Random forest machine learning algorithm predicts virologic outcomes among HIV infected adults in Lausanne, Switzerland using electronically monitored combined antiretroviral treatment adherence.

Machine Learning (ML) can improve the analysis of complex and interrelated factors that place adherent people at risk of viral rebound. Our aim was to build ML model to predict RNA viral rebound from medication adherence and clinical data. Patients were followed up at the Swiss interprofessional medication adherence program (IMAP). Sociodemographic and clinical variables were retrieved from the Swiss HIV Cohort Study (SHCS). Daily electronic medication adherence between 2008-2016 were analyzed retrospectively. Predictor variables included: RNA viral load (VL), CD4 count, duration of ART, and adherence. Random Forest, was used with 10 fold cross validation to predict the RNA class for each data observation. Classification accuracy metrics were calculated for each of the 10-fold cross validation holdout datasets. The values for each range from 0 to 1 (better accuracy). 383 HIV+ patients, 56% male, 52% white, median (Q1, Q3): age 43 (36, 50), duration of electronic monitoring of adherence 564 (200, 1333) days, CD4 count 406 (209, 533) cells/mm3, time since HIV diagnosis was 8.4 (4, 13.5) years, were included. Average model classification accuracy metrics (AUC and F1) for RNA VL were 0.6465 and 0.7772, respectively. In conclusion, combining adherence with other clinical predictors improve predictions of RNA.

Using Search Engine Data as a Tool to Predict Syphilis.

BACKGROUND: Researchers have suggested that social media and online search data might be used to monitor and predict syphilis and other sexually transmitted diseases. Because people at risk for syphilis might seek sexual health and risk-related information on the internet, we investigated associations between internet state-level search query data (e.g., Google Trends) and reported weekly syphilis cases. METHODS: We obtained weekly counts of reported primary and secondary syphilis for 50 states from 2012 to 2014 from the US Centers for Disease Control and Prevention. We collected weekly internet search query data regarding 25 risk-related keywords from 2012 to 2014 for 50 states using Google Trends. We joined 155 weeks of Google Trends data with 1-week lag to weekly syphilis data for a total of 7750 data points. Using the least absolute shrinkage and selection operator, we trained three linear mixed models on the first 10 weeks of each year. We validated models for 2012 and 2014 for the following 52 weeks and the 2014 model for the following 42 weeks. RESULTS: The models, consisting of different sets of keyword predictors for each year, accurately predicted 144 weeks of primary and secondary syphilis counts for each state, with an overall average R of 0.9 and overall average root mean squared error of 4.9. CONCLUSIONS: We used Google Trends search data from the prior week to predict cases of syphilis in the following weeks for each state. Further research could explore how search data could be integrated into public health monitoring systems.