Optimization of a low-dose 320-slice multi-detector computed tomography chest protocol using a phantom

Abstract

Background: Radiation exposure because of medical imaging is a public safety concern. Objective: To optimize a protocol for multi-detector computed tomography (MDCT) of the chest with pulmonary nodules based on targeted standard deviation (SD) to achieve minimal patient radiation exposure, while maintaining acceptable diagnostic information. Methods: A Lungman chest phantom with 5 spheres simulating nodules of 12, 10, 8, 5, and 3 mm diameters at 100 Hounsfield Units (HU) was scanned by 320-slice MDCT by varying targeted SD of 9, 14, 20, and 25, beam pitch of 0.637, 0.813, 1.388, at 120 and 100 kVp, and 10-400 mA. We measured the radiation doses in terms of corresponding volume computed tomography dose index (CTDIvol). Quantitative image quality was determined by the percent contrast-to-noise ratio (%CNR). Two independent radiologists evaluated images for nodule detection acceptability on a 5-point rating scale. Results: We found targeted SD increased results in decreased radiation dose and %CNR. We found CTDIvol reduction between –40% and –88% compared with the default setting for various targeted SD, beam pitch, and kVp. The %CNR was highest at targeted SD 9. There was good agreement between the image quality scores by 2 independent readers (k = 0.66). Conclusion: Targeted SD 20, 10-400 mA at pitch 0.813 and 120 kVp on a nodule size of 5 mm was considered the optimal protocol for low-dose chest CT. Our recommended protocol can reduce the radiation dose substantially under the manufacturer’s default protocol, while preserving acceptable image quality for lung nodule detection. Keywords: CT chest, CT noise index, MDCT, radiation dose reduction, targeted SD