Search

The COVID-19 pandemic shed light on the need for quick diagnosis tools in healthcare, leading to the development of several algorithmic models for disease detection. Though these models are relatively easy to build, their training requires a lot of data, storage, and resources, which may not be available for use by medical institutions or could be beyond the skillset of the people who most need these tools. This paper describes a data analysis and machine learning platform that takes advantage of high-performance computing infrastructure for medical diagnosis support applications. This platform is validated by re-training a previously published deep learning model (COVID-Net) on new data, where it is shown that the performance of the model is improved through large-scale hyperparameter optimisation that uncovered optimal training parameter combinations. The per-class accuracy of the model, especially for COVID-19 and pneumonia, is higher when using the tuned hyperparameters (healthy: 96.5%; pneumonia: 61.5%; COVID-19: 78.9%) as opposed to parameters chosen through traditional methods (healthy: 93.6%; pneumonia: 46.1%; COVID-19: 76.3%). Furthermore, training speed-up analysis shows a major decrease in training time as resources increase, from 207 min using 1 node to 54 min when distributed over 32 nodes, but highlights the presence of a cut-off point where the communication overhead begins to affect performance. The developed platform is intended to provide the medical field with a technical environment for developing novel portable artificial-intelligence-based tools for diagnosis support. [ABSTRACT FROM AUTHOR]

Background: The influenza virus and the novel beta coronavirus (SARS-CoV-2) have similar transmission characteristics, and it is very difficult to distinguish them clinically. With the development of information technologies, novel opportunities have arisen for the application of intelligent software systems in disease diagnosis and patient triage. Methods: A cross-sectional study was conducted on 268 infants: 133 infants with a SARS-CoV-2 infection and 135 infants with an influenza virus infection. In total, 10 hematochemical variables were used to construct an automated machine learning model. Results: An accuracy range from 53.8% to 60.7% was obtained by applying support vector machine, random forest, k-nearest neighbors, logistic regression, and neural network models. Alternatively, an automated model convincingly outperformed other models with an accuracy of 98.4%. The proposed automated algorithm recommended a random tree model, a randomization-based ensemble method, as the most appropriate for the given dataset. Conclusions: The application of automated machine learning in clinical practice can contribute to more objective, accurate, and rapid diagnosis of SARS-CoV-2 and influenza virus infections in children. [ABSTRACT FROM AUTHOR]