JMIR Publications

ABSTRACT

Medical applications of mathematical modeling, including Machine Learning (ML) models, Knowledge Graphs (KG), and Digital Twins (DT) - primarily involve the prediction of individual patient outcomes. However, to date, the systematic evaluation of these models' contributions to healthcare quality has been explored only to a limited extent. This perspective article systematically examines how mathematical modeling can be integrated into health care quality management. It begins with a comprehensive overview of quality assessment in health care, addressing the definitions of procedures, patient outcomes and Quality Metrics with a quantitative focus. The emphasis is subsequently placed on three categories of Patient-Centered Quality of Care (PCQC), namely, patient safety, procedure accuracy and procedure efficacy, which provide both conceptual and mathematical descriptions. Different levels of modeling tasks essential for managing PCQC are identified. This quantitative concept provides the foundation for the subsequent implementation of ML models. This article facilitates a deeper understanding of the topic by assigning relevant publications to these three quality categories. Focus is placed on the applicability of KGs and DTs to improve quality management in healthcare.