Chao Ma1, Xiaoyu Guo1, Li Liu1, Qian Zhan1, Jing Li1, Chengcheng Zhu2, Li Wang3, Jing Zhang4, Xu Fang1, Jianxun Qu5, Shiyue Chen1, Chengwei Shao1, Jian-Ping Lu1. 1. Department of Radiology, Changhai Hospital of Shanghai, the Second Military Medical University, No.168 Changhai Road, 200433, Shanghai, China. 2. Department of Radiology and Biomedical Imaging, University of California, San Francisco, California, USA. 3. Department of Radiology, Changhai Hospital of Shanghai, the Second Military Medical University, No.168 Changhai Road, 200433, Shanghai, China. wangli_changhai@163.com. 4. Department of Pathology, Changhai Hospital of Shanghai, the Second Military Medical University, Shanghai, China. 5. GE Healthcare, MR Group, Shanghai, China.
Abstract
BACKGROUND: To investigate the influence of region of interest (ROI) size on tumor apparent diffusion coefficient (ADC) measurements in pancreatic cancer. METHODS: The study population consisted of 64 patients with pathologically proved pancreatic ductal adenocarcinomas (PDACs), who underwent preoperative magnetic resonance imaging (MRI) examinations including diffusion-weighted imaging (DWI). The tumor ADCs were measured by two independent readers using six round ROIs with sizes ranging from 20 to 214 mm2 (9 to 97 pixels) in both the six separate measurements. The intra- and inter-observer variabilities were analyzed by using the coefficient of variance (CV), the interclass correlation coefficient (ICC) and Bland-Altman analysis. The mean ADCs measured with the 6 different-sized ROIs were compared using one-way repeated analysis of variance. The sample sizes were calculated by using 80% power and a 5% significance level to detect 10 to 25% changes in ADC measurements. RESULTS: The largest ROI (ROI214) yielded the best intra-observer repeatability (CV, 6.3%; ICC, 0.93) and inter-observer reproducibility (CV, 10.1%; ICC, 0.84). The mean differences in ADC measurements ± limits of agreement between the two readers were (0.06 ± 0.47) × 10-3 mm2 for ROI20, (0.08 ± 0.46) × 10-3 mm2 for ROI46, (0.05 ± 0.37) × 10-3 mm2 for ROI82, (0.07 ± 0.42) × 10-3 mm2 for ROI115, (0.05 ± 0.43) × 10-3 mm2 for ROI152 and (-0.02 ± 0.29) × 10-3 mm2 for ROI214. CONCLUSIONS: ROI size had a considerable influence on the ADC measurements of PDACs.
BACKGROUND: To investigate the influence of region of interest (ROI) size on tumor apparent diffusion coefficient (ADC) measurements in pancreatic cancer. METHODS: The study population consisted of 64 patients with pathologically proved pancreatic ductal adenocarcinomas (PDACs), who underwent preoperative magnetic resonance imaging (MRI) examinations including diffusion-weighted imaging (DWI). The tumor ADCs were measured by two independent readers using six round ROIs with sizes ranging from 20 to 214 mm2 (9 to 97 pixels) in both the six separate measurements. The intra- and inter-observer variabilities were analyzed by using the coefficient of variance (CV), the interclass correlation coefficient (ICC) and Bland-Altman analysis. The mean ADCs measured with the 6 different-sized ROIs were compared using one-way repeated analysis of variance. The sample sizes were calculated by using 80% power and a 5% significance level to detect 10 to 25% changes in ADC measurements. RESULTS: The largest ROI (ROI214) yielded the best intra-observer repeatability (CV, 6.3%; ICC, 0.93) and inter-observer reproducibility (CV, 10.1%; ICC, 0.84). The mean differences in ADC measurements ± limits of agreement between the two readers were (0.06 ± 0.47) × 10-3 mm2 for ROI20, (0.08 ± 0.46) × 10-3 mm2 for ROI46, (0.05 ± 0.37) × 10-3 mm2 for ROI82, (0.07 ± 0.42) × 10-3 mm2 for ROI115, (0.05 ± 0.43) × 10-3 mm2 for ROI152 and (-0.02 ± 0.29) × 10-3 mm2 for ROI214. CONCLUSIONS: ROI size had a considerable influence on the ADC measurements of PDACs.
Entities:
Keywords:
ADC; DWI; MRI; Pancreatic cancer; Region of interest
Authors: Yi Wang; Zongming E Chen; Paul Nikolaidis; Robert J McCarthy; Laura Merrick; Laura A Sternick; Jeanne M Horowitz; Vahid Yaghmai; Frank H Miller Journal: J Magn Reson Imaging Date: 2011-01 Impact factor: 4.813
Authors: Andreas Lemke; Frederik B Laun; Miriam Klauss; Thomas J Re; Dirk Simon; Stefan Delorme; Lothar R Schad; Bram Stieltjes Journal: Invest Radiol Date: 2009-12 Impact factor: 6.016
Authors: Thomas J Vogl; Christian Marko; Marcel C Langenbach; Nagy N N Naguib; Natalie Filmann; Renate Hammerstingl; Tatjana Gruber-Rouh Journal: Eur Radiol Date: 2020-09-22 Impact factor: 5.315
Authors: Tara Barwick; Matthew Orton; Dow Mu Koh; Martin Kaiser; Andrea Rockall; Nina Tunariu; Matthew Blackledge; Christina Messiou Journal: Br J Radiol Date: 2021-03-18 Impact factor: 3.039
Authors: Nervana Hussien; Rasha S Hussien; Darine Helmy Amin Saad; Mohamed El Kassas; Walid F Elkhatib; Mai Ezz El Din Journal: Front Oncol Date: 2022-01-13 Impact factor: 6.244