Explainable Artificial Intelligence for Early Prediction of Pressure Injury Risk

Required reading for all learners: Implicit Bias impacts patient outcomes

Pressure injury, an area of damage to the skin, underlying tissue, or both that is caused by pressure or pressure combined with shear, occurs during hospitalization in 4% to 6% of critical care patients. Crucial to the prevention of HAPIs is the exercise of clinical judgment by nurses, who must decide which interventions are most appropriate for each individual patient and when interventions should be implemented. Thus, early and accurate HAPI risk assessment is important for guiding clinical choices and enabling the initiation of prompt preventive measures. However, traditional HAPI risk–assessment methods such as the Braden Scale do not incorporate key risk factors unique to critical care patients, such as oxygenation and perfusion. Artificial intelligence (AI) models can process and analyze extensive amounts of data to identify complex relationships among variables, including critical care–specific factors such as oxygenation and perfusion.