Faculty of Information Technology

Bush fires cause major damage each year in many areas of the world and the earlier that they can be detected the easier it is to minimize this damage. This paper describes a collective intelligence algorithm that performs localized rather than centralized control of a number of unmanned aerial vehicles (UAV) that can survey complex areas for fires, devoting attention in proportion to the user specified importance of each area. Simulation shows that not only is the algorithm able to perform this action successfully, it is also able to automatically adapt to a simulated malfunction in one of the UAVs.

With recent and ongoing improvements to unmanned aerial vehicle (UAV) endurance and availability, they are in a unique position to provide long term surveillance in risky environments. This paper presents a swarm intelligence algorithm for executing an exhaustive and persistent search of a non-trivial area of interest using a decentralized UAV swarm without long range communication. The algorithm allows for an environment containing arbitrary arrangements of no-fly zones, non-uniform levels of priority and dynamic priority changes in response to target acquisition or external commands. Performance is quantitatively analysed via comparative simulation with another leading algorithm of its class.

With sincere and profound gratitude, I thank my advisor and friend, Professor Tim Hendtlass. Without Tim's, frankly, extraordinary advocacy and support, this thesis would not exist, and I would not be where I am today. As an undergraduate, Tim's childlike wonder and boundless enthusiasm inspired the same from us, his students. Even then, his door was always open and he, without exception, made time to humour my eager but undirected interest in artificial intelligence and postgraduate research. It was during one of these chats, while Tim was off on one of his famous tangents, that the problem which was to become my thesis topic first came up.

Introduction: An ideal patient controlled analgesia (PCA) opioid would have both a rapid onset and a long duration of action, attributes, which are not available in currently existing opioids including morphine, the most widely used agent. A mixture of rapid onset and long acting opioids may potentially achieve both these qualities. In this study, we compared a PCA fentanyl-morphine mixture to morphine alone. Methods: The study received Research Ethics Board approval. Patients undergoing major bowel surgery were recruited. The combination solution was prepared according to a 1:75 fentanyl to morphine potency ratio (fentanyl 13.33 mcg/ml and morphine 1 mg/ml). The morphine alone solution contained 2 mg/ml. Patients were randomly allocated to one of the regimens and were then evaluated 4 times during the first 48 hours following surgery. Time to relief, visual analog pain scores, opioid consumption, demands, deliveries, and side effects on an opioid-related symptom distress scale were recorded. Results: 64 patients were recruited and 58 completed the protocol. Groups were well matched for age, weight and sex. There were no significant differences between groups in time to relief, opioid consumption, pain scores or side effects other than the occurrence of nausea, which was lower for the combination group in one visit. Discussion: Despite the theoretical advantage, this study failed to show a significant advantage to a mixture of fentanyl and morphine over that of morphine alone. Time to relief, consumption and pain scores were similar for both groups in all 4 visits during the first 48 hours. Although nausea was consistently lower in the mixture group, the difference reached statistical significance in only one of the four post-operative visits. The results are similar to a previous study which failed to achieve better or faster pain control with an opioid mixture. 1 It is possible that a higher concentration of fentanyl in the mixture (i.e. a 1:50 potency ratio) would have achieved better results. It is also possible that patient satisfaction was better in the mixture group. However, it is a very subjective measure, and was therefore not included in the interviews. There are very few studies, which explored opioid combinations as a way to optimize pain control. 1,2 Further studies are needed to explore the potential of different potency ratios and opioid combinations to achieve rapid and long lasting pain control. References: 1. Ngan Kee WD, Khaw KS, Wong EL. Randomised doubleblind comparison of morphine vs. a morphine-alfentanil combination for patient controlled analgesia. Anaesthesia 1999;54:629-33. 2. Ludbrook GL, Macintyre PE, Douglas H, Ong

Introduction: Invasive procedures carry the risk for complications such as erroneous placement arising from inadequate manual skills and infection secondary to breaches in aseptic technique. With neuraxial anesthesia, infectious complications may be devastating. Although it is assumed that improvement in sterile technique parallels improved dexterity, this assertion remains unproven. In this trial we determined whether increased proficiency in manual skills for epidural anesthesia is associated with improved aseptic technique. Methods: The study received Research Ethics Board approval. Novice second year anesthesia residents were repeatedly videotaped performing epidural anesthesia over 6month periods. Three independent examiners blinded to the identity and level of training of the residents evaluated the procedures for manual skills and aseptic technique. Each procedure was graded using a scaled manual skills checklist, a global rating scale and a scaled checklist for aseptic technique. Results: 35 sessions were videotaped over 1 year. Inter-rater reliability was nearly perfect for all three measures(ICC 0.91, p<0.001). A significant relationship was found between the number of epidural procedures performed and the manual skills checklist scores (F=80.62, df=1,23, p<0.0001).A strong positive association was found between increased experience and manual skills reflected by the correlation between manual skills checklist scores and the number of epidurals performed (R=0.84,p<0.0001) ). In contrast, no relationship was found between the number of epidural procedures performed and the aseptic technique checklist scores (F=1.67, df=1,23, p=0.21)( ). Discussion: Manual skills for invasive procedures improved with increasing experience but aseptic technique did not, despite formal teaching in this area. These findings reflect major gaps in the understanding of the principles of aseptic technique, most likely due to lack of structured training. 1 Increasing virulence and resistance of contaminating organisms is a major concern and emphasis on meticulous aseptic technique is paramount to patient safety.Educational initiatives are needed to correct these teaching gaps.

Background: Invasive procedures such as epidural anesthesia carry risks for complications such as erroneous placement arising from inadequate manual skills and infection secondary to breaches in aseptic technique. Although it is assumed that improvement in aseptic technique parallels improved dexterity, this assertion remains unproven. The aim of this study was to determine whether increased proficiency in the manual skills for epidural anesthesia is associated with improved aseptic technique.

INTRODUCTION: There is a growing demand for objective assessment of manual skills and competencies in all aspects of invasive procedures. At the present time, there is no valid evaluation tool for assessment of technical proficiency in regional anesthesia procedures. The aim of this study was to validate an objective tool for assessing residents' competency in performing epidural anesthesia by using a global assessment scale and a 3-scale 27-stage checklist. We wish to demonstrate that this tool can differentiate operators with different levels of training. METHOD: The study received Research Ethics Board approval. Second year anesthesia residents were recruited after their previous experience with epidural anesthesia was assessed by questionnaire. The residents were videotaped several times during epidural catheter insertion for labour analgesia over a 6-month period. Videotaping was done in a blinded manner masking the identity and level of training of the resident. Three blinded independent examiners evaluated each session, using a specifically devised assessment comprised of a global rating scale and a 27 stage 3-scale checklist to judge the skill level, and grade the videotaped sessions. RESULTS: 21 sessions from 6 residents were videotaped over 6 months. The first session was taped after 12 epidural procedures were performed independently and the last ones after up to 220 insertions. Inter-rater reliability for the different checklist and global rating form items shows moderate to high degree of agreement for most stages. Total scores demonstrate almost perfect agreement (Kappa / ICC +/-SE= 0.90 +/-0.03 and 0.83 +/-0.13 respectively, p<0.0001) between examiners. To test whether higher total scores are associated with greater experience, a series of repeated measures ANCOVAs were performed. In both the global rating form and the checklist, a significant relationship between total scores and number of epidurals insertions performed was found to exist (checklist: F=86.55, df=1,53, p<0.0001; global rating: F=36.49, df=1,54, p<0.0001). DISCUSSION: The results of our study show that scores on a system comprised of a global rating form and a task specific checklist had a very good correlation with experience in performing epidural analgesia and a significant relationship between total scores and the number of epidurals insertions performed was found to exist. The interrater reliability of these assessment tools is very strong, suggesting we will be able to obtain meaningful data from these scales without inter-examiner differences. Evaluating technical skills by an objective tool under direct observation, as opposed to laboratory setting may create a "gold standard" of assessment. Furthermore, residency programs could use these evaluations for identifying deficiencies in teaching programs and trainees that require extra instruction.

Introduction: Airway manipulation during direct laryngoscopy (DL) and intubation can produce a marked physiological response. These changes can be detrimental in scenarios such as ischemic heart disease, aortic dissection or intracranial hemorrhage (1, 2). Minimizing oropharyngo-laryngeal stimulation or avoiding laryngoscopy may attenuate the hemodynamic response. In this prospective randomized control trial, we compared the hemodynamic response following tracheal intubation, using three different techniques of intubation. We also examined post-operative airway morbidities, as our secondary outcome. Methods: Local REB approval was obtained for this study. Sixty, adult ASA I-II patients, scheduled for elective surgery requiring general anesthesia with oro-tracheal intubation were randomly allocated into one of 3 groups (n=20). DL group -endotracheal intubation was achieved with a Macintosh laryngoscope. GS group-intubation using a Glidescope. TL group-trachlight intubation. Anesthesia technique was standardized, and intubation was performed by a single experienced anesthesiologist. The hemodynamic variables, mean arterial pressure (MAP) and heart rate (HR) were measured noninvasively before induction (baseline values) immediately after induction (post induction values), at intubation, and for 5 minutes after intubation at one-minute intervals. We also recorded the number of attempts and total time for intubation using the three techniques. The severity of post-operative airway morbidity upto 48 hours following surgery was assessed using a questionnaire and an analog visual scale (VAS). Results: There was no significant difference in MAP and HR when compared between the groups (graph 1-2). DL intubation times were significantly lower than those in the GS and TL groups (both p<0.0001). Sore throat and VAS assessed in the PACU were significantly higher in the TL group than in the DL group (p=0.0033). Discussion: No benefit was achieved by using any of the 3 intubation techniques for attenuation of hemodynamic changes. There was a higher incidence of sore throat associated with trachlight intubation than with DL and GS.

▪ Abstract:  An ideal patient-controlled analgesia (PCA) opioid would have both a rapid onset and a long duration of action, attributes, which are not available in currently existing opioids including morphine, the most widely used agent. A mixture of rapid onset and long-acting opioids may potentially achieve both these qualities. In a randomized, double-blind study, we compared a fentanyl–morphine combination with morphine alone for PCA, in 54 patients undergoing bowel surgery. The combination solution was prepared according to a 1:75 fentanyl to morphine potency ratio. The mixture contained fentanyl 13.33 μg/mL and morphine 1 mg/mL. The morphine alone solution contained 2 mg/mL. Patients were randomly allocated to one of the regimens and were then evaluated 4 times during the first 48 hours following surgery. Time to effect, visual analog pain scores, opioid consumption, demands, deliveries, and side effects on an opioid-related symptom distress scale were recorded. Groups were well matched for age, weight, and sex. There were no significant differences between groups in time to effect, PCA usage, pain scores or side effects other than the occurrence of nausea, which was lower for the combination group in 1 visit. Further studies are needed to explore the potential of different potency ratios and opioid combinations to achieve rapid and long-lasting pain control. ▪