Electroporation ablation, as an innovative cancer treatment, not only preserves the structure and function of affected organs but also significantly reduces surgical risks, offers patients a safer and more effective therapeutic option, and demonstrates immense potential in the field of oncology. This paper presents the innovative design of a high-voltage nanosecond pulse generator triggered by a silicon carbide (SiC) photoconductive switch. The generator is capable of stably outputting adjustable voltages ranging from 10 kV to 15 kV, with pulse widths precisely controlled between 10 and 15 nanoseconds, and an operating frequency adjustable from 1 Hz to 10 Hz. This device enables instant activation and deactivation of the pulse generator during ablation, enhancing the efficiency of strong electric field applications and preventing overtreatment due to delayed shutdown. This paper introduces the structure and basic principles of this novel SiC photoconductive switch-triggered pulse device and reports on the impact of device-related pulse parameters on the ablation effect of hepatocellular carcinoma cells through cell experiments. Under optimal ablation parameters, the CCK8 results show that the number of viable cells is only 0.7% of that in the untreated control group after 12 h of subculture following ablation. These findings hold significant importance for expanding the application areas of SiC devices.

In this paper, a physics-based ANN assisting method for extracting transient properties of extrinsically triggering photoconductive semiconductor switches (ET-PCSSs) is proposed. It exploits the nonlinear mapping of ANN between transient current (input) and doping concentration (output). According to the basic laws of photoelectric device operating, two types of ANN models are constructed by gaussian and polynomial fitting. The mean absolute error (MAE) of forecasting transient photocurrent can be less than 10 A under low triggering optical powers, which verifies the feasibility of ANN assisting TCAD applied to PCSSs. The results are comparable to computation by Mixed-Mode simulation, yet even thousands of seconds of CPU runtime cost are saved in every period. To improve the robustness of the Poly-ANN predictor, Bayesian optimization (BO) is implemented for minimizing the curl deviation of photocurrent-time curves.