Your browser doesn't support javascript.
Show: 20 | 50 | 100
Results 1 - 2 de 2
Add filters

Document Type
Year range
Healthcare (Basel) ; 10(6)2022 Jun 10.
Article in English | MEDLINE | ID: covidwho-1911281


Lockdown implementation during COVID-19 pandemic has caused many negative impacts in various aspect of life, including in the academic world. Routine disruption to teaching and learning environment has raised concerns to the wellbeing of university staff and students. This study aimed to examine the subjective wellbeing of the university community in Northern Malaysia during lockdown due to COVID-19 pandemic and the factors affecting it. An online cross-sectional survey involving 1148 university staff and students was conducted between March and April 2020. The research tools include the Personal Wellbeing Index (PWI) to assess subjective wellbeing and the Depression, Anxiety and Stress 21 (DASS-21) scale for psychological distress. While we found the subjective wellbeing score in our study population was stable at 7.67 (1.38), there was high prevalence of anxiety, depression, and stress with 27.4%, 18.4%, and 11.5%, respectively. The students reported higher levels of psychological distress compared to staff. The PWI score was seen to be inversely affected by the depression and stress score with a reduction in the PWI score by 0.022 (95% CI -0.037 to -0.007) and 0.046 (95% CI -0.062 to -0.030) with every one-unit increment for each subscale, respectively. Those who perceived to have more difficulty due to the lockdown also reported low subjective wellbeing. Thus, it is crucial to ensure policies and preventative measures are in place to provide conducive teaching and learning environment. Additionally, the detrimental psychological effects especially among students should be addressed proactively.

Diagnostics ; 12(5):1258, 2022.
Article in English | ProQuest Central | ID: covidwho-1871356


Numerous research have demonstrated that Convolutional Neural Network (CNN) models are capable of classifying visual field (VF) defects with great accuracy. In this study, we evaluated the performance of different pre-trained models (VGG-Net, MobileNet, ResNet, and DenseNet) in classifying VF defects and produced a comprehensive comparative analysis to compare the performance of different CNN models before and after hyperparameter tuning and fine-tuning. Using 32 batch sizes, 50 epochs, and ADAM as the optimizer to optimize weight, bias, and learning rate, VGG-16 obtained the highest accuracy of 97.63 percent, according to experimental findings. Subsequently, Bayesian optimization was utilized to execute automated hyperparameter tuning and automated fine-tuning layers of the pre-trained models to determine the optimal hyperparameter and fine-tuning layer for classifying many VF defect with the highest accuracy. We found that the combination of different hyperparameters and fine-tuning of the pre-trained models significantly impact the performance of deep learning models for this classification task. In addition, we also discovered that the automated selection of optimal hyperparameters and fine-tuning by Bayesian has significantly enhanced the performance of the pre-trained models. The results observed the best performance for the DenseNet-121 model with a validation accuracy of 98.46% and a test accuracy of 99.57% for the tested datasets.