Toward Scalable Bionic Cardiac Systems: Energy-Efficient and High-Performance FPGA Modeling of Pacemaker Cells via Nonlinear Simplification

Abstract

This study introduces a novel methodology for digital modeling of biological cells and neurons for neuromorphic hardware systems with medical and bionic applications. The method unifies multiple nonlinear terms into a single expression, improving computational performance and scalability while reducing reliance on expensive nonlinear components. The method is generalizable to a wide range of neuronal and biological models; however, in this study, it is demonstrated on the sinoatrial (SA) node cell model, achieving a $6.784\times $ performance improvement and a 47.3% reduction in energy consumption on a Zynq XC7Z010 FPGA board. It also supports large-scale integration, demonstrated by a 300-cell network and an 800-cell network interconnected via gap junctions. These results highlight the potential for energy-efficient, real-time bionic systems in cardiac modeling and personalized healthcare technologies.