Takeshi Soyama1, Yusuke Sakuhara1, Kohsuke Kudo2, Daisuke Abo1, Jeff Wang3, Yoichi M Ito4, Yu Hasegawa5, Hiroki Shirato3. 1. Department of Diagnostic and Interventional Radiology, Hokkaido University Hospital, N-14, W-5, Kita-ku, Sapporo, Hokkaido, 060-0848, Japan. 2. Department of Diagnostic and Interventional Radiology, Hokkaido University Hospital, N-14, W-5, Kita-ku, Sapporo, Hokkaido, 060-0848, Japan. kkudo@huhp.hokudai.ac.jp. 3. Department of Radiation Medicine, Hokkaido University Graduate School of Medicine, Sapporo, Japan. 4. Department of Biostatistics, Hokkaido University Graduate School of Medicine, Sapporo, Japan. 5. Department of Radiology, Sapporo-Kosei General Hospital, Sapporo, Japan.
Abstract
PURPOSE: This preliminary study compared ultrasonography-computed tomography (US-CT) fusion imaging and conventional ultrasonography (US) for accuracy and time required for target identification using a combination of real phantoms and sets of digitally modified computed tomography (CT) images (digital/real hybrid phantoms). METHODS: In this randomized prospective study, 27 spheres visible on B-mode US were placed at depths of 3.5, 8.5, and 13.5 cm (nine spheres each). All 27 spheres were digitally erased from the CT images, and a radiopaque sphere was digitally placed at each of the 27 locations to create 27 different sets of CT images. Twenty clinicians were instructed to identify the sphere target using US alone and fusion imaging. The accuracy of target identification of the two methods was compared using McNemar's test. The mean time required for target identification and error distances were compared using paired t tests. RESULTS: At all three depths, target identification was more accurate and the mean time required for target identification was significantly less with US-CT fusion imaging than with US alone, and the mean error distances were also shorter with US-CT fusion imaging. CONCLUSION: US-CT fusion imaging was superior to US alone in terms of accurate and rapid identification of target lesions.
RCT Entities:
PURPOSE: This preliminary study compared ultrasonography-computed tomography (US-CT) fusion imaging and conventional ultrasonography (US) for accuracy and time required for target identification using a combination of real phantoms and sets of digitally modified computed tomography (CT) images (digital/real hybrid phantoms). METHODS: In this randomized prospective study, 27 spheres visible on B-mode US were placed at depths of 3.5, 8.5, and 13.5 cm (nine spheres each). All 27 spheres were digitally erased from the CT images, and a radiopaque sphere was digitally placed at each of the 27 locations to create 27 different sets of CT images. Twenty clinicians were instructed to identify the sphere target using US alone and fusion imaging. The accuracy of target identification of the two methods was compared using McNemar's test. The mean time required for target identification and error distances were compared using paired t tests. RESULTS: At all three depths, target identification was more accurate and the mean time required for target identification was significantly less with US-CT fusion imaging than with US alone, and the mean error distances were also shorter with US-CT fusion imaging. CONCLUSION: US-CT fusion imaging was superior to US alone in terms of accurate and rapid identification of target lesions.
Authors: Min Woo Lee; Hyunchul Rhim; Dong Ik Cha; Young Jun Kim; Dongil Choi; Young-sun Kim; Hyo K Lim Journal: AJR Am J Roentgenol Date: 2012-06 Impact factor: 3.959