Syndromic Surveillance through Measuring Lexical Shift in Emergency Department Chief Complaint Texts

Syndromic Surveillance has been performed using machine learning and other statistical methods to detect disease outbreaks. These methods are largely dependent on the availability of historical data to train the machine learning-based surveillance system. However, relevant training data may differ from region to region due to geographical and seasonal trends, meaning that the syndromic surveillance designed for one area may not be effective for another. We proposed and analyse a semi-supervised method for syndromic surveillance from emergency department chief complaint textual notes that avoids the need for large training data. Our new method is based on identification of lexical shifts in the language of Chief Complaints of patients, as recorded by triage nurses, that we believe can be used to monitor disease distributions and possible outbreaks over time. The results we obtained demonstrate that effective lexical syndromic surveillance can be approached when distinctive lexical items are available to describe specific syndromes.