Image processing of computed tomography scan data for automated structure identification and stress analysis

Miscellaneous · Fri Jan 01 00:00:00 EST 1988

OSTI ID:6809450

Meagher, J M

This research investigates a non-invasive method to determine trabecular architecture, and its incorporation into FEM automatically generated from CT scan data. Part of the image processing involves a general method to remove beam hardening error from CT scan data without requiring the original projection data. A FEM of a canine femur is used to investigate bone stress-morphology relationships, and their sensitivity to load distribution, beam hardening errors, and anisotropic modeling of bone. The method to identify bone uses sequential application of edge gradient masks. CT span predictions were verified using histologic sections. The area fraction of bone and trabecular orientation were determined within 10% and 15 degrees when the pixel size was smaller than the intratrabecular spacing. Beam hardening is removed from CT scan data using simulated projections through the uncorrected image. Linearity of the reconstruction algorithm is used to add reconstructed projection corrections to the uncorrected image. The error in theoretical attenuation for a test phantom decreased from 30% to 5% with virtual elimination of the streak artifact. The image improvement equals that achieved with corrections based on the original projection data when all beam hardening occurs within the reconstruction space.

• Research Organization: California Univ., Berkeley, CA (USA)
• OSTI ID: 6809450
• Resource Relation: Other Information: Thesis (Ph. D.)
• Country of Publication: United States
• Language: English

Related Subjects

62 RADIOLOGY AND NUCLEAR MEDICINE
COMPUTERIZED TOMOGRAPHY
OPTIMIZATION
SKELETAL DISEASES
DIAGNOSIS
SKELETON
BONE FRACTURES
DATA COVARIANCES
IMAGE PROCESSING
MOLECULAR STRUCTURE
BODY
DIAGNOSTIC TECHNIQUES
DISEASES
INJURIES
ORGANS
PROCESSING
TOMOGRAPHY
550602* - Medicine- External Radiation in Diagnostics- (1980-)