TABLE I. Key Purposes of the Reviewed Studies.

Purposes	Brief Description	Reference	Frequency
	Identify the infected individual more quickly	[21]
	Screen coronavirus diseases using deep learning	[27]
	Identify the coronavirus patients	[28]
	Develop a CNN-based algorithm to detect COVID-19 from CT images	[20]
	Detect COVID-19 with the help of AI and smartphone sensors	[29]
	Use an anomaly model based on a deep learning network to make the screening process faster for COVID-19 detection from X-ray images	[26]
	Detect COVID-19 from X-ray images using transfer learning with CNN	[30]
	Detect COVID-19 from X-ray images using a deep CNN model	[31]
	Propose an algorithm to detect COVID-19 from CT images using a deep CNN model and SVM classifier	[32]
	Develop a deep learning model CoroNet using the Xception CNN to detect COVID-19 from X-ray images	[33]
	Build a framework that uses smartphone sensors to detect COVID-19	[29]
Diseases detection	Classify patients into non-COVID 19 infection, COVID-19 infection, and no infection from X-ray images using a deep CNN model	[26]	19
	Compare the performance of seven DL models to find the best model for COVID-19 detection	[34]
	Develop and evaluate the performance of an AI model to detect COVID-19 and also evaluate the performance of radiologists to detect the disease by using and without AI support	[35]
	Detect the COVID-19 by identifying the characteristics from chest X-ray using a deep learning model(CAD4COVID-XRay)	[36]
	Detect COVID-19 from X-ray images using generative adversarial network (GAN) and deep learning transfer	[37]
	Develop and evaluate and AI-based system for detecting COVID-19 from a globally diverse and multi-institution dataset	[38]
	Develop several AI models to identify COVID-19 positive patients using blood counts without knowledge of symptoms or history of the individuals	[39]
	Detect COVID-19 with faster R-CNN using X-ray images for real-time assessment	[40]
	Diagnose the identified patients to classify (in to patients’ categories) and tracking the progress COVID-19 patients	[41]
	Distinguish COVID-19 from pneumonia using deep learning	[42]
	Efficiently diagnose COVID-19 using X-ray images through deep CNN models	[43]
	Develop a tool to predict survival and death for severe COVID-19 patients	[44]
	Diagnosis COVID-19 positive case faster using both non-image and image clinical data	[45]
Diseases diagnosis	Develop a system to identify patients who would develop more severe illness among the patients with mild cases of COVID-19	[46]	9
	Develop a system to improve the diagnostic performance from posterior-anterior (PA) X-ray images of lungs with COVID-19 cases	[47]
	Analyse and predicting the risk of COVID-19 patients based on ML models using patient's baseline clinical parameters	[48]
	Develop a deep learning-based model to repurpose commercially available drugs to disrupt viral proteins of SARS-Cov-2	[49]
	Forecast of the COVID-19 to estimate size, lengths and ending time of COVID-19 across China	[19]
Epidemic forecasting	Predict the trend of the infection for the next 80 days using deep learning as well as the progress of the epidemic (epidemic sizes and peaks)	[50]	3
	Predict the growth of the COVID-19 pandemic using mathematical modeling, ML and cloud computing	[51]
Sustainable development	Analyze the correlation among environmental factors and confirmed cases of COVID-19	[52]	1
	Compare the prediction performance of the proposed algorithms with the existing methods	[32]
Performance comparison	Compare seven different DL models to find out the best model for disease detection	[34]	4
	Compare the performance of radiologists in distinguishing COVID-19 from other pneumonia with and without AI assistance	[35]
	Compare the performance of a DL model with six other radiologists	[36]
Patient management	Improve management of COVID-19 ICU patients	[53]	1