Semi-automatic vessel detection for challenging cases of peripheral arterial disease

• Published in: Computers in biology and medicine 2021-06, Vol.133, p.104344-104344, Article 104344
• Main Authors: Mistelbauer, Gabriel, Morar, Anca, Schernthaner, Rüdiger, Strassl, Andreas, Fleischmann, Dominik, Moldoveanu, Florica, Gröller, M. Eduard
• Format: Article
• Language: English
• Subjects: Active contours without edges; Algorithms; Angiography; Arteries; Arteriosclerosis; Atherosclerosis; Automation; Blood vessel segmentation; Blood vessels; Bone imaging; Bones; Coronary vessels; Datasets; Image segmentation; Learning algorithms; Machine learning; Medical imaging; Neural networks; Physicians; Principal components analysis; Pulmonary arteries; Technologists; Training; Veins & arteries; Vessel tracking; Workflow
• ISSN: 0010-4825; 1879-0534
• DOI: 10.1016/j.compbiomed.2021.104344

Manual or semi-automated segmentation of the lower extremity arterial tree in patients with Peripheral arterial disease (PAD) remains a notoriously difficult and time-consuming task. The complex manifestations of the disease, including discontinuities of the vascular flow channels, the presence of calcified atherosclerotic plaque in close vicinity to adjacent bone, and the presence of metal or other imaging artifacts currently preclude fully automated vessel identification. New machine learning techniques may alleviate this challenge, but require large and reasonably well segmented training data. We propose a novel semi-automatic vessel tracking approach for peripheral arteries to facilitate and accelerate the creation of annotated training data by expert cardiovascular radiologists or technologists, while limiting the number of necessary manual interactions, and reducing processing time. After automatically classifying blood vessels, bones, and other tissue, the relevant vessels are tracked and organized in a tree-like structure for further visualization. We conducted a pilot (N = 9) and a clinical study (N = 24) in which we assess the accuracy and required time for our approach to achieve sufficient quality for clinical application, with our current clinically established workflow as the standard of reference. Our approach enabled expert physicians to readily identify all clinically relevant lower extremity arteries, even in problematic cases, with an average sensitivity of 92.9%, and an average specificity and overall accuracy of 99.9%. Compared to the clinical workflow in our collaborating hospitals (28:40 ± 7:45 [mm:ss]), our approach (17:24 ± 6:44 [mm:ss]) is on average 11:16 [mm:ss] (39%) faster. [Display omitted] •Specific challenges encountered with PAD:•vessels touching bone,•discontinuities (e.g. non-uniform opacified vessels),•bone-like regions (e.g. calcifications and stents), and•metal hardware artifacts (e.g. hip joint implants).•We provide a comprehensive description of these cases and propose an approach that:•solves the above-mentioned challenges with PAD,•is faster than our current clinical approach,•provides similar accuracy and robustness, and•requires significantly less user interaction.