This project investigates current challenges in the safe and effective administration of total intravenous anesthesia, and seeks to deliver engineering contributions to support the work of clinicians and improve patient outcomes. A key issue faced by clinicians and current anesthesia support systems alike is that substantial variability exists in the response of different patients to anaesthetic drugs, which makes the tasks of anaesthesia management rather onerous.

The research spans three interrelated areas: sensing, sysem identification and feedback control.

System identification
We investigate methodologies for real-time identification of pharmacological dose-response models for typical intravenous anesthetic drugs. The identification of pharmacological models presents significant challenges including infrequent and mixed-mode observations, model non-linearity and the presence of non-Gaussian disturbance effects.
Work to date in this space has included the development of real-time Bayesian inference methods for the estimation of patient-specific pharmacokinetic-pharmacodynamic parameters for common hypnotic drug propofol.

Sensing
Sensing work considers the availability of information from intra-procedural signal sources to support patient-specific drug administration. Both existing and experimental data sources are considered. Challenges associated with sensing in anesthesia applications relate to requirements of reliability, frequency, timeliness and non-invasiveness of signals. Work to date has focussed on existing EEG-based measures (e.g., BIS) but we are also investigating the viability of new sensor technology such as ANU-developed low-cost sensors for breath-based drug detection.

Feedback Control
Feedback control work investigates ways to harness available intra-operative information to improve the delivery of anesthetic drugs. In a clinical applications, questions of safety and performance are paramount, and system interpretability is pivotal to its acceptance by clinical end-users. Work to date has explored a variety of options, from human-centred advisory solutions through to more traditional engineering controller design. 

This is a multidisciplinary project with multiple opportunities to contribute at varying levels of research expertise. Please contact the project lead, Dr Nicolo Malagutti, for more information.

Other info:

This project was featured in the National Science Week ANU showcase for 2021. A video of the presentation can be found here.