2020/04/16 by Florin Condrea, Condrea, Florin, Victor-Andrei Ivan +3
Engineering · Medicine · #Non-Invasive Vital Sign Monitoring #COVID-19 diagnosis using AI
paper · pdf · doi:10.48550/arxiv.2004.07691
Automatically detecting vital signs in videos, such as the estimation of\nheart and respiration rates, is a challenging research problem in computer\nvision with important applications in the medical field. One of the key\ndifficulties in tackling this task is the lack of sufficient supervised\ntraining data, which severely limits the use of powerful deep neural networks.\nIn this paper we address this limitation through a novel deep learning\napproach, in which a recurrent deep neural network is trained to detect vital\nsigns in the infrared thermal domain from purely synthetic data. What is most\nsurprising is that our novel method for synthetic training data generation is\ngeneral, relatively simple and uses almost no prior medical domain knowledge.\nMoreover, our system, which is trained in a purely automatic manner and needs\nno human annotation, also learns to predict the respiration or heart intensity\nsignal for each moment in time and to detect the region of interest that is\nmost relevant for the given task, e.g. the nose area in the case of\nrespiration. We test the effectiveness of our proposed system on the recent\nLCAS dataset and obtain state-of-the-art results.\n