Dynamic Spatial Reconstructor
The dynamic spatial reconstructor (DSR) is a unique computed tomography (CT)-based scanner valuable for three-dimensional imaging and visualization of high temporal resolution three-dimensional cardiac cycles. Developed in the 1970s and early 1980s, the DSR is the "multi-source, multi-detector high speed synchronous 3D CT scanner for high temporal and spatial resolution scanning of the heart, lungs, and circulation" according to the Mayo Clinic, where the scanner was developed and is located. It is considered a research prototype and is not available commercially.
The DSR was developed as a non-invasive diagnostic device to detect lung cancer and heart disease in their early stages. It emerged as an answer to the tremendous challenge of using CT to provide 3D reconstruction of moving objects such as the cyclic motion of the beating heart. Due to its efficacy, it has become the standard in the field of three-dimensional real-time imaging by which other non-invasive imaging modalities are measured in their effectiveness for achieving various diagnoses.
Only a single DSR exists, at the Mayo Clinic site (Rochester, Minnesota), due to its prohibitive cost and size. The physical machine comprises:
- a gantry 15 ft (4.57 m) in diameter and 20.5 ft (6.24 m) in length, weighing about 17 U.S. tons
- fourteen x-ray guns featured within a hemicylindrical arrangement (surrounding the patient or subject over-head and on the sides) and targeted at an adjacent hemicylindrical fluorescent screen
- fourteen rotating two-dimensional television cameras and eight video disc recorders for recording the x-rays
- electronics and software algorithms for image acquisition
The DSR is theoretically capable of acquiring image data for up to 240 contiguous 0.9 mm thick segments in time periods down as far as 1/60 of a second. The process can then be repeated as rapidly as 60 times per second. Due to limiting physical factors of the machine, however, these values are somewhat diminished in practice. The 14 rotating television cameras, possessing 240 scan lines apiece, receive x-ray photons from the 14 x-ray point sources directly opposite them at a frequency of 1/60 second, which happens to be a physiologically appropriate frequency for internal investigations involving moving organs such as the heart.
Though the DSR is capable of diagnosing a myriad of heart and lung disorders, its cost (and thus limited capability for service) has prevented it from becoming a routinely useful clinically diagnostic tool. Nevertheless, over the years since its inception in 1983, the DSR has made possible the collection and analysis of unique, important data that has been especially employed in cardiac dynamics research and in assuring the legitimacy of other imaging modalities.
An example of typical research usage is estimating the spatio-temporal distribution of the velocity of the left ventricular wall from experimental data obtained on the DSR. Essentially, a dense velocity field may be computed by applying a differential technique. This velocity field is obtained by mathematically applying the three following assumptions to the images: conservation of mass, incompressibility, and smoothness of the velocity field. In the case of this study, the results were in terms of the evolution of the field over time and maximum velocities, which were found to be in good agreement with the known physiological behavior of the heart.
The DSR is not without its problems, however. In addition to its enormous cost and size, another difficulty plaguing the machine is that the gantry rotates only 1.5° per 1/60 second, which hinders the homogeneous distribution in orientation angle of images per time period.
In order to keep the DSR more modern, some alterations have been implemented over the years since its launch, such as converting the old cameras (image isocon) to CCD (charge coupled device) cameras, larger lenses, and utilizing digitized images with corresponding algorithms. The DSR has thus been used to examine cardiopulmonary mechanics and pulmonary ventilation. Studies have confirmed the functionality of volumetric CT (DSR) in accurately resolving lung volumes, cardiac chamber volumes, myocardial muscle mass, and regional lung density, among others.
In general, the greater the number of viewing angles used in collecting data tends to generate better images, at least up to a point. At least 4/60 of a second of scanning is typically desirable for creation of a reasonably good image reconstruction by the DSR. However, the DSR operator, or the research team doing the experiment, must consider the organ of interest in terms of its speed of movement within the body. This is so that interesting and
Computed tomography (CT)—Also known as CAT (computed axial tomography) scans. Computed tomography scans are completed with the use of a 360-degree x-ray beam and computer reconstruction of images using appropriate algorithms. Two-dimensional slices are acquired and stacked up computationally to yield a three-dimensional density map of areas of interest. These scans allow for high resolution cross-sectional views of body organs and tissues.
relevant data can be taken at an appropriate rate, based on the similarity between the DSR temporal resolution and the tissue velocities and periodicity, in order to obtain optimal images.
Maintenance is performed as needed on this unique instrument.
Health care team roles
The DSR is mainly a research machine, so health care roles are generally relegated to interpretation of data sets by a doctor or technician trained to work with the apparatus.
The DSR is used at its sole site in the Mayo Clinic. It is generally used by researchers in fields such as cardiac dynamics which require in vivo 3D data for specific studies or by doctors who wish to measure how well another imaging modality (i.e. electron CT) performs in comparison. It is used on humans as well as a wide range of animals for research.
Very few people are needed to operate the DSR since there is only one. Operating the scanner requires very specialized training that would be provided by technicians and technologists on site at Mayo if the need arises for its use.
Gorce, J.M., D. Friboulet, and I.E. Magnin. "Estimation of three-dimensional cardiac velocity fields: Assessment of a differential method and application to 3d ct data." Medical Image Analysis. 1, no. 3 (1997): 245-261.
The Mayo Clinic. 200 First Street SW, Rochester, MN 55905 507-284-2511. <http://www.mayo.edu>.
The Dynamic Spatial Reconstructor. <http://everest.radiology.uiowa.edu/gallery/dsr.html>.
Bryan Ronain Smith