A recent history of image-guided surgery in the NIST lab in the Montreal Neurological Institute

This talk will track the history of image guided surgery over the past two decades in the NeuroImaging and Surgical Technology (NIST) lab at the Montreal Neurological Institute as an excuse to talk about the joy of working with curious, bright and hardworking students.  The NIST lab was the first to use frameless stereotaxic techniques with tracked intra-operative ultrasound to improve guidance during surgery while accounting for brain shift. The procedure matches intra-operative ultrasound to the pre-operative MRI data using image intensities or blood vessels as features to drive the registration. The result of this mapping is a deformation field that can be used to warp the pre-operative data to fit the anatomy of the patient during surgery, thus correcting for brain shift.  Over the past 15 years, though the work of many students and fellows, this matching procedure has been improved by using additional anatomical landmarks, using simulated ultrasound, using statistically-driven intensity-based similarity functions and implemented with a fast GPU implementations. The NIST lab developed brain atlases that can be customized to a patients anatomy to help in the identification of deep targets for planning movement disorder surgery. In addition to the brain, the NIST lab has developed tools to register preoperative spine CT to the patient, again with intra-operative ultrasound, to help in pedicle screw insertion when instrumenting one or more vertebrae. All these tools have been made publicly available as open source neuronavigation software known as IBIS (at http://ibisneuronav.org/) and templates and atlases (at http://nist.mni.mcgill.ca/atlases/).

 

Dr. Collins is a professor in the departments of Biomedical Engineering and Neurology & Neurosurgery, associate member of the McGill Centre for Studies in Aging, and the Center for Intelligent Machines at McGill and world-renowned expert in image processing for quantitative analysis of medical images. He heads the Neuro Imaging and Surgical Technologies (NIST) laboratory at the Brain Imaging Center of the Montreal Neurological Institute where his team of ~20 students, fellows and engineers develop and use computerized image processing techniques such as non-linear image registration and model-based segmentation to automatically identify structures within the human brain.

In neurology, these tools are used to characterize differences in brain morphology of patients with multiple sclerosis, Parkinson’s disease, Alzheimer’s disease, amyotrophic lateral sclerosis and other neurodegenerative diseases in comparison to the healthy brain as it changes from birth, through adolescence and adulthood and eventually death.  In neurosurgery, these tools are used combine data from different medical imaging modalities to generate 3D models of the skin, skull, dura, cortical surface, blood vessels, functional regions and brain lesions to help neurosurgeons plan surgery.  The pre-operative images and 3D models are then integrated in a neuronavigation platform that uses intra-operative ultrasound to correct for tissue deformations during neurosurgery.

In the past 5y, his group has published more than 140 journal articles (from 390 career total) and 30 conference papers and abstracts (388 total) yielding a Google h-index=119. His work is detailed at http://nist.mni.mcgill.ca/ .