Table 2.

A summary of notable works on risk assessment during the COVID-19 pandemic.

Research	Application	Methodology	Performance
[73]	COVID-19 diagnosis from symptoms	Fuzzy Inference System	No metrics provided
[74]	Vulnerability assessment of individuals	Logistic regression, Gradient Boosted Trees and Closedloop platform	No metrics provided
[80]	Mortality risk assessment from COVID-19	Artificial Neural Network	Accuracy: 86.25%, Sensitivity: 87.50%, Specificity: 85.94%, AUROC: 90.12%
[82]	COVID-19 mortality risk assessment using blood samples	Ensemble learning using deep neural network and random forest	Sensitivity 100%, Specificity 91% and Accuracy 92%
[84]	COVID-19 recovery period estimation and identification of high risk age groups	Decision tree, Support Vector Machine, Naive Bayes, Logistic Regression, Random Forest and K-nearest Neighbour	Accuracy 99.85%
[79]	COVID-19 severity assessment from chest x-ray images	Transfer learning with VGG16	$R^2$ 0.90 and mean absolute error of 8.5%
[85]	Identification of high risk individuals that require hospitalisation and low risk individuals that can recover at home using a chatbot	AI chatbot	No metrics provided
[86]	Predict development of acute respiratory distress syndrome (ARDS) of COVID-19 patients from clinical data	Logistic Regression, KNN, Decision Tree, Random Forests and Support Vector Machine	Accuracy: 70% to 80%