With the spread of COVID-19 pandemic worldwide, medical imaging modalities and deep learning can play an important role in the fight against this disease. Recent years have seen the impressive results obtained using deep neural networks in different fields. Radiology is among the medical fields that can benefit from this recent progress and improve disease’s diagnosis, monitoring and prognosis. In this work, we propose the use of a deep efficient neural network based on EfficientNet B7 to detect COVID-19 in Chest X-rays (CXR). The obtained results on a large dataset are promising and show the high performance of the proposed model, with in average an accuracy of 95%, an AUC of 95%, a specificity of 90% and a sensitivity of 97%. In addition, an explainability model was developed and shows the high performance and high generalization degree of the proposed model in detecting and highlighting the signs of the disease. © 2021, Springer Nature Switzerland AG.
[1]
Phillip M Cheng,et al.
Tackling the Radiological Society of North America Pneumonia Detection Challenge.
,
2019,
AJR. American journal of roentgenology.
[2]
Moulay A. Akhloufi,et al.
Deep Retinal Diseases Detection and Explainability Using OCT Images
,
2020,
ICIAR.
[3]
Ioannis D. Apostolopoulos,et al.
Covid-19: automatic detection from X-ray images utilizing transfer learning with convolutional neural networks
,
2020,
Physical and Engineering Sciences in Medicine.
[4]
Pedro Henriques Abreu,et al.
Using deep learning techniques in medical imaging: a systematic review of applications on CT and PET
,
2019,
Artificial Intelligence Review.
[5]
M. Mildner,et al.
Re-epithelialization and immune cell behaviour in an ex vivo human skin model
,
2020,
Scientific Reports.