Multi-task Learning Dataset for the Development of Remote Patient Monitoring System

The COVID-19 pandemic caused havoc on the world, infecting more than 3.5 billion people and resulting in over 15 million deaths, and overwhelmed existing healthcare infrastructures around the world, as announced by the World Health Organization (WHO). We propose in this work an effective and low-cost strategy for collecting, pre-processing, and extracting meaningful information from different types of patient data that may be useful for statistics and training of Machine Learning (ML) models to respond to pandemics such as COVID-19. Information like medical history, clinical examination, para-clinical testing, and patient RGB videos are collected This achievement will enable further studies to train, test, and deploy on-device decentralized ML models to monitor patients at home.