Implementación de un Sistema Inteligente semiautomático para la asistencia de pacientes en unidades de cuidados críticos

Abstract

Hemodynamic support of patients in cardiac anesthesia (resuscitation) intensive care units (ICU) is complex due to the number of variables provided by the various devices with little integration, allowing the clinician to assess the needs for treatment variation. The assessment of all available clinical parameters and the execution of the necessary modifications in intravenous drug infusions is a time-consuming effort that does not add value to patient care. The integration of data, facilitating decision making and helping to streamline treatment variations can optimize work to achieve more accurate and continuous patient target ranges. The use of clinical decision support systems (CDSS) to assist clinicians can help improve the quality and efficiency of care. This paper describes the development and implementation of a CDSS called IOSC3 (Intelligent Ontology-based System in Critical Care), for the management of cardiac intensive care unit patients. This CDSS implemented an expert knowledge-based system to provide real-time therapeutic advice based on continuous monitoring of patients' cardiovascular vitals. When the proposal is evaluated by the clinician, its appropriateness is determined and accepted, the system acts semi-automatically, controlling the drug infusion pumps and modifying the number of drugs delivered to the patient. IOSC3 has been tested in real time on patients in the ICU of the Complejo Hospitalario Universitario de Vigo. The IOSC3 system was integrated and accepted by the ICU staff as an aid to decision making, since the proposed dosage recommendations are accepted in 90% of the cases. It is seen as a useful tool for their daily work. Further research will be needed in different clinical scenarios to see if the IOSC3 system represents more advantageous endpoints in aspects such as total dose administration, shorter stays or mortality.