Using Computer Vision and Artificial Intelligence to Track the Healing of Severe Burns

Area of Science:

• Medical technology
• Artificial Intelligence
• Computer Vision

Background:

• Accurate burn severity assessment is crucial for effective treatment, particularly differentiating superficial from deep partial-thickness burns.
• Continuous wound healing monitoring in subacute care is essential to prevent complications.
• Existing methods for burn assessment and tracking can be resource-intensive and may lack continuous monitoring capabilities.

Purpose of the Study:

• To develop and evaluate an Artificial Intelligence (AI) and Computer Vision (CV) based pipeline for classifying burn severity and tracking wound healing.
• To create accessible, low-cost tools for both clinical and remote patient use.
• To demonstrate the feasibility of using image-guided therapy for wound parameter assessment.

Main Methods:

• Development of the AI-CV-based Skin Abnormality Tracking Algorithm pipeline.
• Utilizing high-quality 2D color images for wound assessment.
• Tracking key wound parameters including 2D spatial dimensions and color changes indicative of healing or complications.

Main Results:

• The AI-CV pipeline successfully classified burn severity and tracked wound healing.
• Demonstrated the ability to characterize physiological changes within the wound through image analysis.
• Successfully tracked a single burn for 8 weeks (6 weeks clinic, 2 weeks home monitoring).

Conclusions:

• AI and CV present a promising approach for objective and accessible burn care management.
• The developed algorithm can aid clinicians in assessing burn severity and monitoring healing progress.
• This technology supports continuous wound tracking, potentially improving patient outcomes through early complication detection and remote care.