Public health surveillance is challenging due to difficulties accessing medical data in real-time. We present a novel, effective and computationally inexpensive method for syndromic surveillance using Twitter data. The proposed method uses a regression model on a database previously built using named entity recognition to identify mentions of symptoms, disorders and pharmacological substances over GNIP Decahose Twitter data. The result of our method is compared to the reported weekly flu and Lyme disease rates from the US Center of Disease Control and Prevention (CDC) website. Our method predicts the 2014 CDC reported flu prevalence with 94.9% Spearman correlation using 2012 and 2013 CDC flu statistics as training data, and the CDC Lyme disease rate for July to December 2014 with 89.6% Spearman correlation. It also predicts the prevalences for the same diseases and time periods using the Twitter data from the previous week with 93.31% and 86.9% Spearman correlations respectively.