Computed Tomography (CT) is the gold standard for image evaluation of lung disease, including lung cancer and cystic fibrosis. It provides detailed information of the lung anatomy and lesions, but at a relatively high cost and high dose of
radiation. Chest radiography is a low dose imaging modality but it has low sensitivity. Digital chest tomosynthesis (DCT) is an imaging modality that produces 3D images by collecting x-ray projection images over a limited angle. DCT is less expensive than CT and requires about 1/10th the dose of radiation. Commercial DCT systems acquire the
projection images by mechanically scanning an x-ray tube. The movement of the tube head limits acquisition speed. We recently demonstrated the feasibility of stationary digital chest tomosynthesis (s-DCT) using a carbon nanotube
(CNT) x-ray source array in benchtop phantom studies. The stationary x-ray source allows for fast image acquisition. The objective of this study is to demonstrate the feasibility of s-DCT for patient imaging. We have successfully imaged 31 patients. Preliminary evaluation by board certified radiologists suggests good depiction of thoracic anatomy and pathology.
Digital tomosynthesis is a type of limited angle tomography that allows 3D information to be reconstructed
from a set of x-ray projection images taken at various angles using an x-ray tube, a mechanical arm to rotate the tube
about the object, and a digital detector. Tomosynthesis reconstruction requires the precise location of the detector with
respect to each x-ray source, forcing all current clinical tomosynthesis systems to use a physically coupled source and
detector so the geometry is always known and is always the same. This limits the imaging geometries and its large size
is impractical for mobile or field operations. To counter this, we have developed a free form tomosynthesis with a
decoupled, free-moving source and detector that uses a novel optical method for accurate and real-time geometry
calibration to allow for manual, hand-held tomosynthesis and even CT imaging.
We accomplish this by using a camera, attached to the source, to track the motion of the source relative to the
detector. Attached to the detector is an optical pattern and the image captured by the camera is then used to determine
the relative camera/pattern position and orientation by analyzing the pattern distortion and calculating the source
positions for each projection, necessary for 3D reconstruction. This allows for portable imaging in the field and also as
an inexpensive upgrade to existing 2D systems, such as in developing countries, to provide 3D image data. Here we
report the first feasibility demonstrations of free form digital tomosynthesis systems using the method.