Background: The coronavirus disease (COVID-19) poses an urgent threat to global public health and is characterized by rapid disease progression even in mild cases. In this study, we investigated whether machine learning can be used to predict which patients will have a deteriorated condition and require oxygenation in asymptomatic or mild cases of COVID-19. Methods: This single-center, retrospective, observational study included COVID-19 patients admitted to the hospital from February 1, 2020, to May 31, 2020, and who were either asymptomatic or presented with mild symptoms and did not require oxygen support on admission. Data on patient characteristics and vital signs were collected upon admission. We used seven machine learning algorithms, assessed their capability to predict exacerbation, and analyzed important influencing features using the best algorithm. Results: In total, 210 patients were included in the study. Among them, 43 (19%) required oxygen therapy. Of all the models, the logistic regression model had the highest accuracy and precision. Logistic regression analysis showed that the model had an accuracy of 0.900, precision of 0.893, and recall of 0.605. The most important parameter for predictive capability was SpO2, followed by age, respiratory rate, and systolic blood pressure. Conclusion: In this study, we developed a machine learning model that can be used as a triage tool by clinicians to detect high-risk patients and disease progression earlier. Prospective validation studies are needed to verify the application of the tool in clinical practice.
The objectives of our study were : (1) To investigate the feasibility of using the digital KFA 1000 imagery analysis as a means of accurately describing and quantifying soil properties of Andosols, and (2) to reveal the relationship between the soil properties and the extent of clubroot disease incidence in cabbage. The film (Sojuzkarta #25253, scale 1 : 250, 000), acquired on June 7, 1988, was used to examine the approach in the entire area of Tsumakoi-mura of Gunma Prefecture. The film is sensitive in the red (570-680 nm) and near infrared (680-810 nm) regions. The scanner data were converted from analog to digital from by a drum scanner with a 25-μm sampling pitch. This method provided data with a pixel size of approximately 6 m×6 m. The near infrared reflectance was a better indicator for delineating five soil series than the red reflectance, indicating the differences of soil color, organic matter content, and water regime among the five soil series. The large variations in the red reflectance within a single soil series may be influenced by different soil moisture conditions ; five soil series and three drainage classes were identified. The effect of soil series on the retardation of the disease incidence was much greater in Light-colored Andosols than in other types. The disease was more severe in the poorly drained class. A season of abundant rainfall was shown to increase the incidence of the disease. Sojuzkarta KFA 1000 imagery provides good information on soil properties of Andosols for the assessment of the incidence of clubroot disease.
