Adaptive Learning and AI to Support Medication Management.

Artificial Intelligence (AI) has proven to be very helpful in different areas, and its application in the medical field can help clinicians to make more informed decisions, especially in the Intensive Care Unit (ICU), where the COVID-19 pandemic exposed several challenges. If a model is correctly selected, AI can help improve healthcare systems by predicting future occurrences. This study analyzes Machine Learning (ML) techniques, discusses their application in an ICU environment and proposes a methodology to develop ML models to predict mean arterial pressure (MAP) values that can assist healthcare professionals’ decision making. The current study starts by doing a brief review of ML methods and identifying interesting models to train. The chosen models are then trained and the one considered to be the most appropriate is chosen to be tested in a controlled environment, in a system that can generate medical data. Data is then cleaned, classified, and fed to the model, to give a prediction related to future MAP values. Finally, the developed system is integrated into the ICU4Covid project, more precisely in the B-Health IoT Box, which allows for deployments of new applications for ICUs. The proposed system's integration into ICU4Covid enables remote medical operations and can improve the quality care of healthcare services in ICU environments.