Subject(s)
Aggregatibacter actinomycetemcomitans , Endocarditis, Bacterial , Endophthalmitis , Pasteurellaceae Infections , Humans , Endophthalmitis/microbiology , Endophthalmitis/diagnosis , Endophthalmitis/etiology , Endocarditis, Bacterial/microbiology , Endocarditis, Bacterial/complications , Endocarditis, Bacterial/diagnosis , Pasteurellaceae Infections/diagnosis , Pasteurellaceae Infections/complications , Pasteurellaceae Infections/microbiology , Aggregatibacter actinomycetemcomitans/isolation & purification , Male , Female , Eye Infections, Bacterial/diagnosis , Eye Infections, Bacterial/microbiologyABSTRACT
Brain Imaging Data Structure (BIDS) provides a valuable tool to organise brain imaging data into a clear and easy standard directory structure. Moreover, BIDS is widely supported by the scientific community and has been established as a powerful standard for medical imaging management. Nonetheless, the original BIDS is restricted to magnetic resonance imaging (MRI) of the brain, limiting its implantation to other techniques and anatomical regions. We developed Medical Imaging Data Structure (MIDS), conceived to extend BIDS methodology to other anatomical regions and multiple imaging systems in these areas. The MIDS standard was developed to store and manage medical images as an extension of BIDS. It allows the user to handily save studies of multiple anatomical regions and imaging techniques. Besides, MIDS improves the classification of multiple images within the structure, allowing the possibility to unify them in a single study to apply on them preprocessing or artificial intelligence algorithms. Finally, the results generated are saved in the derivatives folder.
Subject(s)
Artificial Intelligence , Brain , Algorithms , Brain/diagnostic imaging , Magnetic Resonance Imaging/methodsABSTRACT
BACKGROUND: The role of chest radiography in COVID-19 disease has changed since the beginning of the pandemic from a diagnostic tool when microbiological resources were scarce to a different one focused on detecting and monitoring COVID-19 lung involvement. Using chest radiographs, early detection of the disease is still helpful in resource-poor environments. However, the sensitivity of a chest radiograph for diagnosing COVID-19 is modest, even for expert radiologists. In this paper, the performance of a deep learning algorithm on the first clinical encounter is evaluated and compared with a group of radiologists with different years of experience. METHODS: The algorithm uses an ensemble of four deep convolutional networks, Ensemble4Covid, trained to detect COVID-19 on frontal chest radiographs. The algorithm was tested using images from the first clinical encounter of positive and negative cases. Its performance was compared with five radiologists on a smaller test subset of patients. The algorithm's performance was also validated using the public dataset COVIDx. RESULTS: Compared to the consensus of five radiologists, the Ensemble4Covid model achieved an AUC of 0.85, whereas the radiologists achieved an AUC of 0.71. Compared with other state-of-the-art models, the performance of a single model of our ensemble achieved nonsignificant differences in the public dataset COVIDx. CONCLUSION: The results show that the use of images from the first clinical encounter significantly drops the detection performance of COVID-19. The performance of our Ensemble4Covid under these challenging conditions is considerably higher compared to a consensus of five radiologists. Artificial intelligence can be used for the fast diagnosis of COVID-19.
ABSTRACT
Brain Imaging Data Structure (BIDS) provides a valuable tool to organise brain imaging data into a clear and easy standard directory structure. Moreover, BIDS is widely supported by the scientific community and has been established as a powerful standard for medical imaging management. Nonetheless, the original BIDS is restricted to magnetic resonance imaging (MRI) of the brain, limiting its implantation to other techniques and anatomical regions. We developed Medical Imaging Data Structure (MIDS), conceived to extend BIDS methodology to other anatomical regions and multiple imaging systems in these areas. The MIDS standard was developed to store and manage medical images as an extension of BIDS. It allows the user to handily save studies of multiple anatomical regions and imaging techniques. Besides, MIDS improves the classification of multiple images within the structure, allowing the possibility to unify them in a single study to apply on them preprocessing or artificial intelligence algorithms. Finally, the results generated are saved in the derivatives folder.
