Grant: CIHR Fellowship

The operating room is a clinical setting in severe need of decision support. Many events in this environment are life-threatening and require immediate attention. An effective monitoring system is required to simplify the information delivered to clinicians, ease the decision-making process in critical care environments and improve safety in the operating room. To ensure safety during anesthesia, anesthesiologists monitor the patient to track changes in heart rate or blood pressure. The monitoring of the carbon dioxide concentration in the exhaled breath has significantly improved the safety of anesthesia. This measurement is called capnography. Under normal conditions, the measured signal has a typical rectangular shape. If the patient enters an abnormal condition this shape can change and indicate a critical event. The clinician can detect such changes and react appropriately to restore normal condition. However, the anesthesiologist must attend to other tasks and is often unable to continuously monitor the capnogram. The triggered alarms of current unintelligent monitoring systems are based on simple thresholds applied on the physiological recordings. This leads to a significant number of false alarms that are so frequent that they represent a nuisance, rather than an improvement. An automated and smarter analysis of the patterns present in the capnogram would facilitate early recognition of important clinical events. The research and development of smart alarms requires collaboration between clinicians and engineers. My research project seeks to develop a smart breath analyzing system to generate intelligent alarms in abnormal clinical situations. The generation of alarms will be designed to be meaningful to the clinicians. A fail-safe decision-support system based on automatic capnogram analysis will improve the way clinicians work. These new procedures will make the clinician's decision making process easier and consequently enhance patient safety.