Fully Automated Thrombus Segmentation on CT Images of Patients with Acute Ischemic Stroke
Mahsa Mojtahedi, Manon Kappelhof, Elena Ponomareva, Henk van Voorst, Efstratios Gavves, Bart J. Emmer, Charles B. Majoie, Henk Marquering
Thrombus imaging characteristics are associated with treatment success and functional outcomes in stroke patients. However, assessing these characteristics based on manual annotations is labor intensive and subject to observer bias. Therefore, we aimed to create an automated pipeline for consistent and fast full thrombus segmentation. We first found the occlusion location using StrokeViewer LVO and created a bounding box around it. We trained dual modality U-Net based convolutional neural networks (CNNs) to subsequently segment the thrombus inside this bounding box. Segmentation results have high spatial accuracy with manual delineations and can therefore be used to determine thrombus characteristics and potentially benefit decision making in clinical practice.
Thursday 7th July
Poster Session 2.2 - onsite 11:00 - 12:00, virtual 15:20 - 16:20 (UTC+2)