Enhanced knowledge about the mechanisms of barotrauma and oxygen toxicity and the introduction of non-invasive patient monitoring facilities helped in the development of patient-tailored strategies of mechanical ventilation thus improving survival rates and morbidity of neonatal intensive care. However, the increased demands on the medical skill and the information overload arising from the many continuously assessed physiologic variables may cause patient management problems at neonatal intensive care units (NICUs). Medical knowledge-based systems may help to organize knowledge, to structure information, and to help in decision-making.
We are therefore developing an open-loop, real-time constrained system for optimizing the mechanical ventilation of newborn infants, called VIE-VENT. It uses quantitative on-line (like ventilator settings, transcutaneous blood gas measurements) and qualitative off-line input (like chest wall expansion, spontaneous breathing effort), and incorporates alarming, monitoring, and therapy planning tasks. Therapy recommendations, based on transcutaneously and invasively determined blood gas measurements, are formulated in terms of recommended changes of the ventilator settings aiming to improve ventilation and oxygenation. The amount of a recommended change depends on various parameters such as the degree of blood gas abnormality, the course of the respiratory disease, and the chosen ventilation strategy (e.g. aggressive or slow weaning). A change of the ventilator's settings is evaluated by monitoring the trend of the subsequent changes of the transcutanous blood gases. A new recommendation is formulated if the short-term trend does not meet present requirements concerning the direction and the amount of the expected change.
The system should support the decision-making of junior and senior
neonatologists by providing comprehensive monitoring data analysis and
understandable therapy recommendations.
VIE-VENT is currently implemented and evaluated at two different
The Project Team
The project is a joint cooperation of the
Austrian Research Institute for
Artificial Intelligence (OFAI), the
Department of Medical Cybernetics and
Artificial Intelligence (IMKAI), the
Neonatal Intensive Care Unit (NICU) of the Department of Pedriatics of
the University of Vienna, and the
Department of Pediatrics, Hospital of Mödling, Austria: