Advancing Wound Care: AI-Powered OCT for Objective Healing Assessment
Monitoring wound healing presents challenges due to the invasiveness of biopsies and the cost and availability of advanced medical imaging. Current methods often rely on visual inspection or basic size measurements.
Biomedical engineers at Duke University, in collaboration with Nokia Bell Labs, have developed a solution.
They utilize a custom-built optical coherence tomography (OCT) imaging system paired with artificial intelligence (AI) models to accurately and objectively measure wound healing progression.
This research also demonstrated that a hydrogel designed to improve wound healing functions more effectively when possessing stiffer mechanical properties. The findings were published online on March 20 in the journal Cell Biomaterials.
The Technology Behind the Breakthrough
OCT, commonly used in eye care for 3D retinal imaging, was adapted to non-invasively visualize tissue architecture and blood flow beneath the skin in wounds. The AI-driven analytical methods, developed by Nokia Bell Labs and trained on data from the Gerecht lab, are crucial for interpreting the extensive data generated by the OCT system. This allows for quantitative tracking of tissue structure and vascular dynamics, enabling objective assessment of healing.
Key Findings: Stiffer Hydrogels Accelerate Healing
To evaluate the technology, researchers applied it to mouse wounds treated with hydrogels from the Gerecht lab, comparing soft and stiff versions. Over two weeks, the platform provided detailed insights, indicating that the stiffer hydrogel facilitated faster formation of granulation tissue and a more rapid transition to regenerated tissue.
This research also demonstrated that a hydrogel designed to improve wound healing functions more effectively when possessing stiffer mechanical properties.
Future Prospects and Clinical Applications
The research collaboration intends to continue developing this platform for potential clinical applications. Future work includes expanding the system's predictive capabilities for various disease states, such as chronic wounds in diabetic patients.
Research Funding
Funding for this research was provided by multiple sources, including the P30 Cancer Center Support Grant, the American Heart Association, the Duke Regeneration Center, Duke Science and Technology, and Nokia Bell Labs.