Can the current radiation dose, in chest tomosynthesis, be reduced with retained image quality? A study in the context of lung cancer screening programs

This study was approved by the Ethical Review Board, and informed consent was obtained from all 50 participating patients (24 men, 26 women; mean age 66 years) with known pulmonary malignancies. Patients were included if they were over 24 years old and had at least one clearly visible lung lesion on CT, which served as the reference standard. CT examinations were performed within two weeks of the DTS exams to avoid changes in tumour size. Three DTS series with different exposure settings were performed for each patient. Chest tomosynthesis was conducted using a GE Definium 8000 system with VolumeRAD technology. The vendor-recommended protocol used 125 kV and standard filtration, producing 60 projection images over ±30°. To achieve 30% and 50% dose reductions, adjustments were made: added aluminium filtration, reduced tube voltage to 100 kV, and varied dose ratios (10, 7, and 5). All other acquisition and reconstruction settings were kept constant. Effective doses for the three protocols were estimated as 0.13 mSv, 0.09 mSv, and 0.06 mSv. Each DTS series was reconstructed into approximately 53 coronal images and evaluated by four observers. A total of 150 image series were assessed using ViewDEX software on calibrated PACS monitors. Observers rated image quality independently and in random order based on predefined anatomical criteria covering vessel demarcation, vessel disturbance, bone structure, and tumour homogeneity, using a five-point ordinal scale. For statistical evaluation, visual grading characteristics (VGC) analysis was used to compare image quality between the reduced-dose protocols and the reference protocol. The area under the VGC curve (AUCVGC) served as the measure of difference, with values near 0.5 indicating no significant difference. A fixed-reader, paired-data approach was applied, with p