Development of a web-app using AI for staging and risk prediction of pressure injuries

Abstract

Despite the existence of international guidelines, pressure injury (PI) staging and risk prediction remain a complex and subjective process in clinical practice. Variability staging accuracy, documentation gaps and organizational barriers indicate the need for a digital solution. Artificial intelligence (AI) has demonstrated promising diagnostic performance in PI assessment. However, healthcare professionals’ acceptance and workflow integration affect the successful implementation. The aim of this study was to explore nurses’ experiences, perceived challenges and expectations regarding PI staging and risk assessment, in order to inform the development of an AI-assisted web application for staging and risk prediction. A qualitative study was conducted using semi-structured, face-to-face interviews with 12 nurses working in diverse settings, and data were analysed using thematic analysis. The Technology Acceptance Model was used as the theoretical framework of this study. The study highlighted 7 main themes: clinical complexity and organizational barriers in pressure injury staging, documentation and communication gaps, improvements, requirements, expectations, trust and AI acceptance and workflow integration. The determinants of intention to use which were expressed by the participants were conditional acceptance of AI, emphasizing perceived usefulness, ease of use, transparency, data security, and preservation of clinical authority. According to the findings, such tools should prioritize usability, workflow compatibility, ethical safeguards, and support of clinical judgment. The user-centered design may enhance adoption and contribute to improved staging accuracy, documentation quality, and patient care.