Abstract

In this paper, we present a robust methodology using mathematical pattern recognition schemes to detect and classify events in action potentials for recognizing toxins in biological cells. We focus on event detection in action potential via abstraction of information content into a low dimensional feature vector within the constrained computational environment of a biosensor. We use generated families of action potentials from a classic Hodgkin–Huxley model to verify our methodology and build toxin recognition engines. We demonstrate that good recognition rates are achievable with our methodology.