Artificial intelligence-assisted design of Chinese herbal medicine based hydrogels

Abstract

The rapid development of artificial intelligence (AI) and the increasing demand for the modernization of Chinese herbal medicine (CHM) have created new opportunities for intelligent biomaterials. This article systematically reviews the latest progress, key challenges, and future directions of AI technology in the intelligent design and development application of natural compounds in traditional Chinese medicine (TCM). It highlights limitations in achieving precise control and targeted delivery in TCM formulations and discusses the advantages of hydrogels as bioactive carriers. The methods based on AI for decoding complex TCM systems, identifying active components, and predicting therapeutic targets were studied. Meanwhile, the progress in the rational design, performance prediction, and release optimization of TCM compound (TCMC) hydrogels was also discussed.

A four-dimensional collaborative design framework integrating disease target prediction, TCMC screening, hydrogel formulation optimization, and active learning of AI was proposed, and its applications in tumor treatment, wound healing, and tissue engineering were discussed. Key challenges, including data quality, model interpretability, and the lack of closed-loop technical systems, are critically evaluated. Future directions such as physics-informed AI and multimodal large models are outlined. This review provides a conceptual and technical foundation for deeper integration of TCM with modern AI and biomaterials science to support the development of precision TCM therapies.

Introduction

Chinese herbal medicine (CHM), a cornerstone of Chinese medical heritage, has accumulated extensive experience in disease prevention and treatment over thousands of years. Its holistic philosophy and syndrome-based therapeutic approach offer notable advantages in managing chronic complex diseases, adjunct cancer therapy, infectious diseases, and functional disorders [1], [2]. However, conventional CHM formulations face major limitations, including insufficient precision, poor targeting, and unclear mechanisms, which restrict their ability to meet modern therapeutic requirements for spatiotemporal control [3], [4]. The multi-component and multi-target nature of CHM further complicates quality control, efficacy assessment, and mechanistic studies. Uncertain identification of active constituents remains a core challenge impeding the modernization and global integration of CHM. Network pharmacology is an emerging interdisciplinary field integrating CHM theory with network science, computer science, polypharmacology, and systems biology. It regards TCM prescriptions, single herbs, and active ingredients as complex systems, and has constructed a “component-target-disease-pathway” network. From a systemic perspective, it clarifies the synergistic mechanism of multiple components, multiple targets, and multiple pathways in TCM, and discovers traditional Chinese medicinal compounds (TCMC) that are of great significance for disease treatment [5], [6].

Hydrogels are polymeric materials with three-dimensional network structures and high water content that mimics the extracellular matrix. They feature tunable physicochemical properties, excellent biocompatibility, and stimulus-responsive drug-release capabilities [7], [8]. Intelligent hydrogels have shown strong potential in controlled drug release, tissue engineering, and regenerative medicine [9]. Integrating active TCMC into intelligent hydrogels enables precise spatial and temporal release, addressing the shortcomings of TCM preparations in terms of targeting and release control [10], [11]. Zhao et al. [12] developed an injectable photocrosslinked composite hydrogel based on oxidized Astragalus polysaccharide and carboxymethyl chitosan/methacrylated sodium alginate with magnesium ions, loaded with curcumin. The incorporated Tribulus terrestris supramolecular assembly enhanced skin repair, and OCS/NX@Cur showed excellent wound-healing performance. Lu et al. [13] developed a (β-D-Pentagalloylglucos)β-D-PGG hydrogel that exhibits excellent antibacterial activity and reduces inflammation by regulating macrophage polarization towards the alternatively activated macrophage phenotype. However, challenges remain in optimizing compatibility between multi-component CHMC and hydrogel matrices, achieving precise release control, and designing intelligent responses for complex diseases. Addressing these issues requires interdisciplinary innovation.

Artificial intelligence (AI) refers to intelligent systems that use machine learning to simulate human cognition, interpret their environment, and take actions to achieve defined goals [14]. AI acts as a bridge between systems-biology-based CHM research and advanced hydrogel materials. It enables the “digital deconstruction” of CHM to identify key active ingredient groups (Q-Markers) and target networks [15], and supports the “intelligent design” of hydrogels tailored to disease microenvironments and delivery requirements. By integrating classical texts and modern data, AI can construct “disease–target” and “drug–component–target” networks to identify critical compounds and targets [6], [16]. For hydrogel design, generative adversarial networks or variational autoencoders can be used to create new hydrogel structures with desired properties [17], [18], [19]. Such integration is expected to overcome bottlenecks in CHM modernization, promoting a shift from “experience-based” to “data-driven” approaches and advancing CHM from population-based to precision therapy. Applications of AI in hydrogel development are summarized in Table 1.

This review aims to systematically examine AI-based technologies for screening active CHM ingredients, elucidating mechanisms, and designing intelligent hydrogels. We propose a collaborative framework integrating “disease–CHM–hydrogel–AI” to support the development of AI-enabled CHM hydrogels for precision medicine.

Continue reading here

Shenggui Lin, Zhaoxuan He, Xinqiao Xia, Wenqian Li, Yanyong Mao, Hao Shi, Artificial intelligence-assisted design of Chinese herbal medicine based hydrogels, Colloids and Surfaces B: Biointerfaces, Volume 265, 2026, 115745, ISSN 0927-7765, https://doi.org/10.1016/j.colsurfb.2026.115745.

Read more interesting articles on Artificial Intelligence here:

• FormulationDE: an updated artificial intelligence system for drug-excipient compatibility prediction
• Application of AI in Tablet Development: An Integrated Machine Learning Framework for Pre-Formulation Property Prediction
• AI-Driven Simulation and Design of Sustained Release Metformin Tablets: Experimental Validation and Predictive Accuracy Assessment