PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.
X-ray image segmentation of different anatomical structures or tissue types is essential for diagnosing lesions of various kinds and for the differentiation between contrast agent and bone tissue. However, the complete separation of multiple targets and tissue types remains a challenge. We describe the use a combination of high-dimensional data clustering and material decomposition methods using spectral information from an energy resolving CdTe Medipix3 photon-counting detector. This paper introduces a flexible, iterative semi-supervised algorithm for multi-material decomposition that uses spectral measurements and the K-edge effects to label and classify CT voxel clusters using a Gaussian Mixture Model (GMM). Preliminary results show excellent quantitative accuracy and separation of more than 3 materials. Results are shown with phantom and mouse CT data. Our correction and calibration methods required for these successful decomposition results will also be described.
Juan C. R. Luna,Ian Harmon, andMini Das
"Tissue classification and contrast agent separation in spectral micro CT using Medipix3 CdTe detector", Proc. SPIE PC12031, Medical Imaging 2022: Physics of Medical Imaging, PC120310N (11 April 2022); https://doi.org/10.1117/12.2613361
ACCESS THE FULL ARTICLE
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.
The alert did not successfully save. Please try again later.
Juan C. R. Luna, Ian Harmon, Mini Das, "Tissue classification and contrast agent separation in spectral micro CT using Medipix3 CdTe detector," Proc. SPIE PC12031, Medical Imaging 2022: Physics of Medical Imaging, PC120310N (11 April 2022); https://doi.org/10.1117/12.2613361