Artificial Intelligence-Enhanced Chronic Ulcer Classification and Assessment

The increasing prevalence of challenging-to-heal wounds, often linked with aging populations and the growing occurrence of chronic illnesses, poses significant challenges within the healthcare sector. Ensuring safe and efficient care for the increasing number of patients with chronic ulcers is a complex undertaking. Due to the multidisciplinary nature and the intricate, dynamic processes involved in wound healing, accurately predicting the precise course of wound recovery is a formidable task. In response to these challenges, healthcare teams have started collecting extensive datasets of wound images acquired during clinical visits. The integration of artificial intelligence systems into clinical practice has the potential to assist clinicians in diagnosing conditions, assessing the effectiveness of therapeutic interventions, and predicting healing outcomes.A notable advancement is the emergence of AI-driven remote consultation systems that utilize smartphones and tablets for data collection and seamless connectivity. These systems expedite timely interventions and enhance communication among healthcare professionals.Another significant challenge in the realm of complex wound management and treatment is the continuous monitoring and measurement of various wound indicators. This study emphasizes the importance of the Model for Telemedicine Assessment and envisions the development of an eHealth-supported wound assessment system, underpinned by AI. Leveraging convolutional neural networks (CNNs), a subset of artificial neural networks known for their proficiency in visual imagery analysis holds great promise for categorizing chronic ulcers.The primary objective of this pilot study is to introduce an AI-driven classification methodology that addresses the increasing demands of wound care and management. Specifically, we explore the feasibility of wound segmentation for diabetic foot ulcers and venous leg ulcers by training a CNN on relevant datasets. Furthermore, we conduct a comparative evaluation of various CNN architectures for wound segmentation, identify the area within the wound, calculate its size, and establish a classification system based on convolutional networks.The outcomes of this study evaluate the effectiveness of our proposed approach as a decision support system for the classification of wound images.

کلیدواژه ها:

Artificial Intelligence ، Convolutional Neural Networks ، Chronic Ulcers ، Chronic Wound Classification ، Telecare