Applying Machine Learning Models with An Ensemble Approach for Accurate Real-Time Influenza Forecasting in Taiwan: Development and Validation Study.

BACKGROUND: Changeful seasonal influenza activity in subtropical areas such as Taiwan causes problems in epidemic preparedness. The Taiwan Centers for Disease Control has maintained real-time national influenza surveillance systems since 2004. Except for timely monitoring, epidemic forecasting using the national influenza surveillance data can provide pivotal information for public health response. OBJECTIVE: We aimed to develop predictive models using machine learning to provide real-time influenza-like illness forecasts. METHODS: Using surveillance data of influenza-like illness visits from emergency departments (from the Real-Time Outbreak and Disease Surveillance System), outpatient departments (from the National Health Insurance database), and the records of patients with severe influenza with complications (from the National Notifiable Disease Surveillance System), we developed 4 machine learning models (autoregressive integrated moving average, random forest, support vector regression, and extreme gradient boosting) to produce weekly influenza-like illness predictions for a given week and 3 subsequent weeks. We established a framework of the machine learning models and used an ensemble approach called stacking to integrate these predictions. We trained the models using historical data from 2008-2014. We evaluated their predictive ability during 2015-2017 for each of the 4-week time periods using Pearson correlation, mean absolute percentage error (MAPE), and hit rate of trend prediction. A dashboard website was built to visualize the forecasts, and the results of real-world implementation of this forecasting framework in 2018 were evaluated using the same metrics. RESULTS: All models could accurately predict the timing and magnitudes of the seasonal peaks in the then-current week (nowcast) (ρ=0.802-0.965; MAPE: 5.2%-9.2%; hit rate: 0.577-0.756), 1-week (ρ=0.803-0.918; MAPE: 8.3%-11.8%; hit rate: 0.643-0.747), 2-week (ρ=0.783-0.867; MAPE: 10.1%-15.3%; hit rate: 0.669-0.734), and 3-week forecasts (ρ=0.676-0.801; MAPE: 12.0%-18.9%; hit rate: 0.643-0.786), especially the ensemble model. In real-world implementation in 2018, the forecasting performance was still accurate in nowcasts (ρ=0.875-0.969; MAPE: 5.3%-8.0%; hit rate: 0.582-0.782) and remained satisfactory in 3-week forecasts (ρ=0.721-0.908; MAPE: 7.6%-13.5%; hit rate: 0.596-0.904). CONCLUSIONS: This machine learning and ensemble approach can make accurate, real-time influenza-like illness forecasts for a 4-week period, and thus, facilitate decision making.

An outbreak of HIV infection among people who inject drugs linked to injection of propofol in Taiwan.

INTRODUCTION: The aim of this study was to report an HIV outbreak related to propofol-injection and the impact of regulating propofol on the HIV epidemic among people who inject drugs (PWID). METHODS: A retrospective cohort study of 252 PWID who were diagnosed with an HIV infection between 2014 and 2017 in Taiwan. The propofol information was collected by routine epidemic surveillance and interviews. We linked several national databases to collect other related factors, including methadone maintenance treatment (MMT) attendance and incarceration. The serums were tested for recent infection by the LAg-avidity EIA assay and relationship of the trains by the Phylogenetic tree analysis. Analyses were conducted using the R Surveillance package for retrospective modeling for outbreak detection. A multiple logistic regression was used to evaluate the association between propofol-injection and other related factors. RESULTS: There were 28 cases reported with propofol-injection, all of which were reported in Central Taiwan. A total of 11 (50%) cases among 22 propofol-injectors with serums were recent infections, which were higher than that 33 (23.4%) of non-propofol group. The phylogenetic tree indicated that 6 propofol-injectors were grouped together with the same cluster in circular. The HIV epidemic curve among PWID revealed an outbreak of 82 in 2015, which then decreased to 43 in 2016 after propofol began to be regulated as a Schedule 4 controlled drug in August 2015. In a multiple logistic regression, attendance at methadone clinics was associated with a significantly higher risk for propofol-injection (adjusted OR = 2.43, 95% CI = 0.98-5.98), and HIV reported in the year 2015 was associated with an increased risk of propofol-injection (adjusted OR = 4, 95% CI = 1.08-14.86). CONCLUSIONS: Our data indicate that the government regulation of propofol as a controlled drug strategy was associated with significant reduction in the spread of HIV among PWID.