ABSTRACT
This chapter on image file structures in Nuclear Medicine is intended to provide a broad background to software developers to assist in extracting information from image files. It covers the basic structure of the binary representation of signed and unsigned integer numbers, floating-point real numbers, character representation as well as big-endian and little-endian byte order, followed by a description of the Interfile and DICOM file structures. The discussion on the Interfile structure introduces the concept of metadata in the form of Keys and Key Values. In the paragraphs covering the DICOM file structure, the concepts of Information Objects, Services and Service Object Pairs are mentioned, but more emphasis is placed on the use and structure of Data Elements and Data Element Tags as described in the DICOM Data Dictionary. Both Interfile as well the DICOM file structures are explained by means of examples. The chapter concludes with a description of the characteristics of the Unique Identifiers (UID’s) that ensure global uniqueness across multiple sites, vendors, and equipment.
