Deniz Jafari (MHSc): Development and evaluation of algorithms for accurate navigation of radiofrequency ablation for the treatment of spinal metastases

When:
November 22, 2017 @ 9:30 am – 10:00 am
2017-11-22T09:30:00-05:00
2017-11-22T10:00:00-05:00
Where:
Rosebrugh Building
Rosebrugh Bldg, Toronto, ON M5S 3G9
Canada

Room: RS 211

Abstract:

Background
The metastasis of cancer into the bony skeleton is common in late stage disease. The vertebrae are the most common sites of bony metastases. The presence of metastatic disease may lead to instability resulting in pain, fracture and spinal cord or nerve root compression, causing severe pain, weakness, and reduction in quality life of cancer patients. The early targeted treatment of these tumours may prevent or reduce negative skeletal related events. New technology improvements in radiofrequency ablation (RFA) has adapted this technology for local bone tumour ablation. However currently, manual treatment planning and execution is dependent on operator experience and without quantitative evaluation of outcome. There is a need to develop and evaluate treatment planning and navigation in the spinal RFA workflow.

Methods
This project has developed a navigation workflow for vertebral RFA utilizing software to integrate pre-procedural imaging data (CT images) with RFA treatment plans and intra-procedural imaging data (3D CT images obtained from O-arm, and later 2D fluoroscopy images). The software is implemented in Slicer, an open source platform for medical imaging. Image registration is performed by manual initial landmark registration, followed by rigid image registration algorithm implemented using elastix toolbox. The RFA probe poses are then tracked using the Medtronic Stealthstation’s optical tracking to ensure that the final probe pose matches the prescribed pose. The treatment visualization throughout the procedure provides the guidance necessary to place the probe according to the plan and perform the procedure accurately. Evaluation of the system will be conducted in phantoms and animal/cadaveric spines.

Results
The developed software is anticipated to yield a clinical workflow capable of implementing optimal / redundant RFA probe poses that are required to ablate the tumour cells within a planned treatment margin with spatial error of less than 1-2mm.

Significance
This work is expected to improve the outcomes of RFA treatment for vertebral metastases by utilizing treatment planning and intraoperative imaging data to guide the procedure. Ultimately, more accurate and targeted RFA treatment is expected to improve the safety and efficacy of this procedure, and enhance the quality of life for patients with spinal metastases.