Machine Learning for Neurotransmitter Monitoring by Fast Voltammetry: Current and Future Prospects

Chemical neuroscience wields tools to uncover the molecular mysteries of the brain. Sensors can be fabricated with properties tailored to the scales needed to decode neurochemical information. Current instrumentation is capable of measurement rates that exceed neurochemical release rates. Modern machine learning models are approaching parameterization near the number of brain synapses. Fast voltammetry has remained a neuroanalytical workhorse technique for nearly half a century and has undergone significant transformations in many aspects due to advances in hardware and computation. Here, we review current and future uses of machine learning coupled with fast voltammetry to quantify neurochemical dynamics in the brains of behaving animal and human subjects. We focus on the advances that machine learning offers to pervasive problems in fast voltammetry. We identify current challenges and limitations for in vivo studies and delineate several routes for future development.