Identification, analysis and prediction of valid and false information related to vaccines from Romanian tweets.

Introduction: The online misinformation might undermine the vaccination efforts. Therefore, given the fact that no study specifically analyzed online vaccine related content written in Romanian, the main objective of the study was to detect and evaluate tweets related to vaccines and written in Romanian language. Methods: 1,400 Romanian vaccine related tweets were manually classified in true, neutral and fake information and analyzed based on wordcloud representations, a correlation analysis between the three classes and specific tweet characteristics and the validation of several predictive machine learning algorithms. Results and discussion: The tweets annotated as misinformation showed specific word patterns and were liked and reshared more often as compared to the true and neutral ones. The validation of the machine learning algorithms yielded enhanced results in terms of Area Under the Receiver Operating Characteristic Curve Score (0.744-0.843) when evaluating the Support Vector Classifier. The predictive model estimates in a well calibrated manner the probability that a specific Twitter post is true, neutral or fake. The current study offers important insights regarding vaccine related online content written in an Eastern European language. Future studies must aim at building an online platform for rapid identification of vaccine misinformation and raising awareness for the general population.

The development of a scoring and ranking strategy for a patient-tailored adverse drug reaction prediction in polypharmacy.

Only few applications are currently dealing with personalized adverse drug reactions (ADRs) prediction in case of polypharmacy. The study aimed to develop a patient-tailored ADR web application, considering characteristics from 734 drugs and relevant patient related factors. The application was designed in Python using a scoring and ranking system based on frequency and severity, computed for each ADR and expressed through an online platform. A neural networks algorithm was used for predicting the severity of ADRs. The application inputs are: age, gender, drugs, relevant pathologies. The outputs are: an overall severity profile (hospitalization and mortality risk), a stratified risk on specific ADR groups and a sorted list of the most important ADRs depending on frequency and severity. The Severity prediction model validation resulted in 79.7-85.1% Area Under the Receiver Operating Characteristic Curve Score, which lies in the good cut-off of 75-90%. The program offers a complex view regarding the ADR profile of a given patient and could be used by the physician and clinical pharmacist during patient safety monitoring, for a coherent therapy choice or medication adjustment, due to the good therapy coverage and the inclusion of relevant patient comorbidities.