Multiple sclerosis (MS) is a chronic, disabling neurodegenerative disease, characterised by lesions in the central nervous system. In spite of being the most common neurological condition among young adults in the UK, many aspects of MS are still not well understood. Dr. Nikos Evangelou and his team at the MS clinic in Nottingham University Hospital (UK) have been investigating a number of hypotheses and, in particular, the role of veins, both inside and outside lesions. Segmenting veins is an ardous task, which most existing approaches fail to tackle in an adequate manner.
We were commissioned by Dr. Evangelou to develop an automated vein segmentation software which took as input the specific magnetic resonance images previously optimised to visually highlight veins in MS patients. Our software produces probabilistic vein maps, which the team can then correlate with clinical parameters.
I contacted Ilixa to develop a fully automated software that would segment out veins in the central nervous system. Drawing on their image analysis experience, Ilixa designed a novel scientific approach, which they piloted on our preliminary data with excellent results.Dr. Nikos Evangelou, consultant neurologist at Nottingham University Hospital
We selected MATLAB® to develop our segmentation algorithm. Its fully integrated visualisation, computation and programming environment made it possible to tightly interact with Dr. Evangelou's team and to make sure that the final product satisfies their specifications.