The objective of the current study is to develop a 3D temperature-field reconstruction method for minimally invasive cryosurgery—the destruction of undesired tissues by freezing. The new method is based on input data from medical imaging and/or an array of wireless implantable temperature sensors. A major difficulty in cryosurgery is the creation of a frozen region to conform to the planned 3D target region, while preserving the surrounding healthy tissues. In minimally invasive prostate cryosurgery for example, a dozen or more cryoprobes are inserted into the gland and operated simultaneously with a urethral warmer, in order to create the desired temperature field. Currently, the procedure is monitored with medical imaging (frequently ultrasound), sometimes with additional input from one or a couple of wired temperature sensors at the tip of a hypodermic needle. The development of ultra-miniature, wireless implantable sensors represents a parallel effort by the current research team. Temperature-field reconstruction in the current study is obtained by solving the classical bioheat equation, with input including temperature data from implantable sensors, and freezing-front location from medical imaging. A practical and computationally inexpensive approach is to assume a quasi-steady process in the frozen domain, based on the observation that the heat transfer during cryosurgery is characterized by a low Stefan number (i.e., the ratio of sensible to latent heat). It follows that the temperature distribution in the frozen region can be approximated as a steady solution at any instant, where the transient nature of the system comes about through boundary conditions. Since the thermophysical properties of the tissue are highly temperature dependent, a predictor-corrector iterative approach is adopted. A full transient bioheat transfer simulation of the procedure is generated by a commercially available code to serve as a benchmark, in order to evaluate the performance of the proposed algorithm. Results of this study suggest that: (a) the uncertainty in simulations based on the proposed algorithm is comparable with uncertainty in ultrasound imaging (1–2 mm) even in cases where only partial freezing front data is available (harvesting such data from medical imaging represents an unmet need); (b) the outcome of the proposed algorithm is equally adequate when temperature data from implantable temperature sensor is the only available input information; and, (c) a hybrid method incorporating data from imaging and temperature sensors seems to be a very promising approach for real-time feedback during cryosurgery. While the significance of the proposed method is in enabling cryosurgeons to visualize the developing thermal field in real time, the broader significance of the current study is in its applicability to other energy modalities in medicine with very little modifications. Source of funding: Award Number R21EB009370 from the National Institute of Biomedical Imaging and Bioengineering (NIBIB). Conflict of interest: None declared. cthaokar@andrew.cmu.edu
Read full abstract