Machine learning applied on chest x-ray can aid in the diagnosis of COVID-19: a first experience from Lombardy, Italy

European Radiology Experimental

Annalisa Polidori¹  Caterina Beatrice Monti²  Francesco Sardanelli³  Isabella Castiglioni⁴  Simone Schiaffino⁵  Davide Ippolito⁶  Davide Gandola⁶  Matteo Interlenghi⁷  Cristina Messa⁸  Christian Salvatore⁹ 

[1] DeepTrace Technologies S.R.L., Via Conservatorio 17, 20122, Milan, Italy;Department of Biomedical Sciences for Health, Università degli Studi di Milano, Via Mangiagalli 31, 20133, Milan, Italy;Department of Biomedical Sciences for Health, Università degli Studi di Milano, Via Mangiagalli 31, 20133, Milan, Italy;Department of Radiology, IRCCS Policlinico San Donato, Via Morandi 30, 20097, San Donato Milanese, Milan, Italy;Department of Physics, Università degli Studi di Milano-Bicocca, Piazza della Scienza 3, 20126, Milan, Italy;Institute of Biomedical Imaging and Physiology, National Research Council, 20090, Segrate, Milan, Italy;Department of Radiology, IRCCS Policlinico San Donato, Via Morandi 30, 20097, San Donato Milanese, Milan, Italy;Department of Radiology, San Gerardo Hospital, Via Pergolesi 33, 20900, Monza, Italy;Institute of Biomedical Imaging and Physiology, National Research Council, 20090, Segrate, Milan, Italy;School of Medicine and Surgery, University of Milano-Bicocca, Piazza dell’Ateneo Nuovo 1, 20126, Milan, Italy;Fondazione Tecnomed, Università degli Studi di Milano-Bicocca, Palazzina Ciclotrone, Via Pergolesi 33, 20900, Monza, Italy;Scuola Universitaria Superiore IUSS Pavia, Piazza della Vittoria 15, 27100, Pavia, Italy;DeepTrace Technologies S.R.L., Via Conservatorio 17, 20122, Milan, Italy;

关键词: Artificial intelligence;    COVID-19;    Neural networks (computer);    Sensitivity and specificity;    X-rays;   

DOI  :  10.1186/s41747-020-00203-z

来源: Springer

【 摘 要 】

BackgroundWe aimed to train and test a deep learning classifier to support the diagnosis of coronavirus disease 2019 (COVID-19) using chest x-ray (CXR) on a cohort of subjects from two hospitals in Lombardy, Italy.MethodsWe used for training and validation an ensemble of ten convolutional neural networks (CNNs) with mainly bedside CXRs of 250 COVID-19 and 250 non-COVID-19 subjects from two hospitals (Centres 1 and 2). We then tested such system on bedside CXRs of an independent group of 110 patients (74 COVID-19, 36 non-COVID-19) from one of the two hospitals. A retrospective reading was performed by two radiologists in the absence of any clinical information, with the aim to differentiate COVID-19 from non-COVID-19 patients. Real-time polymerase chain reaction served as the reference standard.ResultsAt 10-fold cross-validation, our deep learning model classified COVID-19 and non-COVID-19 patients with 0.78 sensitivity (95% confidence interval [CI] 0.74–0.81), 0.82 specificity (95% CI 0.78–0.85), and 0.89 area under the curve (AUC) (95% CI 0.86–0.91). For the independent dataset, deep learning showed 0.80 sensitivity (95% CI 0.72–0.86) (59/74), 0.81 specificity (29/36) (95% CI 0.73–0.87), and 0.81 AUC (95% CI 0.73–0.87). Radiologists’ reading obtained 0.63 sensitivity (95% CI 0.52–0.74) and 0.78 specificity (95% CI 0.61–0.90) in Centre 1 and 0.64 sensitivity (95% CI 0.52–0.74) and 0.86 specificity (95% CI 0.71–0.95) in Centre 2.ConclusionsThis preliminary experience based on ten CNNs trained on a limited training dataset shows an interesting potential of deep learning for COVID-19 diagnosis. Such tool is in training with new CXRs to further increase its performance.

【 授权许可】

CC BY