Improvement of CT image quality with iterative reconstruction idose4

Transcription

1 Improvement of CT image quality with iterative reconstruction idose4 Poster No.: C-0387 Congress: ECR 2014 Type: Scientific Exhibit Authors: M.-L. Olsson, K. Norrgren, M. Söderberg; Malmö/SE Keywords: Physics, CT, Radiation physics, Image verification DOI: /ecr2014/C-0387 Any information contained in this pdf file is automatically generated from digital material submitted to EPOS by third parties in the form of scientific presentations. References to any names, marks, products, or services of third parties or hypertext links to thirdparty sites or information are provided solely as a convenience to you and do not in any way constitute or imply ECR's endorsement, sponsorship or recommendation of the third party, information, product or service. ECR is not responsible for the content of these pages and does not make any representations regarding the content or accuracy of material in this file. As per copyright regulations, any unauthorised use of the material or parts thereof as well as commercial reproduction or multiple distribution by any traditional or electronically based reproduction/publication method ist strictly prohibited. You agree to defend, indemnify, and hold ECR harmless from and against any and all claims, damages, costs, and expenses, including attorneys' fees, arising from or related to your use of these pages. Please note: Links to movies, ppt slideshows and any other multimedia files are not available in the pdf version of presentations. Page 1 of 14

2 Aims and objectives The number of computed tomography (CT) examinations is increasing at radiology departments and the collective effective dose to patients is increasing [1]. Many steps have been made to enable the reduction of radiation dose in CT examinations [2]. One recently important development is the iterative reconstruction algorithms [3, 4]. The purpose of this study is to investigate the effect of different idose 4 levels on the image quality with various scanning parameters and reconstruction filters. Methods and materials The study was done on a Brilliance 64 (Philips Medical Systems, Best, Netherlands) with iterative reconstruction method idose 4 available [5]. Acquisitions were performed using an image quality phantom (Catphan 600, The Phantom Laboratory Inc, Greenwich, USA) and an anthropomorphic whole body phantom (Fig. 1, PBU-60, Kyoto Kagaku, Kyoto, Japan). To simulate a more patient like phantom, an oval body annulus (25-35 cm diameter) was applied around the Catphan (Fig. 2). A helical abdomen protocol with different tube voltages ( kv) and varied tube load ( mas) was used. All other scan parameters were fixed. All collected data were reconstructed with different levels of idose 4 (level 1-6) and filtered back projection (FBP). Different reconstruction filters (B: soft and C: sharp) were used in combination with idose 4 to investigate how the image quality parameters were affected. Image quality was evaluated in terms of noise (SD, standard deviation), low-contrast and spatial resolution. The program AutoQA Lite (Version , Iris QA, LLC, Frederick, USA) was used for evaluation of image noise and low-contrast. Spatial resolution was subjectively evaluated. Image noise in the anthropomorphic phantom was manually evaluated by drawing a region of interest in the liver. Page 2 of 14

3 Fig. 1: The anthropomorphic whole body phantom PBU-60. References: Kyoto Kagaku Page 3 of 14

4 Fig. 2: The Catphan 600 with oval body annulus. References: Department of Clinical Sciences, Medical Radiation Physics, Skåne University hospital - Malmö/SE Images for this section: Page 4 of 14

5 Fig. 1: The anthropomorphic whole body phantom PBU-60. Page 5 of 14

6 Fig. 2: The Catphan 600 with oval body annulus. Page 6 of 14

7 Results Noise - Catphan For the Catphan phantom several different comparisons were made to investigate the impact of idose 4 on the image noise. With idose 4 the noise can be reduced with 13-49% compared to filtered back projection depending on which level of idose 4 was used (Table 1). This is in agreement with the results by Noël et al [6]. Reduction of noise was similar for scans both with and without body annulus. The noise reduction was similar for different tube voltages. Table 1: Measured noise level in Catphan 600 with and without body annulus for 100 mas and different kv. The values in the parenthesis are relative changes compared to filtered back projection (FBP) in percentage. References: Department of Clinical Sciences, Medical Radiation Physics, Skåne University hospital - Malmö/SE The noise reduction for the Catphan phantom with body annulus was investigated for 120 kv and for different mas settings. Using 50 mas and filtered back projection a SD of 24 was obtained (table not shown). This noise level would in most clinical cases be unacceptable. The noise was significantly reduced by applying the iterative reconstruction. With idose 4 level 4 the noise was reduced by 34% giving an image noise of 16. For idose 4 level 6 the noise was reduced even further to a SD of 13. The noise reduction by using idose 4 can significantly improve the possibility to decrease the radiation dose to the patient keeping the noise at a clinically relevant level. Noise - Anthropomorphic phantom Page 7 of 14

8 Figure 3 shows measured noise (SD) for a homogenous part of the liver in the anthropomorphic phantom using 120 kv and three different dose levels (CTDI vol of 2.1, 3.0 and 9.2 mgy). Noise is plotted against the reconstruction method (FBP and different levels of idose 4, level 1-6). The relative dose reduction was similar for all dose levels. Fig. 3: Measured noise level for a homogenous part of the liver in the anthropomorphic phantom. The noise is plotted versus the reconstruction method. Result is given for FBP (filtered back projection) and different idose4 levels (1-6). Data for three different dose levels and for two reconstruction filters are compared. References: Department of Clinical Sciences, Medical Radiation Physics, Skåne University hospital - Malmö/SE Two reconstruction filters were compared, filter B which is a standard filter for abdominal scan protocols and filter C which is a sharper standard filter. The resulting noise reduction was similar for both filters. Low-contrast resolution With Catphan the low-contrast for 200 mas and 120 kv was 0.3% (3 HU) for 8.6 mm, 7.2 mm, 6.3 mm and 5.0 mm respectively for FBP, idose 4 levels 2, 4 and 6. The results showed that increasing level of idose 4 improved the low-contrast resolution. Page 8 of 14

9 Similar results were found when scanning with body annulus. The low-contrast was 0.3% (3 HU) for 29 mm, 24 mm, 20 mm and 16 mm respectively (FBP, idose 4 level 2, 4 and 6). For this more patient like situation it was shown that the image quality was improved with increasing level of idose 4. Figure 4 shows the low-contrast improvement with idose 4 for two different filters (B and C). Fig. 4: Low-contrast images for 200 mas and 120 kv for filter B (upper row) and filter C (lower row). The images from left to right are reconstructed with FBP, idose4 level 2, 4 and 6, respectively. References: Department of Clinical Sciences, Medical Radiation Physics, Skåne University hospital - Malmö/SE Spatial resolution The reconstruction with different idose 4 levels did not affect the resolution (Fig. 5). This is in accordance with the results by Noël et al [6]. By using a sharper filter (C) the resolution was improved. The combination of sharper filter and noise reduction with idose 4 increased the detectability. Page 9 of 14

10 Fig. 5: Spatial resolution in the Catphan phantom. The upper row is reconstructed with filter B, the lower row with filter C. The images from left to right are reconstructed with FBP, idose4 level 2, 4 and 6. The reconstruction with different idose4 levels does not affect the image resolution. The sharper filter (C) increases the resolution. References: Department of Clinical Sciences, Medical Radiation Physics, Skåne University hospital - Malmö/SE Images for this section: Table 1: Measured noise level in Catphan 600 with and without body annulus for 100 mas and different kv. The values in the parenthesis are relative changes compared to filtered back projection (FBP) in percentage. Page 10 of 14

11 Fig. 3: Measured noise level for a homogenous part of the liver in the anthropomorphic phantom. The noise is plotted versus the reconstruction method. Result is given for FBP (filtered back projection) and different idose4 levels (1-6). Data for three different dose levels and for two reconstruction filters are compared. Page 11 of 14

12 Fig. 4: Low-contrast images for 200 mas and 120 kv for filter B (upper row) and filter C (lower row). The images from left to right are reconstructed with FBP, idose4 level 2, 4 and 6, respectively. Fig. 5: Spatial resolution in the Catphan phantom. The upper row is reconstructed with filter B, the lower row with filter C. The images from left to right are reconstructed with FBP, idose4 level 2, 4 and 6. The reconstruction with different idose4 levels does not affect the image resolution. The sharper filter (C) increases the resolution. Page 12 of 14

13 Conclusion By using idose 4 there is a considerable potential to reduce the noise, increase lowcontrast and to maintain or increase the spatial resolution, thereby improving the image quality. Iterative reconstruction can be used to lower radiation dose and maintain image quality or to improve image quality. Personal information Marie-Louise Olsson, M.Sc. Medical Radiation Physics, Department of Clinical Sciences Malmö, Lund University, Skåne Univeristy Hospital, SE Malmö, Sweden Kristina Norrgren, Ph.D. Philips Healthcare, SE , Malmö, Sweden Marcus Söderberg, Ph.D. Medical Radiation Physics, Department of Clinical Sciences Malmö, Lund University, Skåne Univeristy Hospital, SE Malmö, Sweden References 1. National Council on Radiation Protection and Measurements. Ionizing radiation exposure of the population of the United States. NCRP Report No 160, Mattsson S, Söderberg M. Radiation dose management in CT, SPECT/CT and PET/CT techniques. Radiat Prot Dosim 2011; 147: Willemink MJ, de Jong PA, Leiner T, et al. Iterative reconstruction techniques for computed tomography part 1: technical principles. Eur Radiol 2013; 23: Willemink MJ, Leiner T, de Jong PA, et al. Iterative reconstruction techniques for computed tomography part 2: initial results in dose reduction and image quality. Eur Radiol 2013; 23: Page 13 of 14

14 5. Philips Electronics. White paper: idose 4 iterative reconstruction technique Noël PB, Fingerle AA, Renger B, et al. Initial performance characterization of a clinical noise - suppressing reconstruction algorithm for MDCT. AJR Am J Roentgenol 2011; 197: Page 14 of 14