ABSTRACT
As SARS-CoV-2 threatens global public health, computer aided detection has been frequently proposed to assist medical professionals in screening patients for the infection. Based on the state-of-the-art, clinical images, such as chest X-rays (CXR), have been used in artificial intelligence processes like convolutional neural networks in detecting the virus. Additionally, no study has proposed a classification model that accurately predicts the overall SARS-CoV-2 severity scores (OSS) of patients using multiple CXR image features. Therefore, the present study introduces a model that predicts the patients' OSS using their CXR image features. The dataset consisted of 1007 CXR images opacity and geographic extent scores-generated by COVIDNet models in addition to the airspace disease grading scores as the input features while the OSS served as the output label. The dataset was preprocessed using the rescaling, binning, and data splitting techniques to increase the models' predictive accuracy. Among all the trained classification models, the Neural Network-based Trilayered model achieved the highest accuracy of 81.19%, in predicting the OSS specifically for mild severity score. Significantly, the study proves the potential of ML models in accurately predicting the OSS based on CXR image features;however, it is recommended to create a more balanced and larger dataset, and consult more radiologists for the CXR images' OSS validation. © 2022 IEEE.
ABSTRACT
The downside of the integrative aspect of the digital space is how easily fake news can propagate which jeopardized the regulation and control measures of the COVID-19 pandemic. While existing literature expounds on the nature of infodemic phenomenon, recent curiosities lack the exploration of the contributing factors that led to the inability to recognize fake news on social media as it corresponds to the New Media Literacy (NML) levels. NML allows adaptation to technological advancement as it constantly evolves with great sophistication. Anchored from this gap, the study employs a quantitative research design where 385 respondents from Cebu City-a highly urbanized city in the Philippines-were asked to answer a three-part survey questionnaire. The findings purport that a high percentage of respondents can distinguish legitimate from fake news and take proactive measures in reporting or resharing the posts. Moreover, the study reveals that the respondents have high NML levels, particularly in functional prosuming and consuming aspects, which the study probed according to the demographic factors. The salient discussion then revolves around the low critical outcomes of prosuming and consuming NML aspects to push for educational policy formulation methods with interpretive social-scientific approaches. This reinforces the post-truth lens in expanding the fields of concerns arising from the infodemic phenomenon. Furthermore, recommendatory measures are provided in the Philippine educational system- that may be reintegrated into the dimensions of policy theories for educational policy evaluation to probe different areas of improvement in the Media and Information Literacy of the K-12 curriculum.