Office of Technology Transfer – University of Michigan

Detection of Breast Cancer on Digital Tomosynthesis Mammograms

Technology #4576

Questions about this technology? Ask a Technology Manager

Download Printable PDF

Categories
Researchers
Heang-Ping Chan
Managed By
Drew Bennett
Associate Director - Software Licensing 734-615-4004
Patent Protection
US Patent Pending
US Patent Pending

Background

Mammography is the most cost-effective screening method for early detection of breast cancer; however, mammographic sensitivity is often limited by the presence of overlapping dense fibroglandular tissue in the breast. Breast cancer digital tomosynthesis mammography (DTM) is a new modality in which a series of projection view (PV) mammograms of the breast is acquired as the x-ray source is moved about the breast. Subsequently, the breast volume is reconstructed as thin tomographic slice images using a reconstruction algorithm. DTM thus has the potential to reduce the effect of overlapping tissue and improve detection of masses. One of the current challenges with DTM is the detection of subtle microcalcification clusters, which may be obscured as the cluster is split into different DTM slices. As a result of multiple-view reconstruction, sharpness of individual microcalcifications is reduced, thereby lowering the conspicuity of clusters on DTM slices.

Technology

Researchers at the University of Michigan have developed a computer-aided detection (CAD) system that will provide a second opinion to radiologists during their interpretation of DTMs. The system is designed to perform microcalcification detection in the 3-dimensional (3D) volume, using DTM slices generated from the PVs by any reconstruction methods. Additional microcalcification features are extracted from the PV images and combined with the 3D information for differentiating true and false microcalcifications. The design of the specific combination of techniques can achieve high accuracy in detecting clustered microcalcifications in DTM images. The features of the potential clusters may further be analyzed using the 3D image information extracted from the DTM volumes and 2D image information extracted from the PV images before reconstruction. The 3D or the combined 3D/ 2D analysis can provide a microcalcification likelihood score for each potential cluster, identifying the most suspicious clusters.

Applications and Advantages

Applications

  • Detection of clustered microcalcifications in DTM images.

Advantages

  • High accuracy, which may result in earlier diagnosis breast cancer.