Ultrasound-guided Spinal Injections

System setup and Guidance system interface. Live ultrasound images are augmented with the registered model, the needle and the transducer.

Abstract

Facet joint injections of analgesic agents are widely used to treat patients with lower back pain, a growing problem in the adult population. The current standard-of-care for guiding the injection is fluoroscopy, but has significant drawbacks, including the significant dose of ionizing radiation. As an alternative, several ultrasound-guidance systems have been recently proposed, but have not become the standard- of-care mainly because of the difficulty in image interpretation by anesthesiologists unfamiliar with complex spinal sonography. A solution is to register a statistical spine model, learned from pre-operative images such as MRI or CT over a range of population, to the ultrasound images and display as an overlay. In particular, we introduce an ultrasound-based navigation system where the workflow is divided into two steps. Initially, prior to the injection, tracked freehand ultrasound images are acquired from the facet joint and its surrounding vertebrae. The statistical model is then instantiated and registered to those images. Next, the real-time ultrasound images are augmented with the registered model to guide the injection. Feasibility experiments are performed on ultrasound data obtained from nine patients who had prior CT images as the gold-standard for the statistical model. We present three ultrasound scanning protocols for ultrasound acquisition and quantify the error of our model.

Papers

A statistical multi-vertebrae shape+pose model for segmentation of CT images
Abtin Rasoulian, Robert Rohling, and Purang Abolmaesumi
SPIE Medical Imaging, 2013.
Link

Lumbar Spine Segmentation Using a Statistical Multi-vertebrae Anatomical Shape+Pose Model
Abtin Rasoulian, Robert Rohling, and Purang Abolmaesumi
IEEE Transactions on Medical Imaging, 2013.
Link

Volumetric MR Images Using a Statistical Shape+Pose Model
Amin Souzani, Abtin Rasoulian, Sidney Fels, Robert Rohling, and Purang Abolmaesumi
SPIE Medical Imaging, 2014.


Abtin Rasoulian
Last modified: Nov 5, 2012