A Novel Abnormality Annotation Database for COVID-19 Affected Frontal Lung X-rays

Abstract

PurposeTo advance the usage of CXRs as a viable solution for efficient COVID-19 diagnostics by providing large-scale annotations of the abnormalities in frontal CXRs in BIMCV-COVID19+ database, and to provide a robust evaluation mechanism to facilitate its usage.Materials and MethodsWe provide the abnormality annotations in frontal CXRs by creating bounding boxes. The frontal CXRs are a part of the existing BIMCV-COVID19+ database. We also define four different protocols for robust evaluation of semantic segmentation and classification algorithms. Finally, we benchmark the defined protocols and report the results using popular deep learning models as a part of this study.ResultsFor semantic segmentation, Mask-RCNN performs the best among all the models with a DICE score of 0.43 ± 0.01. For classification, we observe that MobileNetv2 yields the best results for 2-class and 3-class classification. We also observe that deep models report a lower performance for classifying other classes apart from the COVID class.ConclusionBy making the annotated data and protocols available to the scientific community, we aim to advance the usage of CXRs as a viable solution for efficient COVID-19 diagnostics. This large-scale data will be useful for ML algorithms and can be used for learning radiological patterns observed in COVID-19 patients. Further, the protocols will facilitate ML practitioners for unified large-scale evaluation of their algorithms.Data Availability StatementThe data associated with this work is available here : Radiologists’ Annotations on COVID-19+ X-rays https://osf.io/b35xu/ via @OSFramework andhttp://covbase4all.igib.res.in/.

Bioinformatics, Computational Informatics, and Modeling Approaches to the Design of mRNA COVID-19 Vaccine Candidates

“…Some of the sources are in real time, for instance web-dashboards [92-95], while others are diverse. COVID-19 data are of different forms, namely genomic sequences of different variants of the SARS-CoV-2 virus [92-95], chest X-rays of COVID-19 patients [96], kidney replacement therapy data for COVID-19 patients [97], lungs data of patients [98], blood data of COVID-19 patients [99][100][101][102][103][104], medical data of patients that were infected and recovered [105], medical data of patients that were infected and died [106], patients' demographic data, patients' bio-data, COVID-19 health facility data for different regions of the world, and COVID-19 phylogenetic data, amongst others. Genomic data collected for the SARS-CoV-2 virus were stored in bioinformatics databases such as NCBI, EBL, and GISAID [20,107,108].…”

Public Covid-19 X-ray datasets and their impact on model bias – A systematic review of a significant problem

“… d AIforCOVID ( Soda et al., 2020 ),( https://aiforcovid.radiomica.it ). e CARING , external annotation ( Mittal et al., 2021 ) ( https://osf.io/b35xu/ ). f Immunoglobulin G (IgG) and Immunoglobulin M(IgM).…”

FedSGDCOVID: Federated SGD COVID-19 Detection under Local Differential Privacy Using Chest X-ray Images and Symptom Information

“…Following the work of a group of researchers from Qatar University and the University of Dhaka, Bangladesh, and collaborators from Pakistan, Malaysia, and medical doctors [ 77 , 78 ], we collected a dataset containing 3616 COVID-19 positives, 10,192 normal and 1345 viral pneumonia chest X-ray images. The COVID-19 data were collected from the various publicly accessible datasets, online sources, and published papers [ 79 , 80 , 81 , 82 , 83 , 84 ], normal data were collected from two different datasets [ 85 , 86 ], and viral pneumonia data were collected from chest X-ray images (pneumonia) database [ 86 ]. Few samples of chest X-ray images are shown in Figure 6 .…”