Accurate AI-based Characterisation of Surface Size, Depth and Tissue Composition in Hard-to-Heal Wounds

Difficult to Heal Wounds

VLU; difficult to heal wounds; diabete; ulcers;

This study aims to determine and evaluate the clinical accuracy, precision, and safety of SeeWound 2, an AI-driven wound assessment application, designed for the measurement of wound surface area (cm²), wound depth (mm), and the estimation of the proportion of fibrin covering (slough) and necrosis (%) in real-world clinical settings for patients with hard-to-heal wounds. The study also seeks to validate the non-invasive method for measuring wound depth, as current standard care involves invasive probing of the wound to estimate depth - a practice that this investigational device is intended to replace with a digital, contact-free measurement approach.

SeeWound 2 is a software-based medical device that utilises artificial intelligence to classify and quantify wound tissue types, specifically fibrin covering (slough) and necrosis, as well as to measure wound surface area and depth through digital image analysis. The system operates as a mobile camera-based application, whereby healthcare professionals capture an image of a hard-to-heal wound. The software then automatically analyses the image using integrated AI models in combination with the LiDAR sensor technology embedded in the mobile camera hardware. The product's capability to automatically measure wound surface area, estimate wound depth in a non-invasive manner, and objectively quantify the proportion of slough and necrosis within the wound bed represents a novel functionality not currently available in clinical practice.

No related papers found yet.

None listed

Sweden