Brainbox

Quantitative analysis of MR brain images for cerebrovascular disease management

Diagnosis, prognosis, and treatment of patients with cerebrovascular diseases critically depend on robust quantification of brain structures and on identification and characterization of abnormalities, to establish pathology status and progression. High-field (3 Tesla, 7 Tesla) multi-sequence MRI is preeminently suited to provide this information. Several large-scale annotated data sets of patients with varying disease characteristics are – or will shortly become – available for development and evaluation of analysis methods.

Our starting point will be volumetric analysis software for mono-sequence (T1-weighted) MRI, developed by Philips. This software and existing methods from academia will be evaluated on the clinical data sets, as concerns accuracy and robustness. Next, analysis methods for multi-sequence MRI protocols will be developed, with two goals:

  1. to improve volumetric analysis of tissues at large (white matter, grey matter, cerebrospinal fluid, etc.)
  2. to provide a quantitative multimodal assessment of tissue and lesion heterogeneity that allows characterization and localization of different types of pathology.

The project is aimed at producing academic output in terms of publications and presentations at international conferences, industrial innovation in terms of improved and readily marketable analysis software for MR brain images, and clinical progress in terms of more efficient and more effective management of patients with cerebrovascular disease.

Koen VinckenKoen Vincken graduated in Informatics (Computer Science) at the Delft University of Technology in 1989. In 1995 he received his PhD at the Utrecht University. The title of his thesis was: Probabilistic Multiscale Image Segmentation by the Hyperstack. He then continued working at the University Medical Center Utrecht at the Psychiatry department in a combined research project with the Image Sciences Institute. Currently, Koen is working as an Associate Professor at the Image Sciences Institute of the Utrecht Medical Center Utrecht. During the last years he was responsible for the external projects that were related to the ISI. In this respect, he developed several applications that are currently being used by other departments, such as Radiology, Anatomy, Orthopaedics and Rheumatology. Koen is project leader of the Brainbox project, that focuses on the quantitative analysis of MR brain images for cerebrovascular disease and the translation from research to industry. In addition, Koen is working on the development of a new digital tool (VQuest) for assessment with two- and three-dimensional images. VQuest is currently being valorized with the help of Utrecht Holdings. VQuest is currently the application used for all radiological assessments in the UMC Utrecht for different levels of expertise (2nd year medical students, radiology residents).