Skip to content
2000
image of MHA-SVR: An Adaptive Soft Sensor Based on Feature Interaction for 
Ultrasonic Phytomedicine Extraction

Abstract

Introduction

Ultrasonic extraction is a crucial technique for isolating active compounds from phytomedicine. However, as a batch process characterized by non-linearity and small sample size, it poses substantial challenges for real-time and prediction of extraction rates during the extraction of phytomedicinal. This work proposes an adaptive soft sensor for ultrasonic phytomedicine extraction.

Methods

An adaptive soft sensor based on an attention mechanism was first proposed. The attention mechanism calculates correlations between samples and assigns weights based on their similarity to the current query. Support vector regression (SVR) is then used to construct the soft sensor for extraction rate measurement. To further enhance sample information analysis, multi-head attention is employed. This allows the model to calculate the similarity between current queries and historical data across different feature spaces, thus improving the modeling capabilities of the intrinsic data structure. Finally, a dual-frequency ultrasonic extraction experiment of puerarin is designed and conducted. The experimental data is collected and labeled from different batches under varying initial extraction temperatures. This data is used to establish the soft sensor model and compare its performance.

Results and Discussion

The experimental results indicate that the proposed MHA-SVR model improved the coefficient of determination (R2) by 5.12% compared to the mainstream model and reduced the online prediction time by 88% compared to the JITL-SVR model. This work performance well exceeds the others while maintaining good real-time capabilities for the dual-frequency ultrasonic extraction of puerarin.

Conclusion

The multi-head attention and SVR-integrated soft sensor method proposed in this study effectively addresses the soft measurement challenges in online monitoring of multi-batch ultrasonic extraction processes. This approach demonstrates significant enhancement in extraction yield detection accuracy across varying batches and diverse initial operating conditions, thereby providing a robust technical solution for real-time quantification of extraction efficiency in botanical material processing.

© 2025 Bentham Science Publishers
Loading

Article metrics loading...

/content/journals/ctmc/10.2174/0115680266356325250510064641
2025-05-20
2025-09-14

  • From This Site

    /content/journals/ctmc/10.2174/0115680266356325250510064641
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. PunF.W. OzerovI.V. ZhavoronkovA. AI-powered therapeutic target discovery.Trends Pharmacol. Sci.202344956157210.1016/j.tips.2023.06.01037479540
     [Google Scholar]
  2. SumathiS. SuganyaK. SwathiK. SudhaB. PoornimaA. VargheseC.A. AswathyR. A review on deep learning-driven drug discovery: Strategies, tools and applications.Curr. Pharm. Des.202329131013102510.2174/138161282966623041208413737055908
     [Google Scholar]
  3. SinghS. KumarR. PayraS. SinghS.K. Artificial Intelligence and Machine Learning in pharmacological research: Bridging the gap between data and drug discovery.Cureus2023158e4435910.7759/cureus.4435937779744
     [Google Scholar]
  4. WangK. HuangY. WangY. YouQ. WangL. Recent advances from computer-aided drug design to artificial intelligence drug design.RSC Med. Chem.202415123978400010.1039/D4MD00522H39493228
     [Google Scholar]
  5. MengF. GuoB. MaY. LiK. NiuF. Puerarin: A review of its mechanisms of action and clinical studies in ophthalmology.Phytomedicine202210715446510.1016/j.phymed.2022.15446536166943
     [Google Scholar]
  6. LvJ. ShiS. ZhangB. XuX. ZhengH. LiY. CuiX. WuH. SongQ. Role of puerarin in pathological cardiac remodeling: A review.Pharmacol. Res.202217810615210.1016/j.phrs.2022.10615235248700
     [Google Scholar]
  7. LigaS. PaulC. Puerarin—A promising flavonoid: Biosynthesis, extraction methods, analytical techniques, and biological effects.Int. J. Mol. Sci.20242510522210.3390/ijms2510522238791264
     [Google Scholar]
  8. WeiH. SunM. WangR. ZengH. ZhaoB. JinS. Puerarin mitigated LPS‐ATP or HG ‐primed endothelial cells damage and diabetes‐associated cardiovascular disease via ROS‐NLRP3 signalling.J. Cell. Mol. Med.20242810e1823910.1111/jcmm.1823938774996
     [Google Scholar]
  9. WanQ. LuQ. LuoS. GuanC. ZhangH. The beneficial health effects of puerarin in the treatment of cardiovascular diseases: From mechanisms to therapeutics.Naunyn Schmiedebergs Arch. Pharmacol.2024397107273729610.1007/s00210‑024‑03142‑338709267
     [Google Scholar]
  10. WeiX. HaoT. Research progress of new technology applied in traditional Chinese medicine extraction.Zhongguo Yesheng Zhiwu Ziyuan20203994750
     [Google Scholar]
  11. Bin MokaizhA.A. NourA.H. KerbouaK. Ultrasonic-assisted extraction to enhance the recovery of bioactive phenolic compounds from Commiphora gileadensis leaves.Ultrason. Sonochem.202410510685210.1016/j.ultsonch.2024.10685238518410
     [Google Scholar]
  12. GaoY. XuM. WanH. LiC. WanY. Determination of isoflavones in Radix puerariae from different origins by ultra-high performance liquid chromatography based on optimal pretreatment method.Foods202312479410.3390/foods1204079436832868
     [Google Scholar]
  13. FengF. GeY. DaiR. XieH. Research progress of ultrasonic-assisted extraction technology.Food Ind.20224304239243
     [Google Scholar]
  14. WangR. SprengelS. BoehmG. BaetsR. AmannM-C. RoelkensG. Application of supercritical fluid extraction technology in traditional Chinese medicine extraction.Science and Innovation201814STh1B.310.1364/CLEO_SI.2018.STh1B.3
     [Google Scholar]
  15. LiuD. Study on the kinetics of ultrasonic extraction.China Sci Technol Inf.20150000033334
     [Google Scholar]
  16. WuE.Y. SunW.J. WangY. ZhangG.Y. XuB.C. ChenX.G. HaoK.Y. HeL.Z. SiH.B. Optimization of ultrasonic-assisted extraction of total flavonoids from abrus cantoniensis (Abriherba) by response surface methodology and evaluation of its anti-inflammatory effect.Molecules2022277203610.3390/molecules2707203635408435
     [Google Scholar]
  17. FengX. LiC. LiP. LiuY. FanB. BaoH. XuT. LiuT. Application of Ultrasonic Extraction in the Extraction of Chemical Components of Traditional Chinese Medicine.J. Chin. Exp. Formulae20142018231234
     [Google Scholar]
  18. GuanY. ChenJ. DongC. Application of virtual sample generation and screening in process parameter optimization of botanical medicinal materials.Curr. Top. Med. Chem.202323861862610.2174/156802662366623011712153136650652
     [Google Scholar]
  19. ReverchonE. De MarcoI. Supercritical fluid extraction and fractionation of natural matter.J. Supercrit. Fluids200638214616610.1016/j.supflu.2006.03.020
     [Google Scholar]
  20. ChenJ. XueZ. Soft sensor based on JITL-SVR for ultrasonic chinese medicine extraction.Proceedings of the 2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)Xiangtan, China, 12-14 May 2023, pp. 869-874.10.1109/DDCLS58216.2023.10166134
     [Google Scholar]
  21. YaoL. GeZ. Online updating soft sensor modeling and industrial application based on selectively integrated moving window approach.IEEE Trans. Instrum. Meas.20176681985199310.1109/TIM.2017.2677622
     [Google Scholar]
  22. ZhaoL. WangR. WangJ. YuT. SuA. Nonlinear state estimation with delayed measurements using data fusion technique and cubature Kalman filter for chemical processes.Chem. Eng. Res. Des.201914150251510.1016/j.cherd.2018.11.020
     [Google Scholar]
  23. ZhangY. YuZ. WuX. New extraction technologies for active components from traditional Chinese medicine and natural plants—microwave-assisted extraction.Zhongguo Zhongyao Zazhi2004021317
     [Google Scholar]
  24. LiuY. ZheW. ZhangR. PengZ. WangY. GaoH. GuoZ. XiaoJ. Ultrasonic-assisted extraction of polyphenolic compounds from Paederia scandens (Lour.) Merr. Using deep eutectic solvent: Optimization, identification, and comparison with traditional methods.Ultrason. Sonochem.20228610600510.1016/j.ultsonch.2022.10600535429898
     [Google Scholar]
  25. XueF. LiC. LiP. LiuY. FanB. BaoH. XuT. LiuT. Application of ultrasonic extraction in the extraction of chemical components from traditional Chinese medicine.J. Chin. Exp. Formula Sci.20142018231234
     [Google Scholar]
  26. YangL. ChenY. ZhangJ. Process Optimization of Dual-Frequency Ultrasound-Assisted Extraction of Pueraria Flavonoids.Preservation and Processing2017065963
     [Google Scholar]
  27. JiangY. YinS. DongJ. KaynakO. A review on soft sensors for monitoring, control, and optimization of industrial processes.IEEE Sens. J.20212111128681288110.1109/JSEN.2020.3033153
     [Google Scholar]
  28. CaoP. LuoX. Modeling for soft sensor systems and parameters updating online.J. Process Contr.201424697599010.1016/j.jprocont.2014.04.020
     [Google Scholar]
  29. QiuK. WangJ. ZhouX. GuoY. WangR. Soft sensor framework based on semisupervised just-in-time relevance vector regression for multiphase batch processes with unlabeled data.Ind. Eng. Chem. Res.20205944196331964210.1021/acs.iecr.0c03806
     [Google Scholar]
  30. GuoX. HanD. XiH. RaoL. LiaoX. HuX. WuJ. Extraction of pectin from navel orange peel assisted by ultra-high pressure, microwave or traditional heating: A comparison.Carbohydr. Polym.201288244144810.1016/j.carbpol.2011.12.02623218318
     [Google Scholar]
  31. YaoL. GeZ. Deep Learning of semisupervised process data with hierarchical extreme learning machine and soft sensor application.IEEE Trans. Ind. Electron.20186521490149810.1109/TIE.2017.2733448
     [Google Scholar]
  32. WangJ. QiuK. WangR. ZhouX. GuoY. Development of soft sensor based on sequential kernel fuzzy partitioning and just-in-time relevance vector machine for multiphase batch processes.IEEE Trans. Instrum. Meas.20217011010.1109/TIM.2021.3123218
     [Google Scholar]
  33. RenJ. MaS.Q. Time series local weighted adaptive kernel PLS soft measurement modeling and its application.J. Instrum. Meas.2018117
     [Google Scholar]
  34. YaoL. GeZ. Dynamic features incorporated locally weighted deep learning model for soft sensor development.IEEE Trans. Instrum. Meas.20217011110.1109/TIM.2021.3073702
     [Google Scholar]
  35. YuanX. LiL. WangY. YangC. GuiW. Deep learning for quality prediction of nonlinear dynamic processes with variable attention‐based long short‐term memory network.Can. J. Chem. Eng.20209861377138910.1002/cjce.23665
     [Google Scholar]
  36. LiuH.B. LiX.Y. YangC. Based on PCA dimensionality reduction model of papermaking wastewater treatment process soft measurement modeling.China Pap. J.201833045360
     [Google Scholar]
  37. RendallR. ChiangL.H. ReisM.S. Data-driven methods for batch data analysis – A critical overview and mapping on the complexity scale.Comput. Chem. Eng.201912411310.1016/j.compchemeng.2019.01.014
     [Google Scholar]
  38. LiS. ZhengY. LiS. HuangM. Data-driven modeling and operation optimization with inherent feature extraction for complex industrial processes.IEEE Trans. Autom. Sci. Eng.20242121092110610.1109/TASE.2023.3259165
     [Google Scholar]
  39. YanW. TangD. LinY. A data-driven soft sensor modeling method based on Deep Learning and its application.IEEE Trans. Ind. Electron.20176454237424510.1109/TIE.2016.2622668
     [Google Scholar]
  40. GuoF. WeiB. HuangB. A just-in-time modeling approach for multimode soft sensor based on Gaussian mixture variational autoencoder.Comput. Chem. Eng.2021146107230, 107230-10723810.1016/j.compchemeng.2021.107230
     [Google Scholar]
  41. YuanX. GeZ. HuangB. SongZ. WangY. Semisupervised JITL framework for nonlinear industrial soft sensing based on locally semisupervised weighted PCR.IEEE Trans. Industr. Inform.201713253254110.1109/TII.2016.2610839
     [Google Scholar]
  42. LuoL. BaoS. MaoJ. TangD. Quality prediction and quality-relevant monitoring with multilinear PLS for batch processes.Chemom. Intell. Lab. Syst.201615092210.1016/j.chemolab.2015.11.004
     [Google Scholar]
  43. GopakumarV. TiwariS. RahmanI. A deep learning based data driven soft sensor for bioprocesses.Biochem. Eng. J.2018136283910.1016/j.bej.2018.04.015
     [Google Scholar]
  44. YuanX. GeZ. SongZ. Soft sensor model development in multiphase/multimode processes based on Gaussian mixture regression.Chemom. Intell. Lab. Syst.20141389710910.1016/j.chemolab.2014.07.013
     [Google Scholar]
  45. LuiC.F. LiuY. XieM. A supervised bidirectional long short-term memory network for data-driven dynamic soft sensor modeling.IEEE Trans. Instrum. Meas.20227111310.1109/TIM.2022.3152856
     [Google Scholar]
  46. XuB. XiongW. YaoL. Chaos least squares support vector machine and its application on fermentation process modeling.CIESC J.2013641245854591
     [Google Scholar]
  47. VaswaniA. ShazeerN. ParmarN. Attention is All You Need.Adv. Neural Inf. Process. Syst.201730
     [Google Scholar]
  48. YinH. ChenJ. QiX. LiuY. Soft measurement of extraction rate of traditional Chinese medicine based on PCA-BP.Comput. Appl. Chem.2008257885888
     [Google Scholar]
  49. LiangT. LiC. ZhangJ. LiuH. Soft measurement of Chinese medicine extraction concentration by support vector machine.Comput. Appl. Chem.2013301113711374
     [Google Scholar]
  50. LvB. ChenJ. LiuB. DongC. AO-BBO: A novel optimization algorithm and its application in plant drug extraction.Curr. Top. Med. Chem.201919213914510.2174/156802661966618113014070930499411
     [Google Scholar]
  51. YaoL. GeZ. Moving window adaptive soft sensor for state shifting process based on weighted supervised latent factor analysis.Control Eng. Pract.201761728010.1016/j.conengprac.2017.02.002
     [Google Scholar]
  52. YamadaN. KanekoH. Adaptive soft sensor ensemble for selecting both process variables and dynamics for multiple process states.Chemom. Intell. Lab. Syst.2021219104443, 104443-10444610.1016/j.chemolab.2021.104443
     [Google Scholar]
  53. YamadaN. KanekoH. Adaptive soft sensor based on transfer learning and ensemble learning for multiple process states.Anal. Sci. Adv.202235-620521110.1002/ansa.20220001338716124
     [Google Scholar]
  54. TulsyanA. KhodabandehlouH. WangT. Spectroscopic models for real-time monitoring of cell culture processes using spatiotemporal just-in-time gaussian processes.AIChE J2021675e1721010.1002/aic.17210
     [Google Scholar]
  55. ZhaoC.H. WangF.L. YaoY. Time period-based statistical modeling, online monitoring, and quality prediction for intermittent processes.Acta Autom. Sin.201036336637410.3724/SP.J.1004.2010.00366
     [Google Scholar]
  56. ZhengJ. ShenF. YeL. Improved mahalanobis distance based JITL-LSTM soft sensor for multiphase batch processes.IEEE Access20219721727218210.1109/ACCESS.2021.3079184
     [Google Scholar]
  57. ShenF. ZhengJ. YeL. MaX. LSTM soft sensor development of batch processes with multivariate trajectory-based ensemble just-in-time learning.IEEE Access20208738557386410.1109/ACCESS.2020.2988668
     [Google Scholar]
  58. ShaoW. GeZ. LiH. SongJ. A semi-supervised dynamic soft measurement modeling approach based on recurrent neural networks.J. Electron. Meas. Instrum.201933713
     [Google Scholar]
  59. LawS.W. WongC.L. WongY. Multi-head CNN-LSTM soft measurement modeling based on attention mechanism.Control Eng.20222918211828
     [Google Scholar]
  60. LuoS.H. WangZ.L. WangX. Based on attention mechanism multi-head CNN-LSTM soft measurement modeling.Control Eng.2022291018211828
     [Google Scholar]
  61. ZuoG. LuoJ. WangN. LianY. HeX. Two-stage variational mode decomposition and support vector regression for streamflow forecasting.Hydrol. Earth Syst. Sci.202024115491551810.5194/hess‑24‑5491‑2020
     [Google Scholar]
  62. AmatriainX. SankarA. BingJ. BodigutlaP.K. HazenT.J. KaziM. Transformer models: An introduction and catalog.arXiv:2302.077302023
     [Google Scholar]
  63. YeC.L. Research progress in new technologies for extraction and separation of Chinese materia medica.Chin. Tradit. Herbal Drugs201546457464
     [Google Scholar]
  64. AlsadiJ. ObaidatM.T. IsmailR. TrradI. AljamalM. DahimM. AbuaddousM. KhodierM. HatamlehR. Al-MattarnehH. Box–behnken design for polycarbonate-pigment blending: Applications and characterization techniques.Polymers20221422486010.3390/polym1422486036432986
     [Google Scholar]
  65. HaqueS.K.M. Optimized Box–Behnken experimental design based response surface methodology and Youden’s robustness test to develop and validate methods to determine nateglinide using kinetic spectrophotometry.Spectrochim. Acta A Mol. Biomol. Spectrosc.202226812071210.1016/j.saa.2021.12071234896681
     [Google Scholar]
  66. KeshtegarB. HeddamS. SebbarA. ZhuS.P. TrungN.T. SVR-RSM: A hybrid heuristic method for modeling monthly pan evaporation.Environ. Sci. Pollut. Res. Int.20192635358073582610.1007/s11356‑019‑06596‑831705408
     [Google Scholar]
  67. ZambomA.Z. DiasR. A review of kernel density estimation with applications to econometrics.Int. Econ. Rev.2013512042
     [Google Scholar]
  68. ChiccoD. WarrensM.J. JurmanG. The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation.PeerJ Comput. Sci.20217e62310.7717/peerj‑cs.62334307865
     [Google Scholar]
  69. de MyttenaereA. GoldenB. Le GrandB. RossiF. Mean absolute percentage error for regression models.Neurocomputing2016192384810.1016/j.neucom.2015.12.114
     [Google Scholar]
  70. ChaiT. DraxlerR.R. Root mean square error (RMSE) or mean absolute error (MAE)? – Arguments against avoiding RMSE in the literature.Geosci. Model Dev.2014731247125010.5194/gmd‑7‑1247‑2014
     [Google Scholar]
  71. TulsyanA. WangT. SchornerG. KhodabandehlouH. CoufalM. UndeyC. Automatic real‐time calibration, assessment, and maintenance of generic Raman models for online monitoring of cell culture processes.Biotechnol. Bioeng.2020117240641610.1002/bit.2720531631322
     [Google Scholar]
  72. TokudaI.T. HoangH. KawatoM. New insights into olivo-cerebellar circuits for learning from a small training sample.Curr. Opin. Neurobiol.201746586710.1016/j.conb.2017.07.01028841437
     [Google Scholar]
  73. DoughertyE.R. DaltonL.A. AlexanderF.J. Small data is the problem.2015 49th Asilomar Conference on Signals, Systems and ComputersPacific Grove, CA, USA, 2015, pp. 418-422.10.1109/ACSSC.2015.7421161
     [Google Scholar]
/content/journals/ctmc/10.2174/0115680266356325250510064641
Loading
MHA-SVR: An Adaptive Soft Sensor Based on Feature Interaction for 
Ultrasonic Phytomedicine Extraction
Bentham Science Publishers ; https://doi.org/10.2174/0115680266356325250510064641
/content/journals/ctmc/10.2174/0115680266356325250510064641
/content/journals/ctmc/10.2174/0115680266356325250510064641
Loading

Data & Media loading...


  • Article Type:     Research Article
Keywords: Support vector regression ; Soft sensor ; Batch process ; Phytomedicine ; Multi-head attention ; Ultrasonic extraction

Most Cited Most Cited RSS feed

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error
Please enter a valid_number test