Digital Breast Imaging Warehouse for Research and Clinical Decision Support

摘要

Breast imaging is used intensively for breast cancer detection. As routine screening examination becomes more popular for women over 40, tremendous amount of breast imaging data has been accumulated. To diagnose breast abnormalities effectively, radiologists expect convenient access to well- organized breast imaging related information. Most breast cancer related data, however, spread across different systems, making them hard to access. For this project, the investigator will design and build an infra-structural information system by incorporating various kinds of breast imaging data, including patient demographics, family history, radiological reports, digital mammography, ultrasound image and MRI, from a diversity of existing clinical systems, into a digital warehouse. Both clinical- and research-oriented applications will be developed to explore the voluminous amount of data collected in the warehouse. The proposed applications include: (1) case finding, and (2) data mining. Case finding capability virtually transforms the warehouse into a digital case library, and facilitates continuous education and on-the-job decision support, by providing exemplary study to compare with case in question. Data mining application demonstrates the value of consolidating enormous amount of data into a centralized data repository for clinical analysis and decision support. The proposed breast imaging warehouse itself, in the long run, will serve as a reusable platform to foster further breast cancer research and clinical decision making.