Automated cement segmentation in

11 Vertebroplasty is a minimally invasive procedure with many benefits. However, the 12 procedure is not without risks and complications. Leakage of the cement out of the 13 vertebral body and into the surrounding tissues is one of the most serious compli14 cations of vertebroplasty. Cement can leak into spinal canal, venous system, soft 15 tissues, lungs and intradiscal space, causing serious neurological complications, tis16 sue necrosis or pulmonary embolism. In this work we present a method for automatic 17 segmentation and tracking of bone cement during vertebroplasty procedures, as a 18 first step towards building a warning system to avoid cement leakage outside the 19 vertebral body. We show that using active contours based on level sets the shape 20 of the injected cement can be accurately detected. We have improved the model 21 for segmentation proposed in our previous work, by including a term that restricts 22 the level set function to the vertebral body. We have applied the method to a set 23 of real intra-operative X-ray images and our results show that the algorithm can 24 successfully detect different shapes with blurred and not well defined boundaries, 25 where the classical active contours segmentation is not applicable. The method was 26 positively evaluated by physicians. 27