Duke University uses OCT and AI for new glimpse into wound healing

24 Mar 2026

Custom-built imaging platform accurately measures progress of healing over time.

A team at Duke University working in collaboration with Nokia Bell Labs has gained new insights into how wounds heal.

Monitoring this natural process is traditionally tricky for clinicians, since biopsies are invasive and disrupt the wound site, while most medical imaging devices that could do the job optically are large and complex.

The new study, published in Cell Biomaterials, indicates that an approach combining OCT imaging with AI-based image analysis can accurately and objectively measure the progress of wounds healing over time.

Using this new approach the researchers also showed that a hydrogel under development to improve wound healing works better when it has stiffer, rather than more flexible, mechanical properties. Taken together the results are "a two-for-one boon in a challenging area for both clinicians and researchers," commented the project.

"Wound healing is a complex process, and what we see on the surface doesn’t always reflect what's happening underneath," said Sharon Gerecht from Duke University.

"For more than a decade, my lab has developed hydrogel-based therapies to guide tissue healing and regeneration. Partnering with Nokia Bell Labs allowed us to combine advanced optical imaging and AI, giving us unprecedented insights into how biomaterials induce healing beneath the surface."

The new platform integrates conventional OCT with quantified speckle variance OCT (qSV-OCT), which captures speckle patterns from red blood cells. While OCT provides label-free, micron-scale depth-resolved imaging of tissue microstructure, qSV-OCT records real-time vascular activity by detecting the motion-induced fluctuations in speckle caused by flowing blood cells.

Stiffer hydrogels promote faster healing

The team's OCT data were fed into bespoke AI-driven analytical methods trained on imaging datasets acquired in the Gerecht lab. This analysis automatically quantified how tissue structure and vascular dynamics evolve over time, as well as objectively assessed the degree of healing.

In trials, the project applied its imaging technique to wounds in mice treated with the Gerecht lab's hydrogel, and compared the wound's response to different hydrogels with different mechanical properties.

Over the course of two weeks, the imaging platform provided a detailed inside look at how granulation tissue - the smooth, glassy tissue that initially fills a wound - filled the space and matured, said the project. The data showed that the stiffer hydrogel helped more initial granulation tissue to form in less time, and accelerated the subsequent transition to intact, regenerated tissue.

The next steps will include further development of the platform for potential clinical use, with the team intending to investigate whether it might be able to predict the healing of chronic wounds in diabetic patients. However, while the OCT-AI platform has proved itself in the relatively simple mouse wound scenario, more work is required to move it beyond monitoring healing progress and make it predictive in a variety of disease states.

"With our developmental technology, we were able to monitor the blood flow near the wound and collectively understand the structural and vascular changes that were happening in real-time," said Jiyeon Song, a postdoctoral researcher in Gerecht’s laboratory.

"The AI helped us quantitatively track those changes and get more objective results rather than us trying to manually analyze the images ourselves."