Program Information
Influence of Tube Current Modulation On Noise Statistics of Reconstructed Images in Low-Dose Lung Cancer CT Screening
J Hoffman1*, F Noo2 , M McNitt-Gray3 , (1) UCLA School of Medicine, Los Angeles, CA, (2) University of Utah, Salt Lake City, UT, (3) UCLA School of Medicine, Los Angeles, CA
Presentations
TH-EF-601-9 (Thursday, August 3, 2017) 1:00 PM - 3:00 PM Room: 601
Purpose: Tube current modulation (TCM) was not widely available during the National Lung Screening Trial, but TCM is now recommended in low-dose CT lung cancer screening, although the full impact on image quality is not well understood. The purpose of this study was to investigate the effect of TCM on image noise statistics within the lungs and assess if observed differences impact task performance.
Methods: Simulated thoracic CT scans of the XCAT phantom were created using both fixed tube-current and a realistic TCM profile. Poisson noise was added that included the effects of the bowtie filter. The total photon fluence at the central detector was equivalent between the two scenarios, and was determined via matching standard deviation values in the heart, lungs, and muscle of an actual screening patient of similar-size. 1000 unique noise realizations (500 with TCM and 500 without) were performed and reconstructed using open-source weighted-filtered-backprojection (FreeCT_wFBP). From the noise realizations, variance and covariance maps were generated and compared. These maps were then used to compare performance for an estimation task defined as computing the mean attenuation value, m, of a small, 10x10x10-voxel ROI place at locations around the lung.
Results: The variance and covariance maps demonstrated differences between fixed tube current and TCM in image noise statistics, particularly at the shoulder and heart levels. These observed differences were found to have an impact on the estimation task. We find that using TCM yields a noisier value for m at 71% of locations where the ROI can be centered within the lungs. But there is a small number of locations (~0.06%) where TCM provides strong improvements.
Conclusion: While TCM improves task performance at the apices of the lungs, task performance may be degraded at heart level. Refinement of TCM may be possible using task-based assessment of image quality.
Funding Support, Disclosures, and Conflict of Interest: * Fred Noo: Institutional Research Agreement, Siemens Healthineers; Research funding from Siemens Healthineers * Michael McNitt-Gray: Institutional Research Agreement, Siemens Healthineers
Contact Email: