Search

It is critical for microwave ablation (MWA) treatment planning to evaluate the changes of thermal coagulation zones. In MWA procedures, the shapes and sizes of thermal coagulation zones are gradually evolving over time. To this end, a novel characterization and mapping method of thermal coagulation zones is presented in this article. Firstly, finite element method (FEM) models of temperature distributions for 40, 45, 50, 55, and 60 W microwave ablations were built to derive thermal ablation data of ex vivo porcine livers and were compared with experimental results. Secondly, growth models of characteristic lengths were fitted. Finally, characterization functions of thermal coagulation zones were developed using these growth models. In addition, shape variation factors were incorporated to handle the minor shape variations of thermal coagulation zones. Experimental results showed that these characterization functions could accurately represent the changes of thermal coagulation zones. The standard deviations between prediction results and simulated values were less than 1 mm. The comparative results were statistically analyzed by paired t test (P > 0.05), indicating no significant differences. The proposed method can simply and effectively predict the changes of MWA coagulation zones with time, thus providing reliable coagulation dimensions for the thermal ablation therapies. [ABSTRACT FROM AUTHOR]

Temperature predictions of microwave ablation (MWA) are currently restricted by the specific temperature‐dependent data of tissue characteristic parameters. To address this issue, a new parameter feedback method based on temperature simulations and single‐thermometry measurements was presented. Experimental data of 2450‐MHz microwave antenna was obtained from ex vivo porcine livers. A temperature distribution model was constructed, and the contributions of characteristic parameters were acquired by the sensitivity analysis method. Subsequently, parameter feedbacks were conducted based on a minimization of the errors between numerical data and single‐thermometry measurements. The temperature distribution model was then optimized using the feedback parameters. Finally, temperature measurements were compared with simulation data to validate the accuracy of the model. According to the temperature distribution model with parameter feedback, the averages of maximum error, average error, as well as SD between the simulation temperatures and the measurements were 2.952, 1.323, and 0.852°C, respectively. The simulated and measured temperature changes were generally in good agreement. The proposed method can be useful in MWA temperature simulations to improve temperature prediction accuracy. [ABSTRACT FROM AUTHOR]