ImageJ shares with several other image-analysis software programs the capability to open and display DICOM images. We then demonstrate the utility of this approach by describing three applications that were implemented as ImageJ macro scripts. We describe two open source plug-ins to the ImageJ software package, Import Dicom Sequence (available at ) and Query Dicom Header (available at ) developed in our laboratory to simplify the utilization of DICOM images in ImageJ. The ability to read DICOM images as well as the recent addition of a macro language for ImageJ has made it possible to develop low-cost image-processing applications on this platform. ImageJ runs on Windows, Macintosh, and UNIX platforms, and can read several image formats including RAW and DICOM part 10 format images (“DICOM images”). ImageJ ( ) is a free, open-source image-processing platform written in Java and supported by the National Institute of Health (Wayne Rasband, National Institute of Mental Health). The information on the DICOM header can also be useful for error checking, allowing errors in the scan parameters used to obtain source images to be more easily detected.
Assuming that the information on the DICOM header is correct, the results of the image analysis may be more reliable using this approach rather than an approach requiring users to input this information. For example, information on the field of view could be used to modify the size of images in the displayed output of the analysis. The information on the header can be useful to automate analyses. Second, the DICOM headers included with these images contain valuable information, such as scan parameters in MR and CT imaging, that may be lost if images are converted to another format. First, it is often possible to obtain these images directly from imaging equipment and/or PACS storage without the need for intermediate conversion steps. The use of images formatted using the specifications in part 10 of the DICOM standard (“DICOM images”) as a starting point for image analysis offers several advantages. Some packages can be implemented only on proprietary workstations or may require that images are available in particular formats. They may be designed to run on a single operating system or on multiple systems. Packages available to perform image processing vary widely in terms of flexibility and cost. If a purchased software package can be programmed or modified, a combination of these two approaches may be used. For most research applications, two approaches are used to obtain software for image analysis: the software is either purchased from a commercial vendor or developed and produced in-house by individuals with computer programming experience. The advent of new radiological applications and the rapidity of advances in image-processing techniques have led to increasing demand for customized, flexible image-analysis software.
0 kommentar(er)
