Evolution of dynamical neural network arrays to correct arrhythmias in a simulated human heart
John C. Gallagher · 2001
The study of evolutionary approaches to create electrical circuits is becoming increasingly widespread. This paper will present early results in evolving electrically implementable dynamical neural networks that correct several classes of arrhythmia in a simulated human heart. In addition to assessing the quality of the evolved circuits, we will examine how the evolved circuits differ from traditional pacemakers and consider what implications these differences may hold. We will also propose some objective function improvements that should increase the number of effective evolved controllers R R