Journal of Water Process Engineering
·
2025
·
IF 6.7
Xu Kou,Rui Qiu,Peng Chen,Qinru Wang,Yang Chen,Hongguang Guo
Antibiotic contamination has emerged as a critical environmental issue due to its adverse effects on ecosystems and public health. Rapid and precise antibiotic detection is essential for mitigating environmental risks. This study introduces a novel deep learning framework for analyzing three-dimensional excitation-emission matrix (3D-EEM) fluorescence spectra to achieve efficient and accurate antibiotic identification. Unlike conventional spectral unmixing techniques such as PARAFAC, our deep learning approach eliminates sample size constraints and labor-intensive preprocessing while significantly enhancing feature extraction capabilities. We established a comprehensive fluorescence database comprising 1632 spectral samples from 65 antibiotic groups spanning sulfonamides, fluoroquinolones, and tetracyclines. Through rigorous comparative analysis of five state-of-the-art deep learning models (YOLOv11, YOLOv8, YOLOv5, RTDETR, and Faster R-CNN), we demonstrated that YOLOv11 achieves superior performance with 98.66 % precision, 98.87 % mAP, and an F1-score of 0.9832. Notably, YOLOv11 exhibited exceptional robustness in identifying fluoroquinolones even under conditions of severe spectral overlap. Further validation across varied component datasets confirmed YOLOv11's exceptional versatility, maintaining 98.7 % precision, 98.9 % mAP, and 0.983 F1-score even when tasked with complex multi-component identification challenges. As the pioneering application of YOLO architecture in antibiotic detection, this work successfully overcomes critical technical limitations inherent to traditional spectral analysis methodologies, substantially reducing data processing time and operational complexity. The framework developed herein provides a valuable reference for deep learning-assisted 3D-EEM analysis and represents a significant advancement toward real-time environmental monitoring systems that offer unprecedented efficiency, accuracy, and intelligence for antibiotic detection in aquatic environments.
研究领域:
环境科学
·
生物信息学
·
医学检测
✓ 已验证
产品关联
DOI
高影响因子
SYNTHETIC METALS
·
2014
·
IF 2.222
Fang Liao,Siwei Yang,Xiubing Li,Lijun Yang,Zhihui Xie,Chenshu Hu,Shuai Yan,Tongyan Ren,Zhiduo Liu
Tunable fluorescent nanomaterials are of great significance. In this paper, the poly(o-phenylenediamine) (PoPD) nanospheres were synthesized via chemical oxidation polymerization of o-phenylenediamine (oPD) monomers by (NH 4 ) 2 S 2 O 8 (APS) at room temperature. It is interesting that when this fluorescent nanosphere was doped by F, S and Se via a simple hydrothermal progress, the fluorescence spectrum remarkably changed. The emission peak of F-doped and S-doped PoPD nanosphere was at 411 nm and 530 nm, the spectrum blue shift can be attributed to the electron-withdrawing effect of F and S. On the other side, the emission peak of Se-doped PoPD nanospheres were at 594 nm and can be attributed to the electronic donor effect of Se. These nanomaterials can be used as ultra-sensitive fluorescent detection system (ppb level) for Cu 2+ , Ca 2+ , Hg 2+ and H 2 O 2 . The tunable spectrum and sensing performance of this system is interesting and significant.
研究领域:
材料科学
·
纳米技术
·
分析化学
·
荧光光谱
✓ 已验证
DOI
高影响因子
FOOD RESEARCH INTERNATIONAL
·
2025
·
IF 8
Xinyao Lin,Chunling Yin,Leqian Hu,Liuchuang Zhao,Mengyao Chen,Xia Hua,Zhimin Liu,Peng Li
Fluorescent sensor arrays are becoming a hot topic in many fields because they can simultaneously detect multiple targets in complex research systems. However, most researches have only collected two-dimensional fluorescence spectral data from fluorescent sensor array interacting with target analytes. In contrast, three-dimensional fluorescence spectra can provide richer information than two-dimensional fluorescence spectra. Based on the hypothesis that collecting three-dimensional fluorescence spectra can obtain more abundant information of green tea samples from different regions, which can improve the accuracy and reliability of origin identification. This study aimed to explore the feasibility of using three-dimensional fluorescent sensor array combined with multi-way pattern recognition methods for the origin discrimination of green tea based on the differences in the contents and types of metal ions in green tea. To investigate this, we first designed a fluorescent sensor array based on amino acid-derived carbon dots and examined its ability to recognize common metal ions in green tea. Excitation-emission matrix spectra of green tea extracts from different geographical origins after interaction with the fluorescent sensor array were collected. Several multi-way pattern recognition methods were used to analyze the three-dimensional fluorescent array data of 100 green tea samples from five origins. The overall classification results of green tea from the five geographical origins were satisfactory, with the best prediction accuracy reaching 96.88%. In comparison, multilinear partial least squares discriminant analysis could make full use of the information of three-dimensional fluorescence data. And its correct identification results for green tea were superior to those of unfold partial least squares discriminant analysis. These results sufficiently demonstrated that the fluorescent sensor array integrated with multi-way pattern recognition, has promising potential for tracing the origin of Chinese green tea.
研究领域:
食品科学
·
分析化学
✓ 已验证
产品关联
DOI
PubMed
高影响因子
Advanced Optical Materials
·
2022
·Vol. 10
·期 8
·2101973
·
IF 10.05
Yan Zhao,Shuo-Hui Cao,Jia-Dai Wang,Na Li,Guo-Chun Lin,Yao-Qun Li
Probing light−matter interactions at the nanoscale is essential for elucidation of their fundamentals and construction of novel nano-optical devices. In this work, it is proposed that polarization- and angle-dependent plasmonic synchronous fluorescence spectroscopy could be powerful to resolve UV–vis plasmon mediated molecular couplings, even under vibrational sublevels. Taking studies of polycyclic aromatic hydrocarbons (PAHs, one general class of graphene analogs with emission range of 350–550 nm) as demonstration, a simple but efficient device of PAHs nanolayer (with thickness in tens to hundreds of nanometers) on an Al nano-film is fabricated. By synchronously scanning excitation and emission wavelengths in a high resolution to depict polarization- and angle-dependent spectrographic profiles, the vibrational transitions coupled to the Al plasmons could be identified, while these could not be attained by conventional fluorescence measurement. As a result, the featured molecular vibrational couplings for both surface plasmon- and plasmonic waveguide-mediated PAHs–Al interactions have been successfully observed at room temperature. Even toward crowded and overlapped spectral components containing multiple vibrations, the transitions could be individually distinguished. These efforts will contribute to the further exploitation and construction of efficient light-emitting devices via luminant molecules on a metal surface.
研究领域:
光学
·
纳米科技
·
材料科学
✓ 已验证
DOI
高影响因子
FOOD HYDROCOLLOIDS
·
2024
·
IF 11
Chang Liu,Ning Wang,Lin Li,Dandan Wu,Liqi Wang,Na Zhang,Dianyu Yu
To investigate the effects of microwave plasma treatment on rice bran protein (RBP), various power levels (150, 160, 170, 180, and 190 W) and processing times (1, 3, 5, 7, and 9 min) were examined for their impact on the conformation, physicochemical properties, and functional properties. The results showed that high-energy active particles generated by microwave plasma could oxidize modify the spatial structure of RBP, thus affecting its functional characteristics. The primary structure of RBP, as revealed by SDS-PAGE, remained unchanged. While the content of β -sheet structures initially decreased and then increased, the random coil content first increased and then decreased, the spatial structure first unfolded and then folded again. The intrinsic fluorescence spectrum showed that hydrophobic amino acids buried within the protein's interior were initially exposed but later reburied. The disulfide bond was first broken to sulfhydryl group and then reformed due to excessive oxidation, which promoted the intermolecular aggregation of RBP. Compared with its natural counterpart, the average particle size decreased from 519.6 nm to 186.7 nm at a power of 160 W and a treatment time of 5 min. Consequently, the emulsification activity index and the stability index increased by factors of 2.18 and 1.40, respectively. The foaming capacity index and foam stability index reached maximum values of 73.07% and 43.34%, respectively. These results provide theoretical evidence for the application of microwave plasma technology to modify plant proteins.
研究领域:
食品科学
·
蛋白质化学
·
生物技术
✓ 已验证
产品关联
DOI
高影响因子
危险品化学品经营许可证（带存储）