Does software optimization influence the radiologists' perception in low dose paediatric pelvic examinations?

Purpose: To investigate whether software optimisation can improve an observers’ perception of image
quality in low dose paediatric pelvic examinations.
Methods: Twenty-five consecutive patients (3e7 years old) were referred for a pelvic digital radiography
(DR) examination. They were prospectively enrolled in the study over a 3-month period. Images were
taken at 80 kV and 2e4 mAs depending on pelvic thickness (9e15 cm). A small focal spot, 130 cm SID:
10 cm air gap and 1 mm Al and 0.2 mm Cu additional filtration were also utilised. Images were acquired
on a Canon DR detector and optimised using five different combinations of the multi-frequency processing
software (Canon DR system version NE, Version 7.1 with SPECTRA) to comply with the ALARA
principle. Image quality was blindly evaluated using the subjective Visual Grading Analysis (VGA) by five
experienced musculoskeletal radiologists, including the evaluation of six anatomical image quality
criteria (scored from 1 to 5).
Results: Consistently, the VGA results indicated that by using software optimised parameters, image
quality was suitable for diagnosis in 48e71% of all images. Based on a VGC analysis all software optimised
images did have significant better image quality then the one with just the clinical settings. Noise
reduction was the software setting which influenced the image quality the most, area under the curve
(AUC) of 0.8172 95%CI 0.7953e0.8375.
Conclusion: Software optimisation improve the radiologists' perception of image quality and should thus
be thoroughly considered within clinical practise. Noise reduction is the software parameter which has
the greatest influence.