UGP system: A deep learning-driven platform for automated identification of ultrafine granular powders using chromatographic fingerprinting

Fei Huang; Ya-ling An; Li-jie Zhang; Jia-wei Wang; Ming-jin Zhang; Zhen-wei Li; Xiao-kang Liu; Dai-di Zhang; Qian-liang Zhang; Li-hua Peng; Wei-lin Qiao; De-an Guo

doi:10.1016/j.jpha.2025.101474

Article Contents

Article Navigation > Journal of Pharmaceutical Analysis > 2025 > Accepted Manu

Fei Huang, Ya-ling An, Li-jie Zhang, Jia-wei Wang, Ming-jin Zhang, Zhen-wei Li, Xiao-kang Liu, Dai-di Zhang, Qian-liang Zhang, Li-hua Peng, Wei-lin Qiao, De-an Guo. UGP system: A deep learning-driven platform for automated identification of ultrafine granular powders using chromatographic fingerprinting[J]. Journal of Pharmaceutical Analysis. doi: 10.1016/j.jpha.2025.101474

Citation:

Fei Huang, Ya-ling An, Li-jie Zhang, Jia-wei Wang, Ming-jin Zhang, Zhen-wei Li, Xiao-kang Liu, Dai-di Zhang, Qian-liang Zhang, Li-hua Peng, Wei-lin Qiao, De-an Guo. UGP system: A deep learning-driven platform for automated identification of ultrafine granular powders using chromatographic fingerprinting[J]. Journal of Pharmaceutical Analysis. doi: 10.1016/j.jpha.2025.101474

Citation:

Fei Huang, Ya-ling An, Li-jie Zhang, Jia-wei Wang, Ming-jin Zhang, Zhen-wei Li, Xiao-kang Liu, Dai-di Zhang, Qian-liang Zhang, Li-hua Peng, Wei-lin Qiao, De-an Guo. UGP system: A deep learning-driven platform for automated identification of ultrafine granular powders using chromatographic fingerprinting[J]. Journal of Pharmaceutical Analysis. doi: 10.1016/j.jpha.2025.101474

PDF( 3137 KB)

UGP system: A deep learning-driven platform for automated identification of ultrafine granular powders using chromatographic fingerprinting

doi: 10.1016/j.jpha.2025.101474

a. School of Chinese Materia Medica, Nanjing University of Chinese Medicine, Nanjing, 210023, China;
b. Zhongshan Institute for Drug Discovery, Shanghai Institute of Materia Medica, Chinese Academy of Science, Zhongshan, Guangdong, 528400, China;
c. Zhongshan Zhongzhi Pharmaceutical Group Co., Ltd. Zhongshan, Guangdong, 528437, China;
d. Shanghai Academy of International Standardization for Traditional Chinese Medicine, Shanghai University of Traditional Chinese Medicine, Shanghai, 201203, China

Funds:

This research was financially supported by the National Natural Science Foundation of China (Grant No: 82130111) and Guangdong Basic and Applied Basic Research Foundation (Grant No: 2023A1515111180).

Received Date: Jun. 23, 2025
Accepted Date: Oct. 21, 2025
Rev Recd Date: Oct. 20, 2025
Available Online: Oct. 23, 2025

Abstract

Abstract

This study developed an intelligent identification system for ultrafine granular powder (UGP) by integrating high performance liquid chromatography (HPLC) fingerprinting with deep learning algorithms. A comprehensive HPLC fingerprint database encompassing 530 batches from 53 UGP varieties across 29 botanical families was established using a standardized 60-min, six-wavelength detection protocol (210, 230, 254, 280, 327, and 380 nm). Chromatographic reproducibility was ensured with quality control sample retention time relative standard deviations (RSDs) below 2%. A three-layer one-dimensional convolutional neural network (1D-CNN) was designed with 32, 64, and 128 filters in successive layers for species classification. Data augmentation techniques including noise interference, baseline drift, and retention time shifts (3.5–60 min) expanded the dataset sixfold and enhanced model generalization capabilities. The optimized model achieved excellent performance on test data with 97.62% accuracy, 97.97% precision, and 97.16% recall, demonstrating consistent reproducibility with mean accuracy of 97.2% ± 0.65% across ten independent training runs. External validation using 63 commercial samples yielded 95.24% identification accuracy, confirming practical applicability. The Flask-based web system enables automated workflows from data upload to species identification and is accessible to users without specialized expertise. This work establishes a standardized approach for intelligent authentication of food-medicine homologous Chinese medicinal UGPs, addressing regulatory and consumer requirements for product authenticity and safety in pharmaceutical and functional food industries.
- Ultrafine granular powder,
- Deep learning,
- HPLC-DAD,
- Automatic identification system,
- Database

FullText(HTML)

References(47)

References

[1]	Y. Jing, M. Yan, D. Liu, et al., Research progress on the structural characterization, biological activity and product application of polysaccharides from Crataegus pinnatifida, Int. J. Biol. Macromol. 244 (2023) 125408.
[2]	M. Dong, J. Li, D. Yang, et al., Biosynthesis and Pharmacological Activities of Flavonoids, Triterpene Saponins and Polysaccharides Derived from Astragalus membranaceus, Molecules 28 (2023) 5018.
[3]	D. Szczuka, A. Nowak, M. Zaklos-Szyda, et al., American Ginseng (Panax quinquefolium L.) as a Source of Bioactive Phytochemicals with Pro-Health Properties, Nutrients 11 (2019) 1041.
[4]	Z. Liu, L. Xu, P. Song, et al., Comprehensive Quality Evaluation for Medicinal and Edible Ziziphi Spinosae Semen before and after Rancidity Based on Traditional Sensory, Physicochemical Characteristics, and Volatile Compounds, Foods 11 (2022) 2320.
[5]	N. Li, Q. Zhang, Z. Jia, et al., Volatile oil from alpinia officinarum promotes lung cancer regression in vitro and in vivo, Food Funct. 9 (2018) 4998-5006.
[6]	N. Zhang, Z. He, S. He, et al., Insights into the importance of dietary chrysanthemum flower (Chrysanthemum morifolium cv. Hangju)-wolfberry (Lycium barbarum fruit) combination in antioxidant and anti-inflammatory properties, Food Res. Int. 116 (2019) 810-818.
[7]	Y. Mai, L. Huang, Y. Qiao, et al., The Potential of Siraitia grosvenorii to Promote Bone Regeneration via Modulating Macrophage Polarization: A Network Pharmacology and Experimental Study, Int. J. Mol. Sci. 26 (2025) 3609.
[8]	J.H. Hwang, Y.S. Park, H.S. Kim, et al., Yam-derived exosome-like nanovesicles stimulate osteoblast formation and prevent osteoporosis in mice, J Control Release 355 (2023) 184-198.
[9]	H. Zhi, Z. Li, Y. Deng, et al., Comparative in vivo constituents and pharmacokinetic study in rats after oral administration of ultrafine granular powder and traditional decoction slices of Chinese Salvia, Biomed. Chromatogr. 33 (2019) e4385.
[10]	T. Rao, Y.F. Gong, J.B. Peng, et al., Comparative pharmacokinetic study on three formulations of Astragali Radix by an LC-MS/MS method for determination of formononetin in human plasma, Biomed. Chromatogr. 33 (2019) e4563.
[11]	H. Li, Research on the Status Quo and Development Policy of New Chinese Decoction Pieces in Yunnan Province, M.S. thesis, Kunming Medical University, Kunming, China, 2020.
[12]	K. Shen, L. Yang, H. Sun, et al., Spatial metabolomics boost understanding the molecular targets and pathways of active components of traditional Chinese medicine in situ and in vivo, TrAC. 191 (2025) 118288 .
[13]	J.L. Ren, L. Yang, S. Qiu, et al., Efficacy evaluation, active ingredients, and multitarget exploration of herbal medicine, Trends Endocrinol. Metab. 34 (2023) 146-157.
[14]	H.Y. Aboul-Enein, I. Ali, Macrocyclic antibiotics as effective chiral selectors for enantiomeric resolution by liquid chromatography and capillary electrophoresis, Chromatographia 52 (2000) 679-691.
[15]	H.Y. Aboul-Enein, I. Ali, A Comparative Study of the Enantiomeric Resolution of Several Tetralone Derivatives on Macrocyclic Antibiotic Chiral Stationary Phases Using HPLC under Normal Phase Mode, Arch. Pharm. 334 (2001) 258-260.
[16]	H.Y. Aboul-Enein, I. Ali, Comparison of the chiral resolution of econazole, miconazole, and sulconazole by HPLC using normal-phase amylose CSPs, Fresenius' Journal of Analytical Chemistry 370 (2001) 951-955.
[17]	P.I. Ali, A. Haque, K. Saleem, Separation and Identification of Curcuminoids in Turmeric Powder by HPLC Using Phenyl Column, Anal. Methods 6 (2014) 2526-2536.
[18]	H.Y. Aboul-Enein, I. Ali, G. Gubitz, et al., HPLC enantiomeric resolution of novel aromatase inhibitors on cellulose- and amylose-based chiral stationary phases under reversed phase mode, Chirality 12 (2000) 727-733.
[19]	Y. Ma, Y. Zhang, Y. Zhao, et al., Comparison of the Antioxidant Capacity of Cell Wall-Broken Decoction Pieces and Traditional Decoction Pieces of Astragli Radix Based on HPLC-ABTS Analytical Method, Biomed. Chromatogr. 39 (2025) e6052.
[20]	Q. Zhang, J. Gan, W. Deng, et al., Fluidized drying kinetics and physicochemical properties evaluation of ultrafine granular powder of Angelicae Sinensis Radix. Chinese Traditional and Herbal Drugs, 55(2024) 3300-3309.
[21]	W. Tan, J. Lin, M. Chen, et al., Study on Consistency Evaluation of Different Types of Decoction Pieces from Astragali Radix Based on HPLC Fringerprint, Traditional Chinese Drug Research and Clinical Pharmacology 36 (2025) 610-616.
[22]	Y. Zhen, Y. Zhi, M. Wang, et al., Comparative study on Cistanches herba formula granules and ultrafine granular power based on UPLC fingerprints and multi-components content determination. Chinese Journal of New Drugs 30 (2021) 1227-1235.
[23]	X. Yu, J. Liu, Y Zhang, et al., Fingerprint Analysis of Cell Wall-Broken Panacis Quinquefolii Radix Pieces by Linear Calibration Using Two Reference Substances, Modern Chinese Medicine 23 (2021) 1617-1622.
[24]	X. Cao, J. Shi, J. Gan, et al., Study on HPLC fingerprint of the Ultrafine Granular Powder of Crataegi Fructus. Journal of North Pharmacy 16 (2019) 1-4.
[25]	Z. Luo, W. Deng, Q. Zhang, et al., Quality evaluation of angelica broken wall powder based on QbD concept of traditional Chinese medicine. Zhongguo Zhong Yao Za Zhi 45 (2020) 1105-1113.
[26]	Q. Meng, J. Zhang, X. Li, et al., ASAP-MS combined with mass spectrum similarity and binary code for rapid and intelligent authentication of 78 edible flowers, Food Chem. 436 (2024) 137776.
[27]	R. Ding, L. Yu, C. Wang, et al., Quality assessment of traditional Chinese medicine based on data fusion combined with machine learning: A review, Crit. Rev. Anal. Chem. 54 (2024) 2618-2635.
[28]	Y. Tan, M. Luo, C. Xu, et al., Deep Learning-Assisted Multiplexed Electrochemical Fingerprinting for Chinese Tea Identification, Anal. Chem. 97 (2025) 11016-11025.
[29]	Y. Pan, H. Zhang, Y. Chen, et al., Applications of Hyperspectral Imaging Technology Combined with Machine Learning in Quality Control of Traditional Chinese Medicine from the Perspective of Artificial Intelligence: A Review, Crit. Rev. Anal. Chem. 54 (2024) 2850-2864.
[30]	Y. Li, N. Logan, B. Quinn, et al., Fingerprinting black tea: When spectroscopy meets machine learning a novel workflow for geographical origin identification, Food Chem. 438 (2024) 138029.
[31]	S. Pan, X. Zhang, W. Xu, et al., Rapid On-site identification of geographical origin and storage age of tangerine peel by Near-infrared spectroscopy, Spectrochim Acta A Mol Biomol Spectrosc 271 (2022) 120936.
[32]	X. Wang, F. Li, L. Wei, et al., Rapid and Precise Differentiation and Authentication of Agricultural Products via Deep Learning-Assisted Multiplex SERS Fingerprinting, Anal. Chem. 96 (2024) 4682-4692.
[33]	X. Wu, Z. Zhao, R. Tian, et al., Total synchronous fluorescence spectroscopy coupled with deep learning to rapidly identify the authenticity of sesame oil, Spectrochim Acta A Mol Biomol Spectrosc 244 (2021) 118841.
[34]	L. Bai, Z.T. Zhang, H. Guan, et al., Rapid and accurate quality evaluation of Angelicae Sinensis Radix based on near-infrared spectroscopy and Bayesian optimized LSTM network, Talanta 275 (2024) 126098.
[35]	T. Liu, Z.Y. Fang, Z. Zhang, et al., A comprehensive overview of graph neural network-based approaches to clustering for spatial transcriptomics, Comput. Struct. Biotechnol. J. 23 (2024) 106-128.
[36]	E. Elizar, M.A. Zulkifley, R. Muharar, et al., A Review on Multiscale-Deep-Learning Applications, Sensors (Basel) 22 (2022) 7384.
[37]	H. Hu, Z. Xu, Y. Wei, et al., The Identification of Fritillaria Species Using Hyperspectral Imaging with Enhanced One-Dimensional Convolutional Neural Networks via Attention Mechanism, Foods 12 (2023) 4153.
[38]	M. Huang, Y. Xu, Y. Liao, Image dataset on the Chinese medicinal blossoms for classification through convolutional neural network, Data Brief 39 (2021) 107655 .
[39]	X. Guo, Z. Ran, F. Li, Kinase-inhibitor binding affinity prediction with pretrained graph encoder and language model, Brief Bioinform 26 (2025).
[40]	X. Li, Z. Luo, J. Lv, et al., m5U-HybridNet: Integrating an RNA Foundation Model with CNN Features for Accurate Prediction of 5-Methyluridine Modification Sites, J. Chem. Inf. Model. (2025).
[41]	N. Zhong, Z. Du, NASNet-DTI: accurate drug-target interaction prediction using heterogeneous graphs and node adaptation, Brief Bioinform 26 (2025).
[42]	D. Shajari, A. Bahari, P. Gill, et al., Synthesis and tuning of gold nanorods with surface plasmon resonance, Opt. Mater. 64 (2017) 376-383.
[43]	V.S.K. Yadav, M.J. Al-Marri, M.A.H.S. Saad, et al., Understanding the progress and challenges in the fields of thermo-catalysis and electro-catalysis for the CO2 conversion to fuels, Emergent Mater. 7 (2024) 1-16.
[44]	A. Bahari, S.I. Esmail, A.M. Alattar, Study of Molecular Interactions, Optical, and Structural Properties through the Green Synthesis of Selenium Oxide Nanoparticles from Hibiscus sabdarriffal: Biocompatibility for Biophysics, Journal of Medical Physics 49 (2024). 557-562.
[45]	X. Su, P. Hu, D. Li, et al., Interpretable identification of cancer genes across biological networks via transformer-powered graph representation learning, Nat. Biomed. Eng. 9 (2025) 371-389.
[46]	G. Li, B. Zhao, X. Su, et al., Discovering Consensus Regions for Interpretable Identification of RNA N6-Methyladenosine Modification Sites via Graph Contrastive Clustering, IEEE J Biomed Health Inform 28 (2024) 2362-2372.
[47]	G. Li, Y. Yang, D. Li, et al., A bijective inference network for interpretable identification of RNA N6-methyladenosine modification sites, Pattern Recognit. 164 (2025) 111541.