Hanyang University Researchers Develop AI-designed Shape-Shifting Microneedles for Diabetic Wound Healing

SEOUL, South Korea, June 26, 2026 /PRNewswire/ -- Chronic diabetic wounds heal slowly and are highly vulnerable to infection. Researchers have developed an AI-guided, 4D-printed microneedle patch that actively closes wounds, fights bacteria, and promotes tissue repair. Inspired by the carnivorous plant Drosera capensis, the device combines shape-changing materials, adhesive DNA nanoparticles, and antibacterial zinc treatment, demonstrating how AI can transform biological inspiration into programmable medical technology.

Chronic wounds remain a major healthcare challenge, especially for people with diabetes, who often experience delayed healing, persistent inflammation, and a higher risk of infection. Traditional wound-closure methods such as sutures, staples, and adhesives can help bring wound edges together, but they do not actively respond to the body's healing process. Scientists are therefore exploring biomaterials that can adapt to biological environments while promoting tissue repair and preventing infection.

Associate Professor Hyun-Do Jung from Hanyang University and his team, based in Korea, have now developed an artificial intelligence (AI)-guided microneedle patch that actively changes shape at physiological temperature (37°C) to help close wounds while delivering regenerative therapy and antibacterial protection. The study combines AI, 4D printing, biomimicry, DNA nanotechnology, and surface engineering in a single wound-healing platform.

This paper was made available online on 30 March 2026 and was published in the journal Advanced Materials.

The study was inspired by Drosera capensis, a carnivorous plant that captures prey through coordinated movement, adhesion, and protective mechanisms. Drawing on these features, the researchers designed a shape-memory microneedle system capable of actively bending after placement in tissue. The microneedles were fabricated using 4D printing, which enables structures to change shape in response to environmental stimuli.

The team used machine-learning models to predict and optimize the shape-recovery behavior of the printed materials, reducing the need for extensive trial-and-error experimentation. By analyzing how material composition and manufacturing conditions affected performance, the researchers identified an optimal fabrication window that balanced mechanical stability with rapid shape recovery. Among the machine-learning approaches evaluated, Gaussian Process Regression provided the most accurate predictions and the most reliable uncertainty estimates.

"This study goes beyond conventional biomimicry by using artificial intelligence to translate nature-inspired principles into a functional biomedical device. The key point of this research is not only that it is inspired by nature, but that AI helps convert biological inspiration into a predictable, programmable, and clinically relevant wound-healing technology," said Dr. Jung.

Laboratory experiments showed that the microneedles rapidly recovered their programmed curved shape at body temperature, helping close wounds and maintain stable tissue contact. The platform also incorporated adhesive DNA nanoparticles designed to support tissue regeneration and a zinc-treated surface that provided antibacterial protection. Tests demonstrated sustained DNA release, favorable responses from endothelial cells and fibroblasts involved in wound healing, and strong antibacterial activity against both Escherichia coli and Staphylococcus aureus.

In preclinical wound-healing experiments, the integrated system accelerated wound closure and improved tissue regeneration compared with conventional approaches.

"Beyond wound healing, the AI-guided 4D-printing strategy could also be extended to soft biomedical robots or tissue-interfacing devices that require programmable motion, controlled shape transformation, and stable contact with biological tissues," said Dr. Jung.

While further research is needed before clinical use, the technology could be adapted for smart wound patches, implants, scaffolds, and stents that respond to the body's environment. In the long term, it may support the development of intelligent biomaterials that improve healing and reduce complications.

Reference
Title of original paper: AI–Guided 4D Printing of Carnivorous Plants–Inspired Microneedles for Accelerated Wound Healing
Journal: Advanced Materials
DOI: 10.1002/adma.202523665

About Hanyang University
Website: https://www.hanyang.ac.kr/web/eng

Contact:
Sung-Rae Cho
82-2-2220-0878
[email protected]

SOURCE Hanyang University