Search

BackgroundSymptomatic dengue infection can result in a life-threatening shock syndrome and timely diagnosis is essential. Point-of-care tests for non-structural protein 1 and IgM are used widely but performance can be limited. We developed a supervised machine learning model to predict whether patients with acute febrile illnesses had a diagnosis of dengue or other febrile illnesses (OFI). The impact of seasonality on model performance over time was examined.MethodsWe analysed data from a prospective observational clinical study in Vietnam. Enrolled patients presented with an acute febrile illness of 90%).ConclusionSupervised machine learning models are able to discriminate between dengue and OFI diagnoses in patients presenting with an early undifferentiated febrile illness. These models could be of clinical utility in supporting healthcare decision-making and provide passive surveillance across dengue endemic regions. Effects of seasonality and changing disease prevalence must however be taken into account—this is of significant importance given unpredictable effects of human-induced climate change and the impact on health.

Abstract Background Minimally invasive mitral valve replacement has become popular across the world. However, annular rupture and patient – prosthetic mismatch (PPM) are still problematic, particularly in the Asian population. To avoid this, a predictor model could be beneficial. Our study aimed to assess the value of mitral valve diameters measured on TTE and CT scan on predicting the actual mitral prostheses. Methods From January 2018 to December 2019, a total number of 96 patients underwent minimally invasive mitral valve replacement. The association between imaging measurements and the outcome was checked by scatter plot and Pearson’s correlation coefficient. Univariable linear regression was used to build the prediction model. Results The three strongest correlations for the whole population are the following features: Mean TTE diameter (0.702), mean diameter on CT lowest plane through the mitral annulus (0.679), and area-derived diameter on CT highest plane through the mitral annulus (0.665). The prosthetic size of the tissue valve group was more correlated to the calculated annulus diameters than that of the mechanical valve group. Tissue valve size predictor models based on these calculated diameters were 16.19 + 0.27 × d (r = 0.744), 12.74 + 0.44 × d (r = 0.756) and 12.79 + 0.38 × d (r = 0.730), respectively. Conclusion Mitral prosthetic size could be predicted based on the mitral diameters measured on TTE and CT scan. The overall correlation coefficient varied from 0.665 (CT Scan) to 0.702 (TTE). These models performed better when applied to bioprosthesis.