University of Pittsburgh
Pittsburgh, Pennsylvania
Sujin Kim

Sujin Kim MS, Rajnish Gupta MS, John Gilbertson, MD
Division of Informatics, Department of Pathology, University of Pittsburgh, Pittsburgh, PA

We have begun the design and implement an advanced prostate tissue banking information system (TBIS) to support the growing demand for highly qualified tissue samples by researchers and industry. Recent advances in genomic and proteomics research has created an unfilled demand for banked tissues that are highly annotated, imaged and pre-analyzed at a sub-specimen level. Existing systems cannot handle this level of complexity. Our next generation tissue banking information system that would meet the needs of medical researchers by capturing demographic, clinical, pathologic, outcomes, morphologic (image) and gene activity (microarray) information from existing clinical information systems, and allow researchers to search for and visualize high annotated and imaged tissues.

The system development is envisioned in three phases. First a textual database will be implemented to manage contact, demographic, epidemoloigic, clinical, pathologic and outcomes data will be designed and implemented in ORACLE. Second high resolution, whole slide images of all blocks will be added using the Interscope imaging system. Finally, gene expression data in the form of micro-array results will be integrated with the textual and morphologic data.

This year will present the textual database. The system is based on an ORACLE 8I database with Developer Forms. The system supports demographic, epidemiologic, clinical, treatment and outcomes data on each patient as well as pathology tissue data down to the block level. All data is structured, enumerated and reviewed by a tissue bank pathologist. The system also supports a separate datawarehouse, on a separate machine, that contains a subset of de-identified tissue data that is available to researcher through the WWW. An interface to the NCI’s CPCTR project is planned.

The system is implemented in the UPMC Pathology Tissue Bank where it is replacing to earlier databases. Over the next year the system will be developed to support tissue banks at other institutions as part of Oracle Health Care’s Cancer Care product as well as a stand alone application.

This work is sponsored in part by CaPCURE.

Keywords: Tissue Bank, Database, Pathology, Bio-informatics