Automatic tube current modulation QA test in CT: phantom scan, analysis, and code - plain & simple

AUTHORS: Juuso Ketola, Satu Inkinen, Henna Kavaluus, Mika Kortesniemi CONTACT INFORMATION: juuso.ketola@hus.fi DESCRIPTION: Introduction Tube current modulation (TCM) aims to achieve constant image quality in computed tomography (CT) slices by adapting the X-ray tube current as a function of patient attenuation. Constancy of TCM is important and should be evaluated with quality assurance (QA) measurements. In this work, we describe a fast TCM QA protocol using CT dose phantom. Phantom set-up A 16-cm diameter CT dosimetry phantom was mounted on the head support extension of the patient table. A 32-cm CT dosimetry phantom was positioned next to the 16-cm phantom on the patient table on its side, such that the circular symmetry was along the coronal plane. Scan isocenter was fixed to the middle of the 32-cm phantom. Positioning the phantoms this way provided a constant attenuation target offset from the isocenter, an instantaneous change in attenuation at the boundary of the two phantoms, and a continuously changing attenuation profile. Imaging protocol Example scan was performed with the Revolution EVO CT scanner (GE Healthcare, Chicago, IL, USA). A posteroanterior topogram was acquired before a helical CT scan (120 kV, 0.625 mm slice thickness, 0.9844 pitch, 0.5 s rotation time, standard FBP kernel). The scan range covered both phantoms and roughly 10 cm of additional volume in both superior and inferior directions. Full range of mA values [10 mA, 540 mA] was allowed for TCM to prevent saturation in areas of low and high attenuation, and the noise index was set to 35. Analysis code Average mA and slice position were extracted from the DICOM header of each cross-sectional image of the scan to obtain the mA-profile along scan length using a Python (v. 3.10) script. Consecutive QA scans could then be compared to this baseline scan visually and numerically. In this shareable code we concentrated on the most relevant TCM parameter, which is the tube current, as we wanted to keep the workflow as simple and quick as possible. Image quality analysis including noise evaluation is an obvious extension to be added to our implementation in the future.