Each year more than 20,000 Canadians are treated surgically for lower back pain. Image guided surgical (IGS) techniques can reduce the number of complications that can arise in classical surgical techniques that can be as high as 20–30%. Intraoperative imaging can further reduce risks by improving accuracy and precision in the placement of pedicular screws used for lumbar fixation. This project will investigate the use of intra-operative ultrasound as an inexpensive alternative to intraoperative CT or MRI. We will develop techniques to precisely identify the boney surface of the vertebrae in both CT and ultrasound and use this information to improve patient-image registration required for image guided spine surgery. This in turn will improve the accuracy in guidance and precision of screw placement and result in better care for the patient.

We have tested the hypothesis that intra-operative ultrasound is viable, precise and clinically relevant to improved precision and will reduce operating time in image guided surgery of the spine. Our specific aims were:

  • To develop an automated model-based method to segment vertebrae of the human spine from 3D computed tomography data and ultrasound data.
  • To develop an automated slice-to-volume registration method using intraoperative 2D ultrasound (US) to align a patient’s vertebra to pre-operative CT images to improve the accuracy of image guided surgery and to reduce the time required for registration.
  • To validate and characterized the accuracy of the registration and segmentation methods in vitro using human plastic spine models, swine models and human cadaver specimens.
  • To evaluate the precision and speed of the ultrasound-based registration method in vivo with patients in the context of lumbo-sacral pedicle screw implantation with respect to landmark-based registration in a commercial navigation system.

Increasing the accuracy of the patient-image registration enabled the surgeon to improve pedicular screw positioning and thus decrease risks for the spinal cord, nerve roots or blood vessels. Improved precision can also increase instrumentation strength, thus preventing loosening of the misplaced hardware. When used with an anterior approach, the IGS system can be used to facilitate removal of disks or tumors, or positioning of artificial disks. By decreasing the time required for the registration procedure, overall operating time will be reduced, implying shorter muscle retraction times with the potential of reduced post-operative pain and reduced risk of infection for the patient. For the clinical team, shorter operating times will reduce fatigue, and may enable completion of more complex procedures with greater assurance. This work was supported by a CIHR operating grant (PI: Collins, co-PIs: Goulet).


[1] Fonov VS, Le Troter A, Taso M, De Leener B, Lévêque G, Benhamou M, Sdika M, Benali H, Pradat PF, Collins DL, Callot V, Cohen-Adad J. Framework for integrated MRI average of the spinal cord white and gray matter: The MNI-Poly-AMU template. Neuroimage. 2014 Sep 7;102P2:817–827

[2] G Forestier, F Lalys, DL Collins, J. Meixensberger, S Wassef, T Neumuth, B Goulet, L Riffaud, P Jannin. Multi-site study of surgical practice in neurosurgery based on Surgical Process Models, Journal of Biomedical Informatics, 46(5), October 2013, Pages 822–829

[3] Yan CX, Goulet B, Chen SJ, Tampieri D, Collins DL. Validation of automated ultrasound-CT registration of vertebrae. Int J Comput Assist Radiol Surg. 2012 Jul;7(4):601–10

[4] Yan, C. X., Goulet, B., Tampieri, D., & Collins, D. L. (2012). Ultrasound-CT registration of vertebrae without reconstruction. International journal of computer assisted radiology and surgery, 7(6), 901–909.

[5] C.X.B. Yan, B. Goulet, J. Pelletier, S.J.S. Chen, D. Tampieri and D.L. Collins, “Towards Accurate, Robust and Practical Ultrasound-CT Registration of Vertebrae for Image-Guided Spine Surgery,” International Journal of Computer Assisted Radiology and Surgery, 2011 Jul;6(4):523–37.