ELISE aimed at developing a digital clinical decision support system (CDSS) for the paediatric intensive care unit to optimize diagnostic and therapeutic routine processes. Especially the early identification and anticipation of life-threatening diagnoses are crucial because the paediatric intensive care setting is a complex knowledge- and experience-based area that continuously challenges healthcare professionals. All diagnostic and therapeutic measures are shaped by highly individual variations due to the age-specific development stages of children and the heterogeneous, partly seldom, diseases within this patient population. To address these challenges, we develop ELISE, which can support clinicians in the early identification of Systemic Inflammatory Response Syndrome (SIRS), sepsis, and associated organ dysfunctions (i.e., hepatic/ hematologic/ respiratory/ renal/ cardiovascular organ dysfunction).

The ELISE CDSS consists of multiple, target-condition-specific knowledge-based detection and data-driven prediction models. These models are developed and evaluated for their diagnostic accuracy (i.e., measured by sensitivity and specificity) before they are validated. Upon achieving an acceptably high level of accuracy to detect and/or predict the presence of diagnostic events, these models may be integrated in a routine application of the CDSS used in the paediatric intensive care.