Abstract
Background: The health crisis resulting from the global COVID-19 pandemic highlighted more than ever the need for rapid, reliable and safe methods of diagnosis and monitoring of respiratory diseases. To study pulmonary involvement in detail, one of the most common resources is the use different lung imaging modalities (like chest radiography) to explore the possible affected areas. Methods: In this work, we performed a comprehensive analysis of sex and age factors in chest X-ray images. The study of these recurrent patient characteristics in pathologies of this type is crucial, since there is a clear scarcity of data that may lead to biases when trying to develop systems that are as representative as possible, as well as to gain knowledge of the disease itself. To identify possible biases, we analyzed 3 different computational approaches for automatic COVID-19 screening: Normal vs COVID-19, Pneumonia vs COVID-19 and Non-COVID-19 vs COVID-19. The presented study was validated using two public chest X-ray datasets, allowing a reliable analysis to support the clinical decision-making process in the context of this dramatic global pandemic. Results: The obtained results for the sex-related imbalance analysis indicate that this factor slightly affects the system performance in the Normal VS COVID-19 and Pneumonia VS COVID-19 approaches, although the identified differences are not relevant enough to worsen considerably the system’s response. Regarding the age-related imbalance analysis, this factor was observed to be again influencing the system in a more consistent way than the sex factor, as it was present in all the approaches. Once again, this worsening is not a major problem for our data and system, as it is not of great magnitude. Conclusions: Multiple studies have been conducted in other fields in order to determine if certain patient characteristics such as sex or age influenced these deep learning systems. However, to the best of our knowledge, this study has not been done for COVID-19 despite the urgency and lack of COVID-19 chest x-ray images. The presented results evidenced that the proposed methodology and tested approaches allow a robust and reliable analysis to support the clinical decision-making process in this pandemic scenario.