The current COVID-19 pandemic, that has caused more than 100 million cases as well as more than two million deaths worldwide, demands the development of fast and accurate diagnostic methods despite the lack of available samples. This disease mainly affects the respiratory system of the patients and can lead to pneumonia and to severe cases of acute respiratory syndrome that result in the formation of several pathological structures in the lungs. These pathological structures can be explored taking advantage of chest X-ray imaging. As a recommendation for the health services, portable chest X-ray devices should be used instead of conventional fixed machinery, in order to prevent the spread of the pathogen. However, portable devices present several problems (specially those related with capture quality). Moreover, the subjectivity and the fatigue of the clinicians lead to a very difficult diagnostic process. To overcome that, computer-aided methodologies can be very useful even taking into account the lack of available samples that the COVID-19 affectation shows. In this work, we propose an improvement in the performance of COVID-19 screening, taking advantage of several cycle generative adversarial networks to generate useful and relevant synthetic images to solve the lack of COVID-19 samples, in the context of poor quality and low detail datasets obtained from portable devices. For validating this proposal for improved COVID-19 screening, several experiments were conducted. The results demonstrate that this data augmentation strategy improves the performance of a previous COVID-19 screening proposal, achieving an accuracy of 98.61% when distinguishing among NON-COVID-19 (i.e. normal control samples and samples with pathologies others than COVID-19) and genuine COVID-19 samples. It is remarkable that this methodology can be extrapolated to other pulmonary pathologies and even other medical imaging domains to overcome the data scarcity.
Respiratory diseases have a significant global impact, and assessing these conditions is crucial for improving patient outcomes. Chest X-ray is widely used for diagnosis, but expert evaluation can be challenging. Automatic computer-aided diagnosis methods can provide support for clinicians in these tasks. Deep learning has emerged as a set of algorithms with exceptional potential in such tasks. However, these algorithms require a vast amount of data, often scarce in medical imaging domains. In this work, a new data augmentation methodology based on adapted generative latent diffusion models is proposed to improve the performance of an automatic pathological screening in two high-impact scenarios: tuberculosis and lung nodules. The methodology is evaluated using three publicly available datasets, representative of real-world settings. An ablation study obtained the highest-performing image generation model configuration regarding the number of training steps. The results demonstrate that the novel set of generated images can improve the performance of the screening of these two highly relevant pathologies, obtaining an accuracy of 97.09%, 92.14% in each dataset of tuberculosis screening, respectively, and 82.19% in lung nodules. The proposal notably improves on previous image generation methods for data augmentation, highlighting the importance of the contribution in these critical public health challenges. Graphical
Coronavirus Disease 2019 (COVID-19), declared a global pandemic by the World Health Organization, mainly affects the pulmonary tissues, playing chest X-ray images an important role for its screening and early detection. In this context, portable X-ray devices are widely used, representing an alternative to fixed devices in order to reduce risks of cross-contamination. However, they provide lower quality and detailed images in terms of spatial resolution and contrast. In this work, given the low availability of images of this recent disease, we present new approaches to artificially increase the dimensionality of portable chest X-ray datasets for COVID-19 diagnosis. Hence, we combined 3 complementary CycleGAN architectures to perform a simultaneous oversampling using an unsupervised strategy and without the necessity of paired data. Despite the poor quality of the portable X-ray images, we provide an overall accuracy of 92.50% in a COVID-19 screening context, proving their suitability for COVID-19 diagnostic tasks.
22 December 2022 in Computers in Biology and Medicine
During the last years, deep learning techniques have emerged as powerful alternatives to solve biomedical image analysis problems. However, the training of deep neural networks usually needs great amounts of labeled data to be done effectively. This is even more critical in the case of biomedical imaging due to the added difficulty of obtaining data labeled by experienced clinicians. To mitigate the impact of data scarcity, one of the most commonly used strategies is transfer learning. Nevertheless, the success of this approach depends on the effectiveness of the available pre-training techniques for learning from little or no labeled data. In this work, we explore the application of the Context Encoder paradigm for transfer learning in the domain of retinal image analysis. To this aim, we propose several approaches that allow to work with full resolution images and improve the recognition of the retinal structures. In order to validate the proposals, the Context Encoder pre-trained models are fine-tuned to perform two relevant tasks in the domain: vessels segmentation and fovea localization. The experiments performed on different public datasets demonstrate that the proposed Context Encoder approaches allow mitigating the impact of data scarcity, being superior to previous alternatives in this domain.
COVID-19 is an infectious disease that caused a global pandemic in 2020. In the critical moments of this healthcare emergencies, the medical staff needs to take important decisions in a context of limited resources that must be carefully managed. To this end, the computer-aided diagnosis methods are extremely powerful and help them to better recognize the evidences of high-risk patients. This can be done with the support of relevant information extracted from electronic health records, lab tests and imaging studies. In this work, we present a novel fully-automatic efficient method to help the clinical decision-making process in the context of COVID-19 risk estimation, using multimodal data fusion of clinical features and deep features extracted from chest X-ray images. The risk estimation is studied in two of the most relevant and critical encountered scenarios: the risk of hospitalization and mortality. This study shows which are the most important features for each scenario, the ratio of clinical and imaging features present in the top ranking and the performance of the used machine learning models. The results demonstrate a great performance by the classifiers, estimating the risk of hospitalization with an AUC-ROC of 0.8452 ± 0.0133 and the risk of death with an AUC-ROC of 0.8285 ± 0.0210, only using a subset of the original features, and highlight the significant contribution of imaging features to hospitalization risk assessment, while clinical features become more crucial for mortality risk evaluation. Furthermore, multimodal data fusion can outperform the approaches that use one data source. Despite the model's complexity, it requires fewer features, an advantage in scenarios with limited computational resources. This streamlined, fully-automated method shows promising potential to improve the clinical decision-making process and better manage medical resources, not only in the context of COVID-19, but also in other clinical scenarios.