3D Modelling of the Residual Freezing for Renal Cryoablation Simulation and Prediction.

Percutaneous cryoablation has become a popular alternative to open surgery for the treatment of abdominal tumors. The preoperative planning of such interventions is an essential but complicated task. It consists in predicting the best placement for several cryoprobes to optimize the resulting iceball shape, that has to cover the whole tumor, while preserving healthy tissue and surrounding sensitive structures. In the past few years, methods have been proposed to simulate the propagation of cold within the tissue, in order to anticipate the final coverage. However, all the proposed models considered the source of cold as limited to the active tip of the cryoprobe, thus omitting a residual freezing along the probe's body. The lack of precision of the resulting models can cause an underestimation of the predicted iceball leading to potential damages to healthy tissue or pain. In this paper, we describe the extension of an existing freezing simulation model to account for this effect. We detail the experimentation of our model on 5 retrospective cases, and demonstrate the improvement of the accuracy and realism of our simulation.