Case-based Retrieval in Mammography

One of the most powerful tools for decision support is the ability to compare a case with a database of similar disease and patient patterns. The database must contain consolidated multi-modal diagnostic tests paired with patient information (like risk profile, disease history, genetic profile) and thus serves as a knowledge base.

Breast cancer detection and diagnosis systems are usually developed for one data modality only, for example, x-ray mammography, mammasonography or MR-mammography. Each of these modalities has advantages and disadvantages concerning the specificity for breast cancer detection. Radiologists need to consult different modalities, analyze them and correlate findings in order to achieve the greatest accuracy in the diagnosis of breast cancer. Although recent studies endorse multimodal breast diagnosis and screenings, there are currently no CAD systems which allow computer aided detection and diagnosis by analyzing different mammography data modalities at once.

Definiens has a new approach for automated multimodal CAD systems delivering accurate and consistent results. Applied in a research study, a wide variety of decisive morphological parameters can be measured and exported including spectral statistics, shape, size, position, texture and relations to neighbor objects.

For further details, please see the poster Definiens Cognition Network Technology® for information centric healthcare in mammography.