Abstract
In the mature field of breast imaging CAD, case-based systems are a relative newcomer. The increased interest in this technology does not really stem from its superiority regarding clinical performance but rather from its ability to provide decision support that emulates the evidence based medicine paradigm. The “reasoning by remembering prior cases” approach followed by case-based CAD systems is not only consistent with human cognition but it easily takes advantage of the digital image data that accumulate rapidly in modern clinical practice. There are however many scientific issues with respect to the development of case-based CAD technology in breast imaging. Challenges regarding knowledge representation, completeness of the knowledge base, the semantic gap between visual and diagnostic relevance, as well as computational efficiency are all active research topics. Furthermore, whether this technology offers clear benefits to the end users over conventional CAD systems is still under investigation. The chapter provides an up-to-date review of the case-based CAD research in breast imaging.