Robust vascular network extraction and understanding within hepatic biomedical images – R-VESSEL-X

Cardiovascular diseases and other blood vessels disorders increase in the world-wide scale and in particular in Occident. The evolution of computer science in researches investigating vascular networks has raised the interest in numerical reconstructing and understanding of those complex tree-like structures. The R-VESSEL-X (Robust vascular network extraction and understanding within hepatic biomedical images) project proposes original and robust developments of image analysis and machine learning algorithms integrating strong mathematical frameworks, e.g. digital geometry and topology, mathematical morphology, or graphs for reconstructing vessels of the liver beyong medical image content. The technical work packages (WPs) of the project reveal this fundamental support with WP1 and WP2 respectively aiming to study vascular enhancement and interactive segmentation of vessels, while WP3 is devoted to the reconstruction of vascular networks by integrating deep learning approaches. The centerlines (or skeletons) of vessels will be important transversal data structures employed in these three packages. Another objective of R-VESSEL-X is to diffuse research works in an open-source way, with the developments of plug-ins compatible with the ITK (Insight ToolKit) and VTK (Visualization ToolKit) librairies largely popularized by the KITWARE company. This project will also include benchmarks composed of images, associated ground-truth and quality metrics, so that researchers and engineers evaluate their novel contributions. Every developments will be managed and validated jointly with doctors actively engaged in inter-disciplinary research between computer science and heath-care. WP4 (cross-fertilization and dissemination) and WP5 (validation) will address these different issues in the project. The consortium of R-VESSEL-X is composed of the following laboratories: Institut Pascal (coordinator, head of WP3/WP5, Clermont-Ferrand), LORIA (head of WP1, Nancy) CReSTIC (head of WP2, Reims), working together with the KITWARE company (head of WP4, Lyon). This is a highly pluridisciplinary group composed of researchers in computer science-related topics (biomedical image processing, numerical simulation and analysis), applied mathematics (digital geometry and topology, mathematical morphology), working with medical doctors (radiologists, hepatologists) and young researchers and developers enrolled for the project (2 PhD students, 2 engineers, 4 MSc students). R-VESSEL-X mainly considers MRI (Magnetic Resonance Imaging), a modality attracting a lot of attention in clinical routine, since it is a non-ionizing acquisition system, and enables better diagnosis for cancerous tissue detection and tracking, mostly with contrast enhancement. Another novelty is to use micro-imaging systems (micro-MRI, micro-MR angiography, synchrotron) to bring anatomical knowledge from rodents' livers to human hepatic vessel extraction. Finally, the robustness of image analysis algorithms will be measured by an original approach, based on the outcomes of finite element numerical simulation of hepatic parenchymal blood flow, in the continuation of previous works. R-VESSEL-X will permit, at long term vision, to connect resulting fine vascular networks to hepatic cancerous tumors, as a computer-aided tool for various clinical activities (biopsy and treatments by surgery, embolization, etc.) and (early) diagnosis of aggressive cancers and in particular HCC (Hepato-Cellular Carcinoma).