Clinicians have unique insight into the diseases and medical conditions they treat, and may develop their own hypotheses they wish to explore by examining existing cases in a data warehouse. To facilitate manual data mining by clinicians and scientists, we have developed an interface for our clinical data warehouse, the Aggregated Biomedical-information Browser (ABB), based on OLAP (On-Line Analytical Processing) technology. The ABB enables clinicians, researchers, and other domain experts to quickly and intuitively explore data in our data warehouse, the Data Warehouse for Translational Research (DW4TR), without needing to involve informatics staff for data extraction. The ABB is capable of handling “on the fly” queries of any data element within the DW4TR. This functionality enables researchers to use their domain knowledge to connect disparate data points as one discovery leads to another. Hypotheses generated through manual data mining combined with domain knowledge, can then be tested using more advanced statistical methods. To illustrate this process a manual data mining example comparing breast cancer pathology in African American and Caucasian American women is performed using the ABB. Analysis of several breast cancer pathology markers suggest African American women will have a worse clinical outcome than Caucasian American women, a clinically meaningful outcome well documented in scientific literature. This report demonstrates the simple yet powerful use of the ABB for manual data exploration in the initial hypothesis generation stage.
Citation: Maskery S, Bekhash A, Kvecher L, Correll M, Hooke JA, et al. (2014) Aggregated Biomedical-Information Browser (ABB): A Graphical User Interface for Clinicians and Scientists to Access a Clinical Data Warehouse. J Comput Sci Syst Biol 7:020-027. doi: 10.4172/jcsb.1000134