OBJECTIVE: The purpose of study was to develop a computer-simulated liver phantom for hepatic CT studies. A computer-simulated liver phantom was mathematically constructed on a computer workstation. MATERIALS AND METHODS: The computer-simulated phantom was calibrated using real CT images acquired by an actual four-detector CT. We added an inhomogeneous texture to the simulated liver by referring to CT images of chronically damaged human livers. The mean CT number of the simulated liver was 60 HU and we added numerous 5-to 10-mm structures with 60+/-10 HU/mm. To mimic liver tumors we added nodules measuring 8, 10, and 12 mm in diameter with CT numbers of 60+/-10, 60+/-15, and 60+/-20 HU. Five radiologists visually evaluated similarity of the texture of the computer-simulated liver phantom and a real human liver to confirm the appropriateness of the virtual liver images using a five-point scale. RESULTS: The total score was 44 in two radiologists, and 42, 41, and 39 in one radiologist each. They evaluated that the textures of virtual liver were comparable to those of human liver. CONCLUSIONS: Our computer-simulated liver phantom is a promising tool for the evaluation of the image quality and diagnostic performance of hepatic CT imaging.
OBJECTIVE: The purpose of study was to develop a computer-simulated liver phantom for hepatic CT studies. A computer-simulated liver phantom was mathematically constructed on a computer workstation. MATERIALS AND METHODS: The computer-simulated phantom was calibrated using real CT images acquired by an actual four-detector CT. We added an inhomogeneous texture to the simulated liver by referring to CT images of chronically damaged human livers. The mean CT number of the simulated liver was 60 HU and we added numerous 5-to 10-mm structures with 60+/-10 HU/mm. To mimic liver tumors we added nodules measuring 8, 10, and 12 mm in diameter with CT numbers of 60+/-10, 60+/-15, and 60+/-20 HU. Five radiologists visually evaluated similarity of the texture of the computer-simulated liver phantom and a real human liver to confirm the appropriateness of the virtual liver images using a five-point scale. RESULTS: The total score was 44 in two radiologists, and 42, 41, and 39 in one radiologist each. They evaluated that the textures of virtual liver were comparable to those of human liver. CONCLUSIONS: Our computer-simulated liver phantom is a promising tool for the evaluation of the image quality and diagnostic performance of hepatic CT imaging.
Authors: John R Mayo; Kun-Il Kim; Sharyn L S MacDonald; Takeshi Johkoh; Peter Kavanagh; Harvey O Coxson; Sverre Vedal Journal: Radiology Date: 2004-07-29 Impact factor: 11.105
Authors: Thomas G Flohr; Stefan Schaller; Karl Stierstorfer; Herbert Bruder; Bernd M Ohnesorge; U Joseph Schoepf Journal: Radiology Date: 2005-04-15 Impact factor: 11.105
Authors: C B Caldwell; S J Stapleton; D W Holdsworth; R A Jong; W J Weiser; G Cooke; M J Yaffe Journal: Phys Med Biol Date: 1990-02 Impact factor: 3.609
Authors: Christian von Falck; Vesela Bratanova; Thomas Rodt; Bernhard Meyer; Stephan Waldeck; Frank Wacker; Hoen-oh Shin Journal: PLoS One Date: 2013-02-13 Impact factor: 3.240