User identity validation is particularly relevant for applications where data privacy is critical, such as Healthcare Information Systems (HIS), where patient records protection and medical acts traceability is extremely important. Current approaches to the problem include biometric solutions, however, traditional modalities only allow momentary verification; readers are generally fixed to a static location, and direct contact or proximity is required. State-of-the-art work has been focusing solutions for continuous, or more frequent assessment in an unobtrusive way. In this paper we present a framework for continuous identity verification, based on knowledge discovery from ECG signals for security enhancement in the HIS context. ECG signals are particularly convenient, as they are frequently already measured in patients, and can also be easily obtained from caregivers interacting with the information system.