Family History Detection from Clinical Text

Electronic Health Record (EHR) systems are becoming highly prevalent today and have the potential to add to the success of spontaneous reporting systems for post marketing surveillance of drugs. EHR systems broadly contain two forms of information viz. coded (structured) information and unstructured text. Researchers have shown that the textual data have much more utility for cohort building than coded information, which, being mainly used for billing and insurance purposes, is biased and does not truly reflect the state of the patient.