Deep learning and its role in COVID-19 medical imaging

Abstract

COVID-19 is one of the greatest global public health challenges in history. COVID-19 is caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and is estimated to have an cumulative global case-fatality rate as high as 7.2%[ 1 ]. As the SARS-CoV-2 spread across the globe it catalyzed new urgency in building systems to allow rapid sharing and dissemination of data between international healthcare infrastructures and governments in a worldwide effort focused on case tracking/tracing, identifying effective therapeutic protocols, securing healthcare resources, and in drug and vaccine research. In addition to the worldwide efforts to share clinical and routine population health data, there are many large-scale efforts to collect and disseminate medical imaging data, owing to the critical role that imaging has played in diagnosis and management around the world. Given reported false negative rates of the reverse transcriptase polymerase chain reaction (RT-PCR) of up to 61%[ 2 , 3 ], imaging can be used as an important adjunct or alternative. Furthermore, there has been a shortage of test-kits worldwide and laboratories in many testing sites have struggled to process the available tests within a reasonable time frame. Given these issues surrounding COVID-19, many groups began to explore the benefits of 'big data' processing and algorithms to assist with the diagnosis and therapeutic development of COVID-19.

Prediction of the COVID-19 infectivity and the sustainable impact on public health under deep learning algorithm

“…Also, the required data are selected and encoded. Further, data redundancy attributes, such as the patient's name and contact information in the infection case table are deleted (Desai et al 2020 ). Additionally, the missing data should be completed, and one-hot-coding should be performed for the disease coding data and area coding data.…”

Applications of artificial intelligence in COVID-19 pandemic: A comprehensive review

“…Recently, deep learning-based algorithms have been used by various researchers for combating the COVID-19 pandemic, including convolutional neural network (CNN), recurrent neural network (RNN), and long short-term memory (LSTM) for the COVID-19 detection, diagnosis, classification. Screening, drug repurposing, prediction, and forecasting ( Bogu and Snyder, 2021 , Desai et al, 2020 , Ghoshal and Tucker, 2020 , He et al, 2020 , Hu et al, 2020 , Khurana et al, 2021 , Baig et al, 2019 , Pan et al, 2021 , Sarv Ahrabi et al, 2021 , Sedik et al, 2021 , Soni and Roberts, 2021 ).…”

Imaging spectrum of abdominal manifestations of COVID-19

“…With increasing awareness and recognition of the extra-pulmonary manifestations of COVID-19, particularly the thrombotic complications, the corresponding imaging findings could also serve as appropriate targets for AI systems[ 67 ]. A subtype of the deep learning technology, namely, convolutional neural networks, has generated great interest amongst the radiology community owing to their varied applications[ 70 ] and could have a potential role in the detection of extra-pulmonary thrombotic complications of COVID-19 such as multi-organ infarction and arterial and venous thrombi on CT imaging. This technique of machine learning, however, would require the availability of large amounts of input data for adequate ‘training’.…”

“…Moreover, it is advisable to include external test data sets of patients from a different demographic from the one used for training, so as to ascertain if the AI algorithm can be applied to various patient populations. Thus, multi-institutional collaboration and global data sharing are a must to generate data sets satisfactory for machine learning[ 70 , 71 ]. The availability of high-quality data sets of thrombotic complications in COVID-19 patients, detected on CT angiography images, could be used to develop innovative machine learning models for detection and quantification of extra-pulmonary disease severity.…”