DAMNNet: Detruncation of Attenuation Maps using Neural NetworksPublic
Downloadable Contentopen in viewer
Attenuation maps derived from Computed Tomography (CT) scans are used for attenuation correction in Single-Photon Emission Computed Tomography (SPECT) imaging. However, CT scans can be truncated due to obese or misaligned patients who extend outside the scanner’s field of view. The goal of this project was to develop a method to reconstruct truncated CT scans using modern image processing and deep learning techniques. We used image processing to extract the contour of the body, and then use a Convolutional Neural Network to infer missing regions of the body to be synthesized with a voxel filling algorithm. Our method accurately corrects CT scans for use in SPECT attenuation correction at minimal cost to medical professionals while facilitating better diagnoses of cardiac patients.
- This report represents the work of one or more WPI undergraduate students submitted to the faculty as evidence of completion of a degree requirement. WPI routinely publishes these reports on its website without editorial or peer review.
- Date created
- Resource type
- Rights statement
- In Collection:
Permanent link to this page: https://digital.wpi.edu/show/9306t194z