Brain image analysis

Magnetic Resonance Imaging (MRI) and its variants are a powerful imaging modality that encompasses rich anatomical and physiological information at a high resolution. In neurosciences these modalities have become a standard in clinical practice However, the interpretation of the images requires the combined use of different modalities, which leads to the need of computer-assisted technologies. During the last years our group has developed several methodologies to analyze MRI images with focus on fast multimodal non-rigid image registration and multimodal image segmentation for brain image tumor analysis studies. The developments are driven by clinical requirements such as computation speed, robustness, and use of standard clinical imaging protocols.

During 2012, we tested our previously developed fast multimodal non-rigid registration algorithm on high-field (7 Tesla) MR images obtaining promising results for the correction of EPI distortion in fMRI.

In brain tumor image analysis, we have further developed algorithms to automatically segment glioblastomas grade III and IV from multimodal images (i.e. T1, T1c, T2, FLAIR). The algorithms are based on supervised and unsupervised classification techniques tailored to the clinical scenario. Through this research, our group was awarded the 2nd prize in the international competition for brain tumor segmentation, held at Miccai 2012, Nice, France.

 

In the software section, we have made available different software tools to perform skull stripping, brain tissue segmentation and others.