Integrating AI with Biosensors and Voltammetry for Neurotransmitter Detection and Quantification: A Systematic Review

The accurate and timely diagnosis of neurodegenerative disorders such as Parkinson’s disease, Alzheimer’s disease and major depressive disorder, critically depends on the real-time monitoring and precise interpretation of authentic neurotransmitter (NT) signal dynamics in complex biological fluids (CBFs) of the body, including cerebrospinal fluids. NT dynamic patterns are determined by both the types and concentrations of NTs at any given time in CBFs. However, current biosensors face significant limitations in sensitivity and selectivity, thereby hindering reliable estimation (detection and quantification) of NTs. To overcome these limitations, nanomaterials and bioenzymes have been employed to modify biosensor interfaces, often through immobilization on biopolymer-functionalized surfaces. Despite these advancements, challenges such as signal convolution and interferences from complex NT interactions, electrode fouling, and inter-NT crosstalk remain unresolved. Recent developments in Artificial Intelligence (AI), particularly in machine learning (ML), pattern recognition (PR) and deep learning (DL), have demonstrated substantial promise in deconvoluting multiplexed NT signals to reveal accurate NT detection and quantification. This review synthesizes findings from 33 peer-reviewed studies, providing a comprehensive overview of the application of AI-based approaches for automated NT estimation. It first surveys the most frequently investigated NTs, associated estimation modalities and data acquisition techniques. Subsequently, it evaluates a range of AI methodologies employed for NT estimation. The review concludes by discussing the prevailing challenges in NT estimation and highlights prospective applications of AI-enhanced NT monitoring, including Closed-Loop Deep Brain Stimulation (CLDBS), which holds promises for advancing therapeutic treatments for neurodegenerative conditions.