To fully understand results derived from genetic research, a patient’s genotype data must be integrated with other information about the individual (vital signs, height/weight, lab values, disease history – the phenotype of the patient) that can be obtained through clinical records. Within the clinical and translational sciences awards (CTSA), significant effort has been supported to expand translational research through the creation and mining of a phenotypic data warehouse (i2b2) that can be further linked to genotype data. However, this is just a first step towards meaningful use of the available information. Much of the information in electronic clinical records is trapped in unstructured free text, and inaccessible. Transforming this information into usable data has great potential to improve personalized healthcare and enhance the scientific enterprise. We are using “collaborative tagging,” a newer web 2.0 phenomenon used to structure information for accessibility online in which groups of individuals can add any word or phrase as a tag to identify an object (a weblog entry, a picture, etc.). Because taggers create whatever they deem as the most important tags, and are not required to select from a complex tree of predetermined tags, folksonomies can be a more palatable form of data entry than selected from a complex, predetermined list of tags. We will present our results on tagging of clinical notes by UMMS and community-based providers. This presentation was part of the retreat mini-symposium entitled: Data-Driven Approaches for Health Informatics.

Presentation on the development of a recommender system for a computer-tailored health communications tool that assists with helping tobacco users to quit smoking. This presentation was part of the retreat mini-symposium entitled: Smartphones, Sensors, and Social Networks: The New Tools of Health Behavior Change.