Duke researchers adapt eye imaging technology with AI to monitor deep tissue wound healing in real-time

Duke engineers combine eye-imaging tech with AI to provide a non-invasive, 3D look at how wounds heal beneath the surface, improving hydrogel therapy research.

By: AXL Media

Published: Mar 21, 2026, 6:00 AM EDT

Source: Information for this report was sourced from Duke University

Bypassing the Biopsy Bottleneck

Monitoring the progression of skin wounds has traditionally been a challenge for clinicians, often limited to simple visual inspections or invasive biopsies that can disrupt the healing process. While advanced imaging exists, most devices are too costly or cumbersome for routine monitoring. To solve this, researchers at Duke University have repurposed Optical Coherence Tomography (OCT)—a technology standard in ophthalmology—to provide a 3D, depth-resolved look at how tissue architecture and blood vessels rebuild themselves beneath the surface of a wound.

The Synergistic Role of AI and Optical Scans

While OCT provides high-resolution images of tissue layers, the resulting data is massive and complex. To translate these raw scans into biological insights, Duke partnered with Nokia Bell Labs to develop custom AI models. This OCT-AI platform is trained to automatically quantify vascular dynamics and structural evolution over time. According to Sharon Gerecht, chair of Biomedical Engineering at Duke, this partnership has provided "unprecedented insights" into how different biomaterials induce healing at a microscopic level that is invisible to the naked eye.

Testing Hydrogel Efficacy in Real-Time

The research team utilized the new platform to evaluate the performance of regenerative hydrogels with varying mechanical properties. Over a two-week period, the OCT-AI system monitored granulation tissue—the initial "filler" tissue in a wound—as it matured into regenerated skin. The data revealed that stiffer hydrogels actually accelerated the healing process compared to softer versions, encouraging faster tissue maturation and more efficient vessel remodeling. This finding demonstrates the platform's utility as a powerful tool for both clinical assessment and material science research.