Particle swarm optimization solution for roll-off control in radiofrequency ablation of liver tumors: Optimal search for PID controller tuning

• Published in: PloS one 2024-06, Vol.19 (6), p.e0300445
• Main Authors: Faria, Rafael Mendes, Rosa, Suélia de Siqueira Rodrigues Fleury, Nunes, Gustavo Adolfo Marcelino de Almeida, Santos, Klériston Silva, de Souza, Rafael Pissinati, Benavides, Angie Daniela Ibarra, Alves, Angélica Kathariny de Oliveira, da Silva, Ana Karoline Almeida, Rosa, Mario Fabrício, Cardoso, Antônio Aureliano de Anicêsio, Faria, Sylvia de Sousa, Berjano, Enrique, da Rocha, Adson Ferreira, Dos Santos, Ícaro, González-Suárez, Ana
• Format: Article
• Language: English
• Subjects: Ablation; Ablation (Surgery); Algorithms; Animals; Biology and Life Sciences; Cancer therapies; Carcinoma, Hepatocellular - surgery; Catheter Ablation - methods; Chemotherapy; Computer and Information Sciences; Computer Simulation; Control algorithms; Controllers; Drug resistance; Electrodes; Heat; Hepatocellular carcinoma; Humans; Hydration; Immunotherapy; Liver cancer; Liver Neoplasms - surgery; Liver tumors; Mean; Medicine and Health Sciences; Optimization; Particle swarm optimization; Physical Sciences; Proportional integral derivative; Radio frequency; Radiofrequency ablation; Radiofrequency Ablation - methods; Research and Analysis Methods; Standard deviation; Statistical analysis; Temperature; Thermal energy; Transfer functions; Tumors; Tuning; Variance; Variance analysis; Veins & arteries
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0300445

The study investigates the efficacy of a bioinspired Particle Swarm Optimization (PSO) approach for PID controller tuning in Radiofrequency Ablation (RFA) for liver tumors. Ex-vivo experiments were conducted, yielding a 9th order continuous-time transfer function. PSO was applied to optimize PID parameters, achieving outstanding simulation results: 0.605% overshoot, 0.314 seconds rise time, and 2.87 seconds settling time for a unit step input. Statistical analysis of 19 simulations revealed PID gains: Kp (mean: 5.86, variance: 4.22, standard deviation: 2.05), Ki (mean: 9.89, variance: 0.048, standard deviation: 0.22), Kd (mean: 0.57, variance: 0.021, standard deviation: 0.14) and ANOVA analysis for the 19 experiments yielded a p-value ≪ 0.05. The bioinspired PSO-based PID controller demonstrated remarkable potential in mitigating roll-off effects during RFA, reducing the risk of incomplete tumor ablation. These findings have significant implications for improving clinical outcomes in hepatocellular carcinoma management, including reduced recurrence rates and minimized collateral damage. The PSO-based PID tuning strategy offers a practical solution to enhance RFA effectiveness, contributing to the advancement of radiofrequency ablation techniques.