artefacts, thanks to implementation of a custom loss function.

Atherosclerotic carotid plaques are an important cause of ischemic stroke. B-Mode ultrasound imaging is commonly used for detection and measurement of these plaques, yet image interpretation is difficult and subjective. See-Mode’s software uses deep learning to automatically detect and measure plaque in B-mode ultrasound.

Computational fluid dynamics (CFD) can be used to model blood flow based on static scans (like CT or MR angiograms), but such models are rarely validated in real clinical settings. We evaluated the accuracy of our automated vascular analysis software that simulates blood flow given a CT or MR angiogram against NASCET grading done by duplex ultrasound.

Intraplaque hemorrhage increases the risk of stroke in patients with carotid artery atherosclerosis. Detecting intraplaque hemorrhage in the medical images is difficult even for expert clinicians. We evaluate the ability of our software that uses machine learning to improve intraplaque hemorrhage detection in b-mode ultrasound.