Fast inverse prediction of the freezing front in cryosurgery

Cryosurgery has become a well-established technique for the ablation of undesirable tissues such as tumors and cancers. The motivation for this study is to improve the efficacy and safety of this technique. This study presents an inverse heat transfer method for monitoring the motion of the freezing front from a cryoprobe. With the help of a thermocouple inserted into the layer of diseased tissue, the inverse heat transfer method estimates simultaneously the blood perfusion rate and the thermal conductivities of both frozen and unfrozen tissues. This information is then fed to the Pennes bioheat equation that: (1) calculates the time-varying temperature distribution inside the layer of tissue and (2) predicts the motion of the freezing front. The effect of the most influential parameters on the inverse predictions is investigated. These parameters are (1) the initial guesses for the unknown Levenberg-Marquardt polynomial parameters of the thermo-physical properties; (2) the temperature of the cryoprobe; (3) the heat transfer coefficient of the impinging jet of liquid nitrogen; and (4) the noise on the temperature data recorded by the thermocouple probe. Results show that the proposed inverse method is a promising alternative to ultrasound and Magnetic Resonance Imaging (MRI) for monitoring the motion of the freezing front during cryosurgery. For all the cryogenic scenarios simulated, the predictions of the inverse model remain accurate and stable.