Department of Diagnostic Radiology, Warren G. Magnusson Clinical Center, National Institutes of Health, Bethesda, Maryland, U.S.A.
A Note from the Editors edical image processing has become a major force in the imaging of cancer. Virtually all cancer imaging requires some level of image postprocessing. Among the most critical postprocessing functions are: image segmentation in which tumors are localized either manually or semiautomatically; image measurement in which physical and physiologic properties of tumors are characterized and mapped onto anatomic images; image visualization in which tumors are displayed in ways that are intuitively easy to grasp; and image registration in which two or more images are fused so that different tumor properties can be combined into one view. Image fusion and computer-aided diagnosis/ detection combine many of these methods to produce synthetic images that display multiple parameters and highlight abnormalities that may be otherwise difficult to detect. Image processing methods will undoubtedly continue to contribute to progress in cancer detection and management.
The development of medical imaging has progressed remarkably over the past few decades. Medical imaging such as radiography, fluoroscopy, computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography (PET), single photon emission computed tomography (SPECT), and ultrasonography (US) can be used for detection, localization, visualization, and characterization of tumors. Many new imaging techniques including angiography, perfusion, and dynamic contrast enhancement imaging have been developed to study tumors. Because the information provided by medical images is enormous and sometimes not very intuitive, image processing and analysis is performed to extract useful information from the images. Image processing and analysis is usually conducted after the acquisition of the images, thus it is also called "postprocessing."
Medical image processing techniques make use of engineering approaches derived from the fields of computer vision, computer graphics, and artificial intelligence research. This chapter presents concepts and techniques for medical image processing and analysis, especially for applications in tumor imaging. It is organized into five sections: image segmentation, image measurement and quantification, image display and visualization, image registration, and computer aided diagnosis/detection.
Was this article helpful?