Kang Bong Joo, Kim Sung Hun, An Yeong Yi, Choi Byung Gil
Department of Radiology, Seoul St. Mary's Hospital, The Catholic University of Korea, 505 Banpo-Dong, Seocho-Ku, Seoul, 137-040, Korea.
Int J Comput Assist Radiol Surg. 2015 May;10(5):509-15. doi: 10.1007/s11548-014-1087-8. Epub 2014 Jun 10.
The storage requirements for full-field digital mammography (FFDM) in a picture archiving and communication system are significant, so methods to reduce the data set size are needed. A FFDM crop tool for this purpose was designed, implemented, and tested.
A total of 1,651 screening mammography cases with bilateral FFDMs were included in this study. The images were cropped using a DICOM editor while maintaining image quality. The cases were evaluated according to the breast volume (1/4, 2/4, 3/4, and 4/4) in the craniocaudal view. The image sizes between the cropped image group and the uncropped image group were compared. The overall image quality and reader's preference were independently evaluated by the consensus of two radiologists.
Digital storage requirements for sets of four uncropped to cropped FFDM images were reduced by 3.8 to 82.9 %. The mean reduction rates according to the 1/4-4/4 breast volumes were 74.7, 61.1, 38, and 24 %, indicating that the lower the breast volume, the smaller the size of the cropped data set. The total image data set size was reduced from 87 to 36.7 GB, or a 57.7 % reduction. The overall image quality and the reader's preference for the cropped images were higher than those of the uncropped images.
FFDM mammography data storage requirements can be significantly reduced using a crop tool.
在图像存档与通信系统中,全场数字化乳腺摄影(FFDM)的存储需求巨大,因此需要降低数据集大小的方法。为此设计、实现并测试了一种FFDM裁剪工具。
本研究纳入了1651例双侧FFDM筛查乳腺摄影病例。使用DICOM编辑器裁剪图像,同时保持图像质量。根据头尾位视图中的乳房体积(1/4、2/4、3/4和4/4)对病例进行评估。比较裁剪图像组和未裁剪图像组之间的图像大小。由两名放射科医生共同独立评估整体图像质量和读者偏好。
四组未裁剪到裁剪的FFDM图像的数字存储需求降低了3.8%至82.9%。根据1/4 - 4/4乳房体积的平均降低率分别为74.7%、61.1%、38%和24%,表明乳房体积越小,裁剪后数据集的大小越小。总图像数据集大小从87GB减少到36.7GB,减少了57.7%。裁剪图像的整体图像质量和读者偏好高于未裁剪图像。
使用裁剪工具可显著降低FFDM乳腺摄影数据的存储需求。