AJR:201, November 2013 1075 synchronous liver or lung metastasis may un- dergo up to 14 CT studies during the same period [7, 8]. The theoretic linear no-threshold model for radiation dose suggests that exposure to CT scans and other radiation-emitting med- ical imaging devices may increase carcino- genic risk, regardless of the radiation dose [9]. Although the linear no-threshold mod- el is widely used as a model for cancer risk estimation from radiation exposure, there is debate whether this model is in fact applica- ble to routine medical imaging procedures. It has been suggested that defense mecha- nisms of the human body against cancer are overwhelmed at high doses, whereas they are stimulated at the low doses at which di- agnostic medical imaging equipment operate [10]. Accordingly, the American Association MDCT of Chest, Abdomen, and Pelvis Using Attenuation- Based Automated Tube Voltage Selection in Combination With Iterative Reconstruction: An Intrapatient Study of Radiation Dose and Image Quality Fernanda Dias Gonzalez-Guindalini 1 Marcos Paulo Ferreira Botelho 1 Hüseyin Gürkan Töre 1 Richard W. Ahn 1 Leo I. Gordon 2 Vahid Yaghmai 1 Gonzalez-Guindalini FD, Ferreira Botelho MP, Töre HG, Ahn RW, Gordon LI, Yaghmai V 1 Department of Radiology, Northwestern Memorial Hospital, Northwestern University–Feinberg School of Medicine, 676 N Saint Clair St, Ste 800, Chicago, IL 60611. Address correspondence to V. Yaghmai (v-yaghmai@northwestern.edu). 2 Division of Hematology/Oncology, Northwestern University–Feinberg School of Medicine, Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago, IL. Medical Physics and Informatics • Original Research AJR 2013; 201:1075–1082 0361–803X/13/2015–1075 © American Roentgen Ray Society B ecause of widespread availability and excellent anatomic resolu- tion, CT is widely used in medical diagnosis [1]. The number of CT examinations tripled from 1996 to 2010, and CT now accounts for almost one half of the collective effective dose from medical proce- dures [2, 3]. Many patients, such as trauma or oncology patients, may frequently undergo multiple di- agnostic imaging procedures for the purpos- es of diagnosis, treatment, and surveillance [4–6]. For example, the National Compre- hensive Cancer Network (NCCN) guidelines now recommend up to 12 CT scans during the first 5 years after diagnosis for surveil- lance in patients with stage IA and IB non- seminomatous testicular cancer, whereas a patient with colon cancer who has resectable Keywords: attenuation-based automated kilovoltage selection, fltered backprojection reconstruction, image quality, iterative reconstruction, radiation dose DOI:10.2214/AJR.12.10354 Received November 21, 2012; accepted after revision Janurary 24, 2013. F. D. Gonzalez-Guindalini, M. P. Ferreira Botelho, and H. G. Töre received support through a Siemens Healthcare Educational Grant. OBJECTIVE. The purpose of this study was intrapatient comparison of image quality and radiation dose between MDCT scans of the chest, abdomen, and pelvis obtained with at- tenuation-based automated kilovoltage selection and sinogram-affirmed iterative reconstruc- tion and scans obtained with standard kilovoltage selection and a filtered backprojection im- age reconstruction algorithm. MATERIALS AND METHODS. One hundred one oncology patients who had under- gone two chest, abdominal, and pelvis CT scans within 1 year were imaged with standard tube voltage selection of 120 kVp using a filtered backprojection reconstruction algorithm (protocol 1) and with attenuation-based automated tube voltage selection using an iterative reconstruction algorithm (protocol 2). Radiation dose parameters (volumetric CT dose index [CTDI vol ], dose-length product, and effective dose) as well as image noise, signal-to-noise ra- tio, and contrast-to-noise ratio were compared. Two independent radiologists evaluated image quality and sharpness. Student t test, Fisher exact test, and Wilcoxon signed-rank test were used for analysis. A p value less than 0.05 was considered significant. RESULTS. Mean ± SD CTDI vol values were 19.9 ± 4.43 mGy and 12.53 ± 4.79 mGy for protocols 1 and 2, respectively ( p < 0.0001). Effective dose was 38.2% lower on average using protocol 2 compared with protocol 1 (12.08 vs 19.55 mSv; p < 0.0001). Objective image qual- ity parameters were significantly better in protocol 2 ( p < 0.0001). Both radiologists found the overall image quality and sharpness to be similar for both protocols ( p > 0.05). CONCLUSION. In patients undergoing CT examination of the chest, abdomen, and pel- vis, the combination of attenuation-based automated tube voltage selection with iterative re- construction significantly reduced radiation dose parameters and maintained objective image quality when compared with standard tube voltage selection associated with filtered backpro- jection reconstruction. Gonzalez-Guindalini et al. MDCT Radiation Dose and Image Quality Medical Physics and Informatics Original Research Downloaded from www.ajronline.org by 52.73.204.196 on 05/17/22 from IP address 52.73.204.196. Copyright ARRS. For personal use only; all rights reserved