Development of a deep learning classifier to accurately distinguish COVID-19 from look-a-like pathology on lung ultrasound

Abstract

ObjectivesLung ultrasound (LUS) is a portable, low cost respiratory imaging tool but is challenged by user dependence and lack of diagnostic specificity. It is unknown whether the advantages of LUS implementation could be paired with deep learning techniques to match or exceed human-level, diagnostic specificity among similar appearing, pathological LUS images.DesignA convolutional neural network was trained on LUS images with B lines of different etiologies. CNN diagnostic performance, as validated using a 10% data holdback set was compared to surveyed LUS-competent physicians.SettingTwo tertiary Canadian hospitals.Participants600 LUS videos (121,381 frames) of B lines from 243 distinct patients with either 1) COVID-19, Non-COVID acute respiratory distress syndrome (NCOVID) and 3) Hydrostatic pulmonary edema (HPE).ResultsThe trained CNN performance on the independent dataset showed an ability to discriminate between COVID (AUC 1.0), NCOVID (AUC 0.934) and HPE (AUC 1.0) pathologies. This was significantly better than physician ability (AUCs of 0.697, 0.704, 0.967 for the COVID, NCOVID and HPE classes, respectively), p < 0.01.ConclusionsA deep learning model can distinguish similar appearing LUS pathology, including COVID-19, that cannot be distinguished by humans. The performance gap between humans and the model suggests that subvisible biomarkers within ultrasound images could exist and multi-center research is merited.

Diagnostic accuracy of point-of-care ultrasound for pulmonary tuberculosis: A systematic review

“…Unlike CXR, it is unclear whether POCUS images are suitable for AI-assisted interpretation. However, early work has been undertaken on the AI-assisted interpretation of LUS findings for paediatric pneumonia and COVID-19 [25][26][27], suggesting it may one day become a possibility for POCUS for PTB. Either way, if POCUS could be shown to have similar diagnostic accuracy to CXR for PTB, it could be a valuable diagnostic tool given its low hardware costs, ability to reach primary care, and near-immediate provision of results.…”

A Comparative and Extensive Study of Covid 19 Diagnosis using Lung Ultrasound Images

“…Also out of 5663 articles found on MI, only 775 articles mentioned AI which is about 14 percent. The figure [3] highlights these results.…”

International Conference on Advances in Science, Engineering and Technology

“…An in-depth study model can distinguish between the emergence of LUS pathology, such as COVID-19, that people can't tell the difference between. The difference in picture quality between humans and models shows that biomarkers are not evident in existing ultrasound images, and that multidisciplinary investigations have been confirmed Arntfield et al [2020]. In most cases, we're creating a new deep network of Spatial Transformer Networks that forecasts the severity of the sickness connected with the input framework while also delivering a subtle state-of-the-art environment.…”