Developing a comprehensive database management system for organization and evaluation of mammography datasets

Area of Science:

• Radiology and Medical Imaging
• Biomedical Informatics
• Health Services Research

Background:

• Mammography practice generates complex, multi-relational data crucial for outcome assessment.
• Existing systems may not adequately support the detailed data requirements for quality control and research in breast imaging.
• Development of decision support tools requires robust and comprehensive clinical datasets.

Purpose of the Study:

• To design and develop a comprehensive mammography database system (CMDB).
• To create a HIPAA-compliant system for collecting multi-relational datasets.
• To support outcome assessment, quality assurance, and the development of decision support tools in mammography.

Main Methods:

• Developed a Health Insurance Portability and Accountability Act (HIPAA) compliant database system.
• Integrated demographic risk factors, mammogram results (using Breast Imaging Reporting and Data System - BI-RADS lexicon), and biopsy pathology outcomes.
• Extended the BI-RADS lexicon to compile a breast pathology lexicon and incorporated institutional breast cancer registry data.

Main Results:

• The CMDB successfully stores multi-relational datasets, managing the complexity of mammography practice.
• Audit results derived from the CMDB align with Mammography Quality Standards Act (MQSA) standards and national benchmarks.
• The system demonstrates capability in handling lexicon development for pathology within breast imaging.

Conclusions:

• The developed CMDB provides a framework for a detailed data repository for breast imaging quality control and research.
• The system is anticipated to enhance existing resources and augment current capabilities in breast cancer diagnosis.
• The CMDB is expected to be valuable for efficient quality assurance audits and the development of improved diagnostic decision support tools.