Wound healing is a complex biological process that moves through stages, including clotting, immune response, scabbing, and scarring. For many patients, especially those in remote areas or with limited mobility, effective treatment is difficult to access, and outcomes can be slow or inconsistent. To address this challenge, researchers have now developed a wearable system that uses artificial intelligence (AI) and bioelectronics to optimize healing, showing significantly faster recovery rates in preclinical studies.

A team of engineers at the UC Santa Cruz (Santa Cruz, CA, USA), working alongside researchers at UC Davis (Davis, CA, USA), has created a device called a-Heal. The system integrates camera, AI, and bioelectronics into a single closed-loop system that monitors the wound and applies therapy. The device attaches to a standard bandage, takes images every two hours, and uses a machine learning model to determine the healing stage and whether intervention is needed.

When healing lags, the AI system delivers treatment either through a bioelectronic actuator that applies fluoxetine, a drug shown to reduce inflammation and promote tissue closure, or by administering an electric field that enhances cell migration. The AI model applies reinforcement learning to mimic physician decision-making, adapting drug dosage and electric field strength over time. The device continuously sends data to a secure interface so clinicians can oversee and adjust treatment.

To validate the technology, UC Davis researchers tested it in preclinical wound models. The results, published in npj Biomedical Innovations, showed wounds treated with the system healed about 25% faster than those receiving standard care. The device not only accelerated closure of acute wounds but also showed potential for restarting healing in chronic wounds, which are especially difficult to treat.

The findings demonstrate how AI-driven feedback control and continuous imaging can transform wound therapy. By tailoring treatment to each patient’s unique healing trajectory, the system could improve recovery outcomes and make therapy more accessible. Researchers now plan to explore applications for chronic and infected wounds, as well as refine the reinforcement learning algorithm to enhance accuracy.

“It’s essentially a microscope in a bandage,” said fellow Associate Professor of ECE Mircea Teodorescu. “Individual images say little, but over time, continuous imaging lets AI spot trends, wound healing stages, flag issues, and suggest treatments.”

http://m.allmycbs.com/en/index.asp