Improving the capnogram processing and alarm generation during anesthesiaContact Person
To ensure safety during anesthesia, anesthesiologists monitor the patient to track changes in heart rate or blood pressure. The monitoring of the carbon dioxide (CO2) concentration in the inhaled and exhaled breath has significantly improved the safety of anesthesia. This measurement is called capnography. The resulting capnogram is an important parameter to survey because it allows for the evaluation of defects in the artificial breathing circuitry, such as a leak in the breathing circuit. The anesthesiologist can detect changes in the capnogram and react appropriately to restore normal condition. However, the anesthesiologist must attend to other tasks and is usually unable to continuously monitor the capnogram. We are developing new methods for analyzing the breathing of the patient with a computerized monitoring system. We seek to develop a smart capnogram analysis system to generate intelligent alarms in abnormal clinical situations.
An expert system for detecting ventilatory events during anesthesiaContact Person
In an operating room, there are many monitors that collect and display information about the function of the patient's body. But it is not possible for a person to do their work and watch both the patient and the monitors, all at the same time. Therefore, it is possible to make a mistake or miss something important, like a change in the patient's breathing.
We are working to develop a computer that has built-in knowledge to help the doctors and nurses to pay attention in the operating room. We have trained the computer to automatically detect important changes in the patient. These changes are combined in a set of rules that help the doctors to make a rapid and correct decision about actions they need to take. The rules include letting the doctor know if there has been a serious problem (such as accidental anesthetic overdose or leaks in the breathing circuit). The rules were developed with the help of expert anesthesiologists. In this study, we are testing the part called eVENT that monitors the patient's breathing.
Knowledge Authoring ToolContact Person
This project is to develop a knowledge authoring tool to be used by the every day anesthesiologist. Rules will be created by the clinician to be used in a real-time decision support inference engine.
In addition the tool will provide memory aids, 'just in time' information and encourage the implementation of clinical guidelines. This project is an essential step in the development of an Intelligent Anesthesia Navigator (IAN). It extends the focus of our team who have already developed methods to extract key features from the mass of clinical physiological data and have developed a real time inference engine. We now need to harvest the expert knowledge that can be used in the expert system. The iKnow software is available for download
. An introductory video
can been viewed on online.
Neuromuscolar Blockade Control
Conventional incremental bolus administration of neuromuscular blocking (NMB) drugs is associated with limitations in intraoperative control, potential delays in recovery, and residual blockade in the postanesthetic period. To overcome such limitations, we developed the Neuromuscular Blockade Advisory System (NMBAS). The NMBAS advises the anesthesiologist on the timing and dose of NMB drugs based on the history of the patient's electromyographic responses. It uses a novel form of modeling that combines model swapping and continuous adaptation to accommodate the patient variation seen with NMB drugs. New methods of handling nonlinearities at the neuromuscular junction to allow application of adaptive control techniques were used. In a clinical trial we have demonstrated that compared to standard practice, NMBAS-guided care was associated with improved NMB quality and higher TOF ratios at the end of surgery, potentially reducing the risk of residual NMB and improving perioperative patient safety. We are developing a closed-loop version of this system.
Closed Loop Control in Anesthesia
Computer controlled, or automatic, drug delivery is the process of administering a therapeutic regime to a patient with computer assistance for calculation of optimal dose and delivery schedules. Computer control can improve drug therapy by reducing drug usage and costs, by permitting health care staff to work more efficiently and to provide better standard of care, by allowing the safe use of drugs that are difficult to administer, and by compensation for human failings with computer strengths, such as unlimited attention span and patience, and capacity for quick, accurate and redundant calculation. Our goal is to develop an automatic control system for anesthesia and to demonstrate its efficacy, safety and benefits in an operating room. Although closed-loop anesthesia has previously been proposed and tested, it has yet to have a significant impact on clinical practice. Recent developments in sensing for anesthesia have opened new possibilities for ‘closing the loop’. Our research will focus on the deployment of new sensors optimized for controlled drug delivery, robust control methodology and extensive clinical validation. In addition, through advanced control techniques, we aim to provide a mathematical guarantee of the safety of patients under anesthesia.
Epileptic Seizure Detection & Prediction
Detection and prediction of seizures using scalp EEG signal analysisContact Person
As one of the most serious brain disorders, epilepsy is associated with recurrent, unprovoked seizures resulting from an abnormal, sudden, synchronous firing of a neural population. The high incidence of this disease among different age groups around the world labels epilepsy as the most common neurological disorder after strokes. Epileptic seizures cause involuntary and temporary disturbances in consciousness and body movements increasing the chance of accidental injury and death. The objective of this project is to develop a real-time algorithm capable of detecting epileptic seizures as well as predicting their onset in advance. Automatic detection of seizures would be helpful since monitoring the patient's electroencephalogram (EEG) for several days is a necessary step in diagnosis and is an expensive and time consuming process. In addition, predicting the onset of seizures may enable clinicians to control seizures. We have already developed a wavelet-based index to discriminate between ictal and interictal periods in scalp EEG signals. This algorithm is based on measuring the rhythmicity and energy of the EEG signal. The preliminary results show the capability of this method in detecting epileptic seizures with high sensitivity and low false detection rate using surface EEG. The next steps are to complete non-linear analysis of the underlying dynamics of EEG time series to find a measure revealing transition among different brain states in patients with epilepsy.
Autonomic-Cardiac Regulation Monitoring
Homeostasis revisited: Autocatalytic loops in the genesis of stress reactivityContact Person
Autonomic-cardiac regulation contains many parts such as the heart, blood vessels, nerve fibers, and the brainstem which harmonically interact with each other. In fact, autonomic-cardiac regulation operates through interactions between the autonomic nervous system and the cardiovascular system. We aim to develop a physiologically inspired mathematical model of autonomic-cardiac regulation capable of describing physiological interactions within the in vivo system. We, then, develop a non-invasive model-based method to monitor autonomic-cardiac regulation based on a computationally efficient system identification technique using routine clinical measurements such as heart rate, blood pressure, and cardiac output.
Children are increasingly given sedation to facilitate non-painful or minimally invasive diagnostic and therapeutic procedures. The intravenous anesthetic agent propofol, which has rapidly titratable and predictable sedative characteristics, has gained popularity for sedation procedures in a number of settings. While propofol has many advantages, rapid administration of a loading dose causes significant respiratory depression by impairment of the chemoreceptor response to carbon dioxide (CO2). Respiratory depression may lead to hypoxemia and the need to provide manual positive pressure ventilation. However, if propofol is administered less rapidly, spontaneous ventilation is maintained as the accumulation of CO2 in the blood from ongoing metabolic processes continues to stimulate the chemoreceptors despite their impaired sensitivity. We are proposing to identify a clinical dosing schedule for the administration of propofol in children that will ensure rapid onset of sedation while maintaining spontaneous ventilation.
We have developed a software tool (iAssist) to assist clinicians as they monitor the physiological data that guide their actions during anesthesia. The system tracks the statistical properties of multiple dynamic physiological processes and identifies new trend patterns. iAssist has been tested in real-time in the operating room environment.
Pain is conveyed to the brain using the nociception system, causing an increase in activity of the autonomic nervous system (ANS). The ANS responds by activating the sympathetic nervous system (SNS), which creates a stress response in the body. Anesthesiologists try to minimize the stress response, by giving patients drugs that stop nociception. Anesthesiologists rely on the patient's vital signs to estimate the level of ANS activation. With this estimate, they can decide whether to give the patient more or less of the drugs. Unfortunately, these vital signs alone are not enough to estimate ANS activity, because they are often affected by other factors and don't always give a good estimate. Anesthesiologists therefore have to interpret these signs in light of the patient's health and medical history, as well as the type of surgery being performed. We are developing a nociception monitor that automatically determines the level of activation of a patient's ANS. It will use custom algorithms that analyze very small, fast changes in the patient's heart rate, called heart rate variability (HRV). Previous research has shown that HRV responds to ANS activation much more predictably than other vital signs. While it is impossible for an anesthesiologist to monitor HRV, computer systems are well suited to the task.
Visual cues to improve change detection in physiological monitoringContact Person
Monitoring can be jeopardized by the inability of a clinician to recognize important changes in the visual display of data throughout the duration of the monitoring task. We hypothesized that the addition of a visual cue imparting contextual information to a physiological display would improve the detection ability and response time of a clinician to a change in a patient variable. We have investigated the addition of a visual cue for a single variable and the addition of four visual cues to display the interactions between groups of physiological variables, implemented using ecological interface design theory, to a monitor display.
Development of Ambulatory Tools for the Assessment of Human Circadian RhythmsContact Person
In industrialized countries, it is estimated that nearly 30% of workers are affected by work on irregular schedules. This type of working environment increases fatigue which raises the risk of severe accidents and has important societal and economic implications. Circadian rhythms have significant health impacts on fatigue and safety in workplaces, and are increasingly being incorporated in medical therapies. Developing suitable technologies to investigate and counteract these risks is fundamental to achieving the parallel objectives of workers health, public safety, and increased productivity. The ultimate goal of this collaboration with D. Boivin at McGill’s Centre for Study and Treatment of Circadian Rhythms is to promote access to multidisciplinary expertise to address central issues related to monitoring and modeling of human circadian physiology.
Pulse oximetery on a cell phoneContact Person
In poorer parts of the world, most preventable anesthesia morbidity and mortality is related to airway and respiratory problems leading to lack of oxygen (hypoxia). A pulse oximeter gives an early warning of hypoxia by monitoring the percentage of hemoglobin in the blood that is oxygenated. An early, rapid and effective response to early signs of hypoxia, detected and displayed by a pulse oximeter, can rescue the patient from the permanent effects of lack of oxygen, such as brain damage or death.
In this project, we propose to develop and evaluate a wireless pulse oximeter that uses a cell phone to analyze the information received from a sensor placed on the finger. The inherent capabilities of a standard cell phone (which are widely available in underdeveloped areas) will be used to intelligently analyze and creatively communicate information from the sensor. The primary goal of this project is to demonstrate the potential for enhanced delivery of information from a pulse oximeter to enhance the safety of anesthesia care throughout the world.
See: World Health Organization Global Pulse Oximetry Project