ABSTRACT
Currently, the relevance of remote express diagnostics of various diseases is beyond doubt. The active spread of various types of epidemics and pandemics necessitates the improvement of various types of express diagnostics. The authors conduct research in the field of remote visual diagnostics using modern methods of processing telemedicine video information. This paper discusses the possibility of improving the quality of visualization of diagnostic signs using the digital dermatoscopy method for express diagnostics of skin rashes in COVID-19 in comparison with the manifestations of atopic dermatitis. The prospect of this work is the study of illumination conditions during registration and selection of skin areas for the analysis of diagnostic images. © 2021 IEEE.
ABSTRACT
This paper presents a novel ultrasound imaging point-of-care (PoC) COVID-19 diagnostic system. The adaptive visual diagnostics utilize few-shot learning (FSL) to generate encoded disease state models that are stored and classified using a dictionary of knowns. The novel vocabulary based feature processing of the pipeline adapts the knowledge of a pretrained deep neural network to compress the ultrasound images into discrimative descriptions. The computational efficiency of the FSL approach enables high diagnostic deep learning performance in PoC settings, where training data is limited and the annotation process is not strictly controlled. The algorithm performance is evaluated on the open source COVID-19 POCUS Dataset to validate the system's ability to distinguish COVID-19, pneumonia, and healthy disease states. The results of the empirical analyses demonstrate the appropriate efficiency and accuracy for scalable PoC use. The code for this work will be made publicly available on GitHub upon acceptance. © 2021 IEEE.