Exploring Machine Learning Strategies in COVID-19 Prognostic Modelling: A Systematic Analysis of Diagnosis, Classification and Outcome Prediction (preprint)

Background: The COVID-19 pandemic, which has impacted over 222 countries resulting in incalculable losses, has necessitated innovative solutions via machine learning (ML) to tackle the problem of overburdened healthcare systems. This study consolidates research employing ML models for COVID-19 prognosis, evaluates prevalent models and performance, and provides an overview of suitable models and features while offering recommendations for experimental protocols, reproducibility and integration of ML algorithms in clinical settings. Methods: We conducted a review following the PRISMA framework, examining ML utilisation for COVID-19 prediction. Five databases were searched for relevant studies up to 24 January 2023, resulting in 1,824 unique articles. Rigorous selection criteria led to 204 included studies. Top-performing features and models were extracted, with the area under the receiver operating characteristic curve (AUC) evaluation metric used for performance assessment. Results: This systematic review investigated 204 studies on ML models for COVID-19 prognosis across automated diagnosis (18.1%), severity classification (31.9%), and outcome prediction (50%). We identified thirty-four unique features in five categories and twenty-one distinct ML models in six categories. The most prevalent features were chest CT, chest radiographs, and advanced age, while the most frequently employed models were CNN, XGB, and RF. Top-performing models included neural networks (ANN, MLP, DNN), distance-based methods (kNN), ensemble methods (XGB), and regression models (PLS-DA), all exhibiting high AUC values. Conclusion: Machine learning models have shown considerable promise in improving COVID-19 diagnostic accuracy, risk stratification, and outcome prediction. Advancements in ML techniques and their integration with complementary technologies will be essential for expediting decision-making and informing clinical decisions, with long-lasting implications for healthcare systems globally.

Reactions to the media coverage during the first wave of COVID-19 in Bangladesh.

BACKGROUND: The news media play a critical role in disseminating accurate and reliable information during an outbreak like COVID-19, especially in LMICs. Studying how people react and reflect on the information provided and how it affects their trust in health systems is essential for effective risk communication. This study was undertaken to explore and analyse newspaper readers' reactions to the unfolding news of the COVID-19 outbreak in Bangladesh and how this affected and shaped their compliance with the mitigation measures advised by the Government. METHODS: We collected readers' comments on relevant news and features on the COVID-19 outbreak (n = 1,055) which were posted in the online versions of the four top circulating Bangla newspapers and one online news portal published during Jan.-Apr. 2020. A search protocol was developed and a team of three researchers searched and extracted data for content analysis according to some pre-determined study themes. RESULTS: Data analysis revealed several characteristics with implications for risk-communication: a faith-based and fatalistic attitude to the unfolding pandemic, a "denial" syndrome in the initial stage, a returning expatriate-bashing for specific countries, and a concern about the safety of the frontline health workers. The readers were resentful of the all-pervasive corruption in the health sector even in times of a pandemic and the Government's poorly coordinated, fragmented, and delayed COVID-19 response. The pandemic severely shook their trust in the already weak health system and perceived it to be incompetent, corrupt, and non-responsive. They had deplorable personal and family experiences while seeking treatment for COVID-19 patients. Expert committees were formed to advise the Government, but few recommendations were implemented on the ground. This helpless scenario made people sharply critical of the political leadership, especially for the failure of providing stewardship at the moment of crisis. CONCLUSIONS: The COVID-19 related information reaching the people, including misinformation, disinformation, and rumours was equivocal in the early months of the pandemic and failed to build the trust and transparency that is necessary for an inclusive response across constituencies. The Government should pay attention and weightage to people's perceptions about its COVID-19 response and take appropriate measures to re-build trust for implementing pandemic control measures.

Cyclone “Amphan” evacuation dilemmas of coastal households amidst COVID-19 pandemic: A study of South-Western region of Bangladesh (preprint)

Cyclone ‘Amphan’ battered the coastal communities in the south-western part of Bangladesh in 2020 during the Covid-19 pandemic. The coastal communities experienced the situation for the first time and were in dilemma whether to stay at home to embrace the cyclone alone or risking the Covid-19 contagion in the cyclone shelters by evacuating themselves. This article intended to explore cyclone Amphan evacuation dynamics among the coastal households amidst COVID-19 pandemic. The study investigated evacuation behaviors among households and explored the impacts of COVID-19 on the evacuation processes. We adopted household survey for collecting primary information and determined 378 samples for interviews at a precision level of 0.05 in fourteen villages. Results demonstrated that despite the utmost effort from the government, 96.6% people in the coastal area received an evacuation order before the landfall, and only 42% people respected the evacuation order. Majority households choose to stay at home because of fear due to Covid-19 in the crowded shelters. Although half of the evacuees were housed in the cyclone shelter, visible COVID-19 protecting facilities were unavailable. Thus, this study would assist future government policies and enhance disaster evacuation plans by incorporating the pandemic to reduce disaster risks in the global south.

Is Bitcoin a Hedge? How Extreme Volatility Can Destroy the Hedge Property (preprint)

A large literature tests whether Bitcoin can hedge portfolio risk, i.e. reduce the risk if added to a portfolio. Intuitively, given the extreme volatility and thus risk of Bitcoin and cryptocurrencies, the idea that Bitcoin is a hedge may be puzzling. Indeed, we show that for extreme levels of volatility, Bitcoin does not reduce the risk if added to a benchmark equity portfolio. This is not only true on average but also holds for sub-samples, including the COVID-19 crisis period. We conclude that a focus on correlations is not sufficient for extreme levels of volatility.

COVID-19 Breaking News in the Print Media in Bangladesh: People's Reactions to the Evolving Pandemic, and Perceptions Regarding Government's Response, in the First Four Months (preprint)

Background: The news media play a critical role in disseminating accurate and reliable information during an outbreak like COVID-19, especially in LMICs. Studying how people react and reflect on the information provided and how it affects their trust in the health systems, are important for designing effective risk communication. This study was undertaken to explore and analyse newspaper readers’ reactions to the unfolding news of the COVID-19 outbreak in Bangladesh and how this affected and shaped their compliance with the mitigation measures advised by the government.Methods: We collected readers’ comments on relevant news and features on the COVID-19 outbreak (n=1,055) which were posted in the online versions of the four top circulating Bangla newspapers and one online news portal published during Jan.-Apr. 2020. A search protocol was developed and a team of three researchers searched and extracted data for content analysis according to some pre-determined study themes.Results: Analysis of data revealed several characteristics with implications for risk-communication: a faith-based and fatalistic attitude to the unfolding pandemic, a “denial” syndrome in the initial stage, a returning expatriate-bashing for specific countries, and a concern about the safety of the frontline health workers. The readers were resentful of the all-pervasive corruption in the health sector even in times of a pandemic and the Government’s poorly coordinated, fragmented, and delayed COVID-19 response. The pandemic severely shook their trust in the already weak health system and perceived it to be incompetent, corrupt, and non-responsive. They had deplorable personal and family experiences while seeking treatment for COVID-19 patients. Expert committees were formed to advise the government, but few recommendations were implemented on the ground. This helpless scenario made people sharply critical of the political leadership, especially for the failure of providing stewardship at the moment of crisis.Conclusions: The COVID-19 related information () reaching the people including misinformation, disinformation, and rumours was equivocal in the early months of the pandemic and failed to build the trust and transparency that is necessary for an inclusive response across constituencies. The government should pay attention and weightage to people’s perceptions about its COVID-19 response and take appropriate measures to re-build trust for implementing pandemic control measures.Funding Information: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.Declaration of Interests: The authors declare that they have no competing interest.Ethics Approval Statement: The very nature of the data (readers’ reactions and comments) did not allow us to get individual consent for participation. Also, the reactions/comments were posted by name only without an address, thus further follow-up being impossible. However, the study followed ethical principles in conducting the study and the confidentiality of the data was strictly maintained by the study team and the readers’ reactions /comments were used anonymously for research purposes only.

CovidEnvelope: A Fast Automated Approach to Diagnose COVID-19 from Cough Signals (preprint)

The COVID-19 pandemic has a devastating impact on the health and well-being of global population. Cough audio signals classification showed potential as a screening approach for diagnosing people, infected with COVID-19. Recent approaches need costly deep learning algorithms or sophisticated methods to extract informative features from cough audio signals. In this paper, we propose a low-cost envelope approach, called CovidEnvelope, which can classify COVID-19 positive and negative cases from raw data by avoiding above disadvantages. This automated approach can pre-process cough audio signals by filter-out back-ground noises, generate an envelope around the audio signal, and finally provide outcomes by computing area enclosed by the envelope. It has been seen that reliable datasets are also important for achieving high performance. Our approach proves that human verbal confirmation is not a reliable source of information. Finally, the approach reaches highest sensitivity, specificity, accuracy, and AUC of 0.92, 0.87, 0.89, and 0.89 respectively. The automatic approach only takes 1.8 to 3.9 minutes to compute these performances. Overall, this approach is fast and sensitive to diagnose the people living with COVID-19, regardless of having COVID-19 related symptoms or not, and thus have vast applicability in human well-being by designing HCI devices incorporating this approach.