Clinically validated classification of chronic wounds method with memristor-based cellular neural network.

Area of Science:

• Biomedical Engineering
• Medical Technology
• Computational Medicine

Background:

• Chronic wounds affect approximately 4% of the global population, necessitating advanced monitoring solutions.
• The COVID-19 pandemic highlighted the need for remote and reliable patient monitoring technologies.
• Current wound assessment methods can be subjective and time-consuming.

Purpose of the Study:

• To introduce and evaluate a novel device, the Wound Viewer (WV), for chronic wound assessment.
• To demonstrate the WV's capability in automatically identifying, classifying, and measuring chronic wounds.
• To validate the WV's performance against established clinical scales.

Main Methods:

• Development of a Wound Viewer (WV) device utilizing a memristor-based Discrete-Time Cellular Neural Network (DT-CNN).
• Prospective observational clinical trial involving 150 patients with chronic wounds.
• Implementation of the Wound Bed Preparation (WBP) scale for wound classification (granular tissue, slough, necrosis).

Main Results:

• The WV device demonstrated high reliability in automatically assessing chronic wounds.
• Classification accuracy using the WBP scale reached over 90% in the clinical trial.
• The technology provides essential clinical information for caregivers, improving wound management.

Conclusions:

• The memristor-based DT-CNN technology integrated into the Wound Viewer offers a reliable and accurate solution for chronic wound assessment.
• The WV device has the potential to significantly enhance the monitoring and management of chronic wounds.
• This technology supports objective and efficient wound evaluation, crucial for patient care.