[New England
      Complex Systems Institute]
[Home] [Research] [Education] [Current Section: Activities & Events] [Community] [News] [The Complex World] [About Complex Systems] [About NECSI]
International Conference on Complex Systems (ICCS2006)

New Approaches to Computer-based Interventional Neuroradiology Training

James Rabinov
MGH/Harvard University

     Full text: Not available
     Last modified: June 22, 2006


New Approaches to Computer-based Interventional Neuroradiology Training

Rabinov J, Luboz V, Krissian K, Neumann P, Cotin S, Rabinov J, Dawson S

Interventional Neuroradiology is a growing field of minimally invasive therapies one of which is the treatment of acute stroke. Stroke is the third leading cause of death in the United States. Each year, more than 750,000 strokes result in over 150,000 deaths. Approximately 80% of strokes are ischemic, where decreased intracerebral flow results from intravascular thrombosis or embolism. Early intracerebral catheterization and lytic therapy within the first three hours after symptom onset can dissolve the clot and restore flow, reducing morbidity and improving patient outcome.

Because treatment is delivered using image guidance, skillful instrument navigation and a thorough understanding of the vascular anatomy are critical in order to avoid irreversible complications. CT angiography can diagnose intracranial vessel occlusion, but to date, nearly all training occurs on actual critically ill patients as they evolve a stroke. This training paradigm has resulted in fewer than 200 fully trained interventionalists who are qualified to treat the nationís 750,000 annual stroke patients. Clearly, a new approach to training is needed.

We have developed a real-time high-fidelity interventional neuroradiology simulator for physician training. The system creates accurate anatomical representations and uses new algorithms for fluoroscopic rendering and physics-based modeling of catheter-vessel interactions. In particular, our method for segmenting and reconstructing the vascular anatomy consists of a streamlined transition from the patientís CTA scans to three-dimensional surfaces. This method reconstructs topologically correct surface representations of tubular structures including bifurcating vessels. The present cerebrovascular model includes 6,600 vessels in both arterial and venous distributions and is optimized for real-time collision detection with virtual instruments.

The approach has been tested against a vascular phantom and actual clinical data sets. The results proved that our reconstructed surfaces are accurate and robust, yet use twenty times fewer triangles than conventional methods while producing ten times smoother surfaces.

In summary, we have defined a common set of endovascular simulation components and applied them to a complete training system for the treatment of aneurysm, arteriovenous malformation, atherosclerosis, stroke and other neurovascular procedures. This endovascular training system is the first fully physics-based, patient-specific simulator that allows multiple levels of skill acquisition and early practical experience without putting patients at risk. The cost-effective design and system compactness will permit cross-specialty interventional training which may alter basic methods of physician education and result in safer, more effective endovascular therapy.

Conference Home   |   Conference Topics   |   Application to Attend
Submit Abstract/Paper   |   Accommodation and Travel   |   Information for Participants

Maintained by NECSI Webmaster    Copyright © 2000-2005 New England Complex Systems Institute. All rights reserved.