Related Experiment Video
Updated: Sep 13, 2025
Application of Laparoscopic Hepatectomy Combined with Intraoperative Microwave Ablation in Colorectal Cancer Liver Metastasis
Published on: March 3, 2023
Optimization of surgical parameters for liver tumor microwave ablation assisted by hydrodissection: Solution space
Chang Yuan1, Xiaotong Yan1, Kai Yue2
1School of Energy and Environmental Engineering, University of Science and Technology Beijing, Beijing 100083, PR China.
Background And Objective:
Hydrodissection effectively protects adjacent tissues from thermal burns during microwave ablation of liver tumors. However, residual heat in the tumor and adipose tissues following the removal of the hydrodissection layer can result in thermal injuries to nearby organs. The objective of this study is to develop an optimization method for surgical parameters based on a solution space features and active learning approach to minimize the risk of postoperative thermal injury.
Methods:
A method was developed in this study to optimize surgical parameters by constructing a solution space to prevent such injuries. A three-dimensional model of the liver and colon was reconstructed using medical imaging, and an electromagnetic-thermal-porous medium flow coupling model was established. Heat transfer and tissue damage characteristics during the surgical procedure were analyzed. Tumor ablation completeness, carbonization zone size, intestinal wall damage, and surgery duration were treated as objective functions and constraints for optimization. Expressions describing the solution space features were fitted based on simulation data. Optimal surgical parameters were determined by combining active learning with neural network training, incorporating individual patient physiological parameters. Ex vivo bovine liver ablation experiments were performed to validate the accuracy of the simulation model.
Results:
The model effectively simulated tissue damage and temperature variations, with an error of approximately 5.34 % compared to ex vivo experiments. For each fitting of the solution space parameters, the goodness of fit exceeded 0.99. The optimization method theoretically achieved optimal solutions within 16 simulations, significantly reducing computation time. Additionally, the active learning algorithm reduced the root mean square error of model predictions by 25.7 %, keeping the prediction error below 6 %. Tumor conductivity was the most influential factor impacting optimization results.
Conclusions:
This study can provide a theoretical framework for optimizing ablative treatment planning to reduce the risk of adjacent tissue burns in the therapy area after microwave ablation surgery.
