Cerebral Endovascular Navigation
 

Computer-Assisted Endovascular Navigation for the Treatment of Brain Aneurysms

This work is a collaboration between the Laboratory of Mathematics in Imaging and the Department of Neurosurgery at Brigham and Women's Hospital.

Project team

  • Sonia Pujol, Ph.D, Laboratory of Mathematics in Imaging, Department of Radiology, Brigham and Women's Hospital, Harvard Medical School
  • Kai Frerichs, M.D., Department of Neurosurgery, Brigham and Women’s Hospital, Harvard Medical School
  • Carl-Fredrik Westin, Ph.D., Laboratory of Mathematics in Imaging, Department of Radiology, Brigham and Women's Hospital, Harvard Medical School

LevelSetModel3.PNG

Clinical Context

The development of endovascular techniques has led to new therapeutic methods for the reconstructive treatment of neurovascular diseases. Interventional catheter-based therapy presents several challenges resulting from the use of fluoroscopic guidance. The procedure involves a clinically significant radiation dose to patient and medical team, along with a difficult mental registration of 2D projective X-Ray images with 3D DSA models. Our hypothesis is that merging imaging and localization data can provide real-time 3D navigation without radiation.

The objectives of this research work are to improve patient safety by reducing intra-operative radiation exposure, and to provide clinicians with a real-time 3D navigation system for endovascular therapy.

We have developed a navigation platform that integrates magnetic tracking of surgical tools and 3D visualization of their position inside reconstructed models of the vasculature. We developed our system in the 3DSlicer software environment. Our strategy relies on the co-registration of 3D imaging and magnetic localization data.

Figure2.JPG
Endovascular Phantom

Validation

We built a phantom of the head composed of a styrofoam skull and a model of the cerebral vasculature. A computer-assisted endovascular procedure was then performed on the phantom. The pre-operative phase consisted in a CT acquisition with a Siemens Sensation 16 multidetector CT (slice thickness 2 mm, reconstruction interval 1.5 mm), and the construction of a 3D model of the vasculature. During the intra-operative phase, a magnetic localizer (MicroBird, Ascension Technology) tracked the position of an endovascular guide whose tip was equipped with a magnetic sensor. The indications of the navigation system were used to reach the control points with the endovascular guide. The accuracy was defined as the distance between the actual location of the guide, and its expected position measured for each control point in CT data.

Results

The navigation system displays in real-time within the 3D CTA model, the position of the magnetically tracked guide. The mean accuracy of the system was 3.72 mm, and the maximal error on the distance was 4.94 mm.

Kai_Frerichs_OR.png
Dr. Frerichs in the OR

Conclusion

We have developed a navigation system to assist neuroradiologists in the endovascular treatment of brain aneurysms. These phantom experiments demonstrated that we can provide real-time 3D visualization of the position of endovascular components in cerebral arteries with clinically relevant accuracy, and thus expect to reduce the use of fluoroscopy during neuroradiological interventions.

Benefits

This new navigation platform is a light-weight and robust prototype, that requires minimum additional training for the clinicians.

 

 
HomePersonnelResearchPublicationCollaborators