Optimization of Ablation Area and Electrode Positioning in High Frequency Irreversible Electroporation via Machine Learning

The adoption of high-frequency irreversible electroporation in oncology opens new perspectives in terms of types of treatable tumours, and treatment effectiveness. Nevertheless, a large number of parameters can influence the efficiency of this procedure. In this paper, we present a machine-learning strategies (more specifically artificial neural networks) as an appropriate approach to predict the ablation area and some electrode characteristics, thus possibly rendering final electroporation results superior, and achievable in a reduced time.