AI Chat Paper
Note: Please note that the following content is generated by AMiner AI. SciOpen does not take any responsibility related to this content.
{{lang === 'zh_CN' ? '文章概述' : 'Summary'}}
{{lang === 'en_US' ? '中' : 'Eng'}}
Chat more with AI
PDF (37.6 MB)
Collect
Submit Manuscript AI Chat Paper
Show Outline
Outline
Show full outline
Hide outline
Outline
Show full outline
Hide outline
Research Article | Open Access

Machine learning-guided one-step fabrication of targeted emodin liposomes via novel micromixer for ulcerative colitis therapy

Xinkun Chen1,§Yuli Pan2,§Tao Tang1Jing Fu2Xueye Chen1 ( )Cheng Bao2 ( )
School of Transportation, Shandong Province Key Laboratory of High Performance Hard Alloys and Precision Tools, Ludong University, Yantai 264025, China
School of Life Sciences, Ludong University, Yantai 264025, China

§ Xinkun Chen and Yuli Pan contributed equally to this work.

Show Author Information

Abstract

This study demonstrates a one-step synthesis strategy for fabricating aptamer-conjugated emodin liposomes (Apt-EMO@Lip) through the integration of microfluidic technology and machine learning. We developed a novel vein groove and horseshoe-shaped micromixer (VGHM) that synergistically combines biomimetic vein-groove microstructures (VGM) with horseshoe-shaped splitting-confluence channels (HSM), achieving exceptional mixing efficiency (99.93% at outlet). Both blank liposomes (Blank@Lip) and Apt-EMO@Lip prepared via VGHM display monodisperse size distributions with narrow polydispersity indices, along with superior in vitro stability and biocompatibility. Systematic investigation of microfluidic parameters reveals that flow rate ratio (FRR) and solvent selection critically influence liposomal characteristics, while total flow rate (TFR) shows negligible impact on nanoparticle synthesis. Compared with conventional thin-film hydration methods, the VGHM approach reduces liposome preparation time by 95% while maintaining equivalent physicochemical properties, significantly lowering production costs and establishing a more efficient platform for nanocarrier fabrication. Innovatively, we developed a CNN-LSTM-Attention multivariate regression model incorporating a Newton-Raphson-based optimization (NRBO) algorithm, achieving superior predictive accuracy for liposome size (R² = 0.9574, root-mean-square error (RMSE) = 6.52 nm). This machine learning framework provides an intelligent parameter optimization tool for nanomedicine development. In vitro experiments demonstrated that emodin-loaded liposomes (EMO@Lip) exhibited sustained-release properties (58.62% cumulative release over 48 h) and effectively suppressed lipopolysaccharide (LPS)-induced secretion of pro-inflammatory mediators (NO, tumor necrosis factor-α (TNF-α), IL-6, IL-1β)—in RAW264.7 macrophages. The Caco-2 scratch assay further confirms EMO@Lip's ability to enhance intestinal epithelial barrier repair and comprehensively ameliorate ulcerative colitis pathology. This strategy significantly enhances drug enrichment efficiency at inflammatory sites via aptamer modification, establishing a novel, efficient, and safe nanomedicine delivery platform for precision-targeted ulcerative colitis therapy. The modular design enables high-throughput continuous production, thereby demonstrating exceptional clinical translation potential and industrial scalability.

Graphical Abstract

This work integrates microfluidic technology with machine learning to demonstrate a one-step fabrication of aptamer-conjugated emodin liposomes via a novel micromixer, achieving efficient production, accurate size prediction, and effective therapy for ulcerative colitis.

Electronic Supplementary Material

Download File(s)
7713_ESM.pdf (978.3 KB)

References

【1】
【1】
 
 
Nano Research
Article number: 94907713

{{item.num}}

Comments on this article

Go to comment

< Back to all reports

Review Status: {{reviewData.commendedNum}} Commended , {{reviewData.revisionRequiredNum}} Revision Required , {{reviewData.notCommendedNum}} Not Commended Under Peer Review

Review Comment

Close
Close
Cite this article:
Chen X, Pan Y, Tang T, et al. Machine learning-guided one-step fabrication of targeted emodin liposomes via novel micromixer for ulcerative colitis therapy. Nano Research, 2025, 18(8): 94907713. https://doi.org/10.26599/NR.2025.94907713
Topics:

3145

Views

639

Downloads

21

Crossref

23

Web of Science

26

Scopus

0

CSCD

Received: 08 May 2025
Revised: 19 June 2025
Accepted: 20 June 2025
Published: 25 July 2025
© The Author(s) 2025. Published by Tsinghua University Press.

This is an open access article under the terms of the Creative Commons Attribution 4.0 International License (CC BY 4.0, https://creativecommons.org/licenses/by/4.0/).