Machine learning-assisted optical nano-sensor arrays in microorganism analysis

Microbial infection can cause problems for public health, and to realize efficient microorganism detection is of great importance. However, the simultaneous identification of microorganism still faces challenges due to the high similarity of the surface microenvironment. With the assistance of machine learning algorithms, nanomaterials-based optical sensor arrays are emerging as a promising analysis technique for microorganism discrimination with the merits of high sensitivity, time-saving and easy operation. We present here the recent development of machine learning assisted optical sensor arrays for microor-ganism identification. In the first part, five types of optical nano-sensor arrays that include fluorescent sensor arrays, colorimetric sensor arrays, multi-response-based sensor arrays, SERS-based sensor arrays and FTIR-based sensor arrays are discussed. Then, eight commonly used machine learning algorithms in the array-based sensors are introduced. Detailed calculation principles involved in the statistical analysis of array-based sensors are overviewed. It is ended by outlining the current challenges and perspectives.(c) 2023 Elsevier B.V. All rights reserved.

Automated summary

Microbial infection poses challenges for public health, and efficient microorganism detection is crucial. However, simultaneous identification of microorganisms is difficult due to the similarities in their surface microenvironments. Machine learning assisted optical sensor arrays, based on nanomaterials, are emerging as promising analysis techniques for microorganism discrimination with high sensitivity, time-saving, and easy operation. This article discusses recent developments in machine learning assisted optical sensor arrays for microorganism identification. It covers five types of optical nanosensor arrays and eight commonly used machine learning algorithms in array-based sensors, while providing an overview of the statistical analysis principles involved. The current challenges and future perspectives are also outlined.