ABSTRACT
We recently read an interesting study which demonstrated that self-inflating resuscitation bag (SIRB) lacking expiratory valve has unreliable performance in oxygen delivery during spontaneous breathing mimicked by mechanical lung simulator. It was postulated that the absence of an expiratory valve and the resulting air entrainment via the exhaust port accounts for the poor oxygen delivery performance. The current disposable SIRB in-use in our institutions (Med-Rescuer Disposable BVM Resuscitator 4000, BLS Systems Limited, ON, Canada) has a duckbill valve but no expiratory valve. Safety concerns regarding its oxygen delivery performance during spontaneous breathing were raised, as this SIRB was commonly used to preoxygenate critically ill patient with potentially transmissible respiratory infection (e.g. COVID-19) before tracheal intubation. We therefore performed an experiment on this SIRB using one of us as a healthy volunteer. Our small experiment demonstrated that air entrainment could occur via the exhaust port and affect oxygen delivery performance. Our experiment also demonstrated that attaching a positive end-expiratory pressure (PEEP) valve to the exhaust port improves the oxygen delivery performance. The findings of this experiment were sent to the relevant department of our institutions for safety consideration. © The Author(s) 2022.
ABSTRACT
The COVID-19 pandemic led to all institutions of education having to transition to fully online learning almost immediately. However, research showed that online learning still lacked adequate interactions with students. This is even more problematic when students are online learning on their own, when adequate online scaffolding activities are absent. This study investigated the impact of chatbots as a scaffolding agent to assist student learning during their independent online learning times. A total of 62 Diploma level students participated in this mixed method research study and presented with a multimedia-based AI chatbot named MERLIN. Data was collected on their attitudes towards using it. Results showed that students were motivated to learn more using MERLIN, improved their learning, and wanted more chatbots in their other courses. These findings have important implications for using AI chatbots as a scaffolding and instructional tool in 21st-century learning environments. © 2022, International Journal of Technology. All Rights Reserved.
ABSTRACT
Wearing face masks in public spaces has become an essential step to prevent the spread of COVID-19. This step poses some challenges to conventional face recognition due to several reasons: 1) the absence of large real-world masked face recognition dataset, and 2) the loss of some visual cues due to the occlusion by the face masks. To address these challenges, this paper presents a real-world masked face recognition dataset that consists of 80500 masked face images of 161 subjects, referred to as MFRD-80K dataset. Every subject contributes 500 masked face images, which are then partitioned into 60:20:20 for train, validation and test. Subsequently, we conduct some benchmark studies to evaluate the performance of the existing face recognition and classification methods on the MFRD-80K dataset. The methods include k-Nearest Neighbour, Multinomial Logistic Regression, Support Vector Machines, Random Forest, Multilayer Perceptron and Convolutional Neural Networks. Since the parameter settings affect the performance of each method, a grid search is performed to determine the optimal parameter settings. The empirical results demonstrate that Convolutional Neural Network achieves the highest test accuracy of 97.16% on MFRD-80K dataset. © 2021, International Association of Engineers. All rights reserved.