Advances in metal oxide semiconductor gas sensor arrays based on machine learning algorithms

Journal of Materials Chemistry C, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 1, 2025

In this article, we summarize the progress of materials, mechanisms and ML-assisted gas sensing data processing for MOS sensor arrays, with a view to providing breakthrough direction future research.

Language: Английский

---

CO Concentration prediction in E-nose based on MHA-MSCINet

Journal of the Taiwan Institute of Chemical Engineers, Journal Year: 2025, Volume and Issue: 169, P. 105981 - 105981

Published: Jan. 22, 2025

Language: Английский

---

Olfactory Diagnosis Model for Lung Health Evaluation Based on Pyramid Pooling and SHAP-Based Dual Encoders

ACS Sensors, Journal Year: 2024, Volume and Issue: 9(9), P. 4934 - 4946

Published: Sept. 9, 2024

This study introduces a novel deep learning framework for lung health evaluation using exhaled gas. The synergistically integrates pyramid pooling and dual-encoder network, leveraging SHapley Additive exPlanations (SHAP) derived feature importance to enhance its predictive capability. is specifically designed effectively distinguish between smokers, individuals with chronic obstructive pulmonary disease (COPD), control subjects. structure aggregates multilevel global information by features at four scales. SHAP assesses from the eight sensors. Two encoder architectures handle different sets based on their importance, optimizing performance. Besides, model's robustness enhanced sliding window technique white noise augmentation original data. In 5-fold cross-validation, model achieved an average accuracy of 96.40%, surpassing that single 10.77%. Further optimization filters in transformer convolutional layer size module increased 98.46%. offers efficient tool identifying effects smoking COPD, as well approach utilizing technology address complex biomedical issues.

Language: Английский

---

Recent Advances in Signal Processing Algorithms for Electronic Noses

Talanta, Journal Year: 2024, Volume and Issue: 283, P. 127140 - 127140

Published: Nov. 1, 2024

Language: Английский

---

Rapid and Simultaneous Detection of Petroleum Hydrocarbons and Organic Pesticides in Soil Based on Electronic Nose

Sensors, Journal Year: 2025, Volume and Issue: 25(2), P. 380 - 380

Published: Jan. 10, 2025

The rapid detection of petroleum hydrocarbons and organic pesticides is an important prerequisite for precise soil management. It also a guarantee quality, environmental safety, human health. However, the current methods are prone to sample matrix interference, complex development processes, short lifespan, low accuracy. Moreover, they face difficulties in achieving simultaneous pesticides. In this paper, we developed electronic nose system based on gas technology, which includes sampling module recognition model. can simultaneously acquire odor signals soil. established model quickly distinguish between healthy soil, contaminated by hydrocarbons, achieve specific pesticide types types. performance was verified real products, experiment shows that has accuracy 100% three tasks: conditions identification, identification.

Language: Английский

---

Advanced Autonomous System for Monitoring Soil Parameters

Technologies, Journal Year: 2025, Volume and Issue: 13(1), P. 38 - 38

Published: Jan. 16, 2025

Context: This research investigates the advantages of real-time monitoring soil quality for various land management practices. It also highlights significance spatio-temporal modeling and mapping in providing a clear visual understanding how aridity changes over time across different locations. Aims: paper aims to provide comprehensive guide key processes required development laboratory-based system. Methods: The applied methodologies involved sensor deployment, data acquisition infrastructure establishment, calibration. These procedures culminated assessment model that was subsequently subjected two months laboratory testing using three distinct types. analysis yielded strong positive linear correlation between measured predicted values. Key Results: As expected, assimilation prior estimates within framework demonstrated significant enhancement accuracy estimations. Conclusions: promotes importance iterative improvements need long-term perspective plan maintenance continuous improvement such systems ecosystem is important improve ease making predictions avoid aridization. results this will be useful researchers practitioners design implementation systems.

Language: Английский

---

Response to letter to the Editor from Y. Takefuji on “Beyond principal component analysis: Enhancing feature reduction in electronic noses through robust statistical methods”

Trends in Food Science & Technology, Journal Year: 2025, Volume and Issue: unknown, P. 104918 - 104918

Published: Feb. 1, 2025

Language: Английский

---

An EWS-LSTM-Based Deep Learning Early Warning System for Industrial Machine Fault Prediction

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(7), P. 4013 - 4013

Published: April 5, 2025

Early warning systems (EWSs) are crucial for optimising predictive maintenance strategies, especially in the industrial sector, where machine failures often cause significant downtime and economic losses. This research details creation evaluation of an EWS that incorporates deep learning methods, particularly using Long Short-Term Memory (LSTM) networks enhanced with attention layers to predict critical faults. The proposed system is designed process time-series data collected from printing machine’s embosser component, identifying error patterns could lead operational disruptions. dataset was preprocessed through feature selection, normalisation, transformation. A multi-model classification strategy adopted, each LSTM-based model trained detect a specific class frequent errors. Experimental results show can failure events up 10 time units advance, best-performing achieving AUROC 0.93 recall above 90%. Results indicate approach successfully predicts events, demonstrating potential EWSs powered by enhancing strategies. By integrating artificial intelligence real-time monitoring, this study highlights how intelligent improve efficiency, reduce unplanned downtime, optimise operations.

Language: Английский

---

Gas Sensor Drift Compensation Using Semi-Supervised Ensemble Classifiers with Multi-Level Features and Center Loss

ACS Sensors, Journal Year: 2025, Volume and Issue: unknown

Published: April 8, 2025

The drift compensation of gas sensors is a significant and challenging issue in the field electronic noses (E-nose). Compensating sensor has great benefit improving performance E-nose systems. However, conventional methods often perform poorly due to complex data relationships before after drifting, or require label information for both nondrift (source data) (target enhance performance, which hard achieve even unrealistic. In this study, we propose semisupervised domain adaptive convolutional neural network (CNN) based on ensemble classifiers multilevel features, pretraining, center loss tackle problem. main idea make full use features extracted from apply Hilbert space's maximum mean discrepancy (MMD) evaluate similarity at different levels. Then corresponding MMD used as weight weighted fusion predictions classifier module, so obtain more reliable result. Furthermore, optimize training, pretraining help feature extractors learn robust common two domains. Center also applied focused learning same class. results sets demonstrate effectiveness our method. average classification accuracies under settings reach 76.06% (long-drift) 82.07% (short-drift), respectively, R2 score reaches 0.804 regression task, improvements compared with several methods. Our work provides an effective method algorithm level solve problem sensors.

Language: Английский

---

Rapid and Label-free Detection of Aflatoxin B1 in Peanut Oil Using Surface-Enhanced Raman Spectroscopy Combined with Deep Learning Model

LWT, Journal Year: 2025, Volume and Issue: unknown, P. 117738 - 117738

Published: April 1, 2025

Language: Английский

---