Effects of water and other filters on fracture resolution in industrial micro-CT scanning

Filtering is known to improve image quality of CT scans. Water immersion is one type of filtering that has been used for CT scanning dry (skeletonized) bones, and it has been suggested that this approach can increase measurement accuracy and improve fracture resolution. These tests have previously involved clinical CT scanners. Here we use an industrial micro -CT scanner and test whether water immersion and other x-ray filtering options increases fracture resolution in reconstructed scans of dry bones. Eleven dry non-human bones were CT scanned using the same acquisition parameters, while varying filter options. Bones were scanned (1) in an unfiltered "dry" air environment, (2) using metal filters at the x-ray source, and (3) with the bones immersed in water. A small subset of bones (N = 3) was also scanned using the same parameters except increasing the number of projections acquired from 500 to 1500. Reconstructed scans were evaluated by the authors, in part using a Likert scale comparing filtered with unfiltered scans, to assess fracture resolution (overall appearance and extent). Results showed that increasing the projections resulted in the greatest improvement in fracture resolution, followed by filtering at the x-ray source. Water immersion performed poorly overall, possibly due to movement artifacts that result from this type of scanning, in which the specimen rotates on a stage during the scan. When using this type of CT scanner, if increased fracture resolution is desired, water immersion is not recommended; increasing the number of projections or filtering at the x-ray source is suggested instead.