Chemometrics‐based signal processing methods for biosensors in health and environment: A review

Abstract

AbstractThe increasing apprehension for health, safety and quality of life in modern society has resulted in the widespread use of biosensors. Biosensors are characterised by their high sensitivity, real‐time monitoring, and easy integration, making them indispensable for environmental monitoring on‐site, as well as invasive and non‐invasive health monitoring. Signal processing and analysis are crucial to biosensor applications, with an important role being played by chemometrics in this regard. This review presents a review of recent research findings in the fields of environmental and health monitoring. In addition, it investigates the role that chemometrics plays in the processing and analysis of biosensor data. The research comprises conventional statistical techniques, including principal component analysis and wavelet transform, as well as modern techniques of artificial intelligence, such as machine learning with neural networks. Through the examination of various algorithm strengths and weaknesses, significant recommendations are offered for biosensor applications. Furthermore, the assessment delivers focused proposals for surmounting signal processing difficulties in biosensors. Additionally, the review contains a concise analysis and reflection on the issue of multiple detection and analysis. The review intends to give essential guidance to future researchers in selecting efficient and sensible methods of data processing for their studies.