Intelligent detection and classification of tetracycline drugs by rare earth fluorescence sensing platform based on deep learning algorithm and STM32 microcontroller

The excessive use of tetracycline antibiotics has led to an increase in antibiotic resistance among bacteria, complicating the treatment of infections. Additionally, residues of these antibiotics have become environmental pollutants, posing a threat to ecosystems. It is imperative to develop a rapid...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Sensors and actuators. B, Chemical Jg. 445; S. 138638
Hauptverfasser: Zhu, Ziqi, Xu, Jun, Chen, Xiangzhen, Li, Yongxin, Zhang, Lina, Jia, Lei, Li, Jiaying, Zhu, Taofeng, Zhao, Tongqian
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 15.12.2025
Schlagworte:
Deep learning
Fluorescent detection
Intelligent classification
STM32 single-chip microcomputer
Tetracycline antibiotics
Fluorescent detection
Deep learning
Intelligent classification
STM32 single-chip microcomputer
Tetracycline antibiotics
ISSN:0925-4005
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The excessive use of tetracycline antibiotics has led to an increase in antibiotic resistance among bacteria, complicating the treatment of infections. Additionally, residues of these antibiotics have become environmental pollutants, posing a threat to ecosystems. It is imperative to develop a rapid, point-of-care, intelligent detection method for multiple antibiotics. In this study, a multi-channel sensor based on a clay mineral with inherent blue fluorescence was successfully assembled, denoted as CDs@LAPONITE-Eu-GMP/Cit. This sensor exhibited fluorescence color changes upon detecting tetracycline, oxytetracycline, chlortetracycline, and doxycycline. The difference of coordination ability between different tetracycline antibiotics and Eu3+ was used to distinguish the spectral signals. Furthermore, the deep learning detection model was used to process the fluorescence spectrum data, and the sensitive and selective detection of tetracycline antibiotics was realized by real-time classification and improvement of accuracy. Furthermore, a pre-trained deep-learning classification model was integrated into the STM32 microcontroller to facilitate rapid data processing and real-time reporting. The developed device achieved miniaturization, simplification, operability, and visualization in practical detection scenarios. This work provided a novel approach for rapid identification and on-site intelligent classification of multiple antibiotics in complex environments, inspiring the design and manufacture of advanced devices for multi-target analysis. [Display omitted] •A multi-channel sensor was assembled for the detection of multiple antibiotics.•Combined with deep learning technology, multiple antibiotics can be distinguished.•The detection model achieves detection through a real-time classification process.•A STM32 microcontroller-assisted device was developed for on-site rapid analysis.•The device can be used for detection in environmental water and food samples.
ISSN:0925-4005
DOI:10.1016/j.snb.2025.138638