Advancing Biosensors with Machine Learning

Abstract

Chemometrics play a critical role in biosensorsbased detection, analysis, and diagnosis. Nowadays, as a branch of artificial intelligence (AI), machine learning (ML) have achieved impressive advances. However, novel advanced ML methods, especially deep learning, which is famous for image analysis, facial recognition, and speech recognition, has remained relatively elusive to the biosensor community. Herein, how ML can be beneficial to biosensors is systematically discussed. The advantages and drawbacks of most popular ML algorithms are summarized on the basis of sensing data analysis. Specially, deep learning methods such as convolutional neural network (CNN) and recurrent neural network (RNN) are emphasized. Diverse ML-assisted electrochemical biosensors, wearable electronics, SERS and other spectra-based biosensors, fluorescence biosensors and colorimetric biosensors are comprehensively discussed. Furthermore, biosensor networks and multibiosensor data fusion are introduced. This review will nicely bridge ML with biosensors, and greatly expand chemometrics for detection, analysis, and diagnosis.

Sensitivity‐Enhancing Strategies in Optical Biosensing

“…[116][117][118][119][120] With its capability of processing a large amount of data, machine learning enables the detection of complex and/or marginally varying sensing signals in an accurate and rapid way. [121][122][123] Along with their contributions to developing autonomous systems and optimizing sensor designs, machinelearning-based approaches can fully benefit multiplexed and real-time sensing (e.g., wearable health monitoring systems) where complex and fluctuating signal matrices must be crossinterpreted to draw diagnostic outcomes. To conclude, high sensitivity combined with user-friendliness and facile readouts will eventually enable larger-scale uses of various sensors for self-and home-diagnosis and the Internet of Things.…”

Surface-Enhanced Raman Scattering (SERS) Spectroscopy for Sensing and Characterization of Exosomes in Cancer Diagnosis

“…To overcome this limitation, the authors employed deep learning algorithms to analyze the spectroscopic signals of exosomes. Deep learning (DL) is a machine learning method based on artificial neural networks that effectively process big sensing data for complex matrices or samples, allowing classification, identification, and pattern recognition [ 116 ]. Notably, DL algorithms have shown to be extremely beneficial for analyzing spectroscopic data in biosensing applications [ 116 ].…”

“…Deep learning (DL) is a machine learning method based on artificial neural networks that effectively process big sensing data for complex matrices or samples, allowing classification, identification, and pattern recognition [ 116 ]. Notably, DL algorithms have shown to be extremely beneficial for analyzing spectroscopic data in biosensing applications [ 116 ]. In this abovementioned work [ 90 ], besides a mere classification of the SERS data from healthy controls and cancer patients, deep learning was used to establish a correlation between the exosome data from individual lung cells with the overall patient’s histological characteristics.…”

Nanostructures for Biosensing, with a Brief Overview on Cancer Detection, IoT, and the Role of Machine Learning in Smart Biosensors

“…Some of the primary machine-learning algorithms which are being widely used for such purposes include support-vector machine, random forest, artificial, and convolutional neural networks, Naïve Bayes, convolutional neural network, and κ-nearest neighbor (κNN). A more in-depth review of the various algorithms for implementing machine learning for biosensing purposes can be found here [236]. Due to their advanced pattern-recognition abilities, machine-learning algorithms can assist nano-biosensors in extracting information from raw data that would be otherwise not apparent.…”