Overcoming limitations of breast volume imaging in 3D dedicated mammotomography

Period of Performance: 09/28/2007 - 09/27/2008


Phase 1 SBIR

Recipient Firm

Zumatek, Inc.
Chapel Hill, NC 27514
Principal Investigator


DESCRIPTION (provided by applicant): The emergence of compact, high-quality digital X-ray imaging detectors, together with nearly exponential growth in computing power and corresponding price reductions, allows for the potential of dedicated 3D breast imaging. X-ray mammography (XRM) has recognized limits including relatively high false negative rates. Other disadvantages include breast compression, relatively high dose to the breast, and low positive predictive value, especially for radiographically dense breasts. The dedicated breast Computed mammoTomography (CmT) solution we are proposing has the potential for improved detection of breast lesions through the removal of contrast-reducing overlying tissue, as well as the added benefits of uncompressed breast imaging for greater patient comfort, and breast dose equal to or less than that of current dual view mammography. Ultimately, we expect this technology to result in improved positive predictive value with earlier cancer detection and reduced morbidity and mortality. In the long term, we seek to demonstrate the feasibility of this device in a clinical setting by overcoming the current limitations in imaging the full breast volume and chest wall. These physical limitations are currently imposed by the fact that the X-ray tube and detector must trace an orbital path which includes moving under a patient support. For other "standard" designs, this makes it difficult to image close to the chest wall, though our prototype device has the 3-dimensional positioning capability to potentially overcome some of these limitations. Specific Aim 1 will investigate of partial angle tomographic scanning using the patient bed to move the breast further into the field of view to capture the entire breast and chest wall while limiting the orbit of the device to the sides of the bed. The assembly will be placed under a patient bed at the Multi-Modality Imaging Lab (MMIL) at Duke University Medical Center and partial angle scans made of available torso, breast, and lesion phantoms. Observers will determine the extent of visibility of lesions close to the chest wall. Specific Aim 2 will demonstrate the feasibility of utilizing current patient bed vertical elevation capabilities to change the relationship between breast and system center of rotation during scans to provide full 360 degree acquisitions while avoiding patient bed limitations. A patient bed with multiple degrees of freedom of movement will be used to lower the patient phantom further into the field of view while the device is perpendicular to the bed axis, then raise the bed to allow to device to move under the bed. Results will be evaluated as in Specific Aim 1. Specific Aim 3 will model patient bed designs to determine the degree to which the chest wall can be moved into the field of view with reasonable patient comfort by potentially utilizing smooth bed contours around a patient's hips and head. Successful implementation of the above alternatives, pending results of Phase I, will result in a next generation advanced system being produced in Phase II that can be placed in a clinical setting for the beginning of patient studies, leading to a commercially available product.