Contribution of AI and Machine Learning in the Optimization of Ultrasensitive Biosensors: Review
Abstract
Artificial intelligence (AI) and machine learning (ML) are reshaping the design, fabrication, and deployment of ultrasensitive biosensors by enabling data-driven optimization across the entire sensing pipeline. This review examines how ML assisted strategies enhance analytical performance, including limit of detection, selectivity, dynamic range, stability, and response time, in electrochemical, optical (surface plasmon resonance, surface-enhanced Raman spectroscopy, fluorescence), piezoelectric, and field-effect transistor-based biosensors. The principal ML examples used in biosensing are summarized: supervised learning for calibration and multi-analyte quantification, unsupervised learning for drift correction and background separation, and deep learning for feature extraction from high-dimensional spectra, impedance signatures, and microfluidic images. Special attention is given to AI-enabled signal processing methods that suppress noise, compensate for temperature and matrix effects, and improve robustness in complex biological samples. Moreover, AI applications in sensor optimization are discussed, including Bayesian optimization, active learning, physics-informed machine learning, and digital twin models. The role of AI in intelligent point-of-care platforms, edge deployment, smartphone readouts, and federated learning is also examined. Persistent challenges, including limited labeled data, dataset shift, lack of reporting standards, model interpretability, and regulatory validation, are analyzed alongside recommended best practices such as uncertainty quantification, explainable AI, and cross-laboratory validation.
Related
- Machine Learning-Based NOx Prediction and Combustion Optimization for a Coal-Fired Boiler: The Role of Feature Engineering
- Integrated experimental and machine learning-driven multi-objective optimization for phenol removal via silica aerogel/ZSM-5 nanocomposite: Isotherm and kinetic analysis, thermodynamic insights and pareto-based process design
- Sugarcane bagasse-MnFe2O4 magnetic biochar composite for ammonium removal: isotherms, kinetics, RSM optimization and machine learning prediction
- Green chemistry synthesized sulfur and nitrogen co-doped graphene quantum dots: a sustainable fluorescent nanosensing platform for ultrasensitive Fe³⁺ detection
- Simple, Fast, and Ultrasensitive Fluorescent Detection of Peroxymonosulfate in Chemical, Enzymatic, and Cellular Systems.
BibTeX
@article{bassir2026contribution,
title = {{Contribution of AI and Machine Learning in the Optimization of Ultrasensitive Biosensors: Review}},
author = {David Bassir and Muhammad Awais and Ata Jahangir Moshayedi},
journal = {MATEC Web of Conferences},
year = {2026},
month = sep,
doi = {10.1051/matecconf/202642204001},
url = {https://hal.science/hal-05748406v1},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Contribution of AI and Machine Learning in the Optimization of Ultrasensitive Biosensors: Review
AU - David Bassir
AU - Muhammad Awais
AU - Ata Jahangir Moshayedi
PY - 2026
DA - 2026/09/09
JO - MATEC Web of Conferences
DO - 10.1051/matecconf/202642204001
UR - https://hal.science/hal-05748406v1
AB - Artificial intelligence (AI) and machine learning (ML) are reshaping the design, fabrication, and deployment of ultrasensitive biosensors by enabling data-driven optimization across the entire sensing pipeline. This review examines how ML assisted strategies enhance analytical performance, including limit of detection, selectivity, dynamic range, stability, and response time, in electrochemical, optical (surface plasmon resonance, surface-enhanced Raman spectroscopy, fluorescence), piezoelectric, and field-effect transistor-based biosensors. The principal ML examples used in biosensing are summarized: supervised learning for calibration and multi-analyte quantification, unsupervised learning for drift correction and background separation, and deep learning for feature extraction from high-dimensional spectra, impedance signatures, and microfluidic images. Special attention is given to AI-enabled signal processing methods that suppress noise, compensate for temperature and matrix effects, and improve robustness in complex biological samples. Moreover, AI applications in sensor optimization are discussed, including Bayesian optimization, active learning, physics-informed machine learning, and digital twin models. The role of AI in intelligent point-of-care platforms, edge deployment, smartphone readouts, and federated learning is also examined. Persistent challenges, including limited labeled data, dataset shift, lack of reporting standards, model interpretability, and regulatory validation, are analyzed alongside recommended best practices such as uncertainty quantification, explainable AI, and cross-laboratory validation.
SN - 2261-236X
ER -