ROI DICOM: What Is the ROI Tool in a DICOM Viewer?
What Is the ROI Tool and What Is It Used For?
This is a tool whose English acronym stands for “Region of Interest,” that is, region of interest. It refers to a part of the image that one wishes to filter or operate on in some way, for example to make an annotation or measurement.
In a DICOM viewer, the ROI is one of the basic functions. Different PACS systems offer different ways to mark the region of interest, or ROI, on a study, for example circles, ellipses, polygons, and rectangles, among others.
Through the properties of the viewer and the tool itself, both its appearance and its behavior can be customized.
Its main use is to perform measurements once the region of interest, or ROI, has been detected in an area of the medical image. This allows the physician to make comparisons with previous studies, adjust contrast or brightness, take measurements, and generally use all the tools of the DICOM viewer around the desired area.
Other tools of the DICOM viewer that most commonly interact once a region of interest, or ROI, has been detected include:
- Length measurement
- Angle measurement
- Area measurement
- Volume measurement
- Annotations
- Polygonal area selection
- Freehand area selection
What Is DICOM?
An acronym for Digital Imaging and Communication in Medicine, DICOM is a standard for digital imaging and communication in medicine. It was designed to specify a protocol of rules for data exchange, the digital image format of medical studies, and a file structure for medical images and associated related information exchanged between imaging equipment from different vendors.
DICOM was originally developed by the National Electrical Manufacturers Association (NEMA) and the American College of Radiology for computed axial tomography (CT) and magnetic resonance imaging (MRI) scans. It is currently maintained by the DICOM Standards Committee and supports a wide range of medical images in the fields of radiology, cardiology, pathology, and dentistry.
DICOM is used in virtually every modern hospital. It ensures the interoperability of the systems used to produce, monitor, transmit, query, process, retrieve, and print medical images and derived structured documents, as well as to manage clinical workflow through a PACS system.
How Does the ROI Relate to the DICOM Viewer?
Generally, diagnostic modalities such as magnetic resonance imaging, computed tomography, and computed radiography, among others, can be costly in terms of both image storage and infrastructure. For this reason, a way was sought to compress these images to make them more manageable for both storage and distribution, without affecting their resolution or diagnostic capability.
The ROI tool is also a form of compression, such that the area of the image belonging to the region of interest can be compressed using a lossless algorithm in order to recover the original data without alteration. Likewise, the outer area is compressed using a lossy, or irreversible, algorithm, which can likewise achieve results with a visually indistinguishable reduction in quality.
The DICOM protocol introduces the ROI tool through the capabilities of the JPEG 2000 image standard, allowing the area of greatest diagnostic importance within the image to be compressed losslessly, while the remaining region is compressed with moderate loss.
The ROI tool can be specified through a set of coordinates or as a binary mask the same size as the image.
There are three alternatives for ROI-based encoding:
- Tiling mechanisms
- CodeBlock Selection
- Coefficient Scaling (referred to as the Maxshift algorithm in the JPEG 2000 standard)
Among the functionalities of the DICOM protocol are extracting images in various formats, parsing the header, and packaging files into new DICOM files, as it is capable of writing headers that comply with the standard.
The ROI, in turn, must define a region of interest based on the interpretation of a set of coordinates or an image in PGM (portable graymap) format. The resulting ROI is returned in a standard format using JPEG 2000 compression. It accepts different input image formats and produces a sequence of output images with the (.jp2) extension. Its capabilities include the ability to apply lossy or lossless compression and interact with the ROI component to use the region-of-interest technique during compression.
References
- DICOM general description Overview.
- ROI-Based Processing. Math Works.
- González, L. “An Approach to Medical Image Compression Based on Regions of Interest and Motion Compensation”. Congreso Argentino de Informática y Salud (Argentine Congress on Informatics and Health), CAIS 2013