X-RAY IMAGING: Trimodal animal scanner uses cone-beam tomography

Since its invention in 1967 by the late Godfrey Hounsfield at EMI Central Research Laboratories, computer-aided tomography (CAT) has become a standard procedure in nearly every hospital worldwide. Hounsfield realized that by taking a series of images from a photon source rotated around the subject, a series of slices could be imaged and then reconstructed to generate a three-dimensional image.

In the more than four decades that have passed, a number of variations of his original technique have been developed that include faster machines such as the helical or spiral computed tomography (CT) scanner and variations of the technique such as cone-beam tomography.

“Unlike a conventional CT scanner that uses one or multiple arcs of x-ray detectors to capture slices of the patient,” says Ivan Charaminsinau, software engineer at Animage (Pleasanton, CA, USA; www.animage-llc.com), “cone-beam systems measure the strength of x-rays projected onto a flat-panel detector in a 3-D solid angle from a point source.” The main advantage of this method is that it reduces the data collection time required, which is important for real-time imaging of the heart or the tracking of contrast agent flow through the body.

Founded in 2008 to bring advanced x-ray imaging products to the veterinary market, Animage has now developed a system based on this technique targeted for use by veterinarians. The system, Fidex, can be used as a standard 2-D x-ray scanner, a 2-D fluoroscopy system that generates 2-D x-ray videos, and a CT scanner that generates 3-D images (see figure).


Animage’s Fidex cone-beam tomographic animal scanner uses NI LabVIEW to control the x-ray source and detector, I/O interface, and the motion of the gantry and table. CT data is reconstructed into 3-D images using a modified Feldkamp algorithm running on an Nvidia GeForce processor.
Click here to enlarge image

In the design of the system, Animage enlisted JAMCO Engineering (Cottage Grove, OR USA; www.jamcoengineering.com) to perform the mechanical design and Animage’s parent company Exxim Computing Corporation (Pleasanton, CA, USA; www.exxim.com) to provide the software needed for 3-D image reconstruction.

In the design of Fidex, a single 60–125-kV, 8-kW x-ray source is placed in one arm of a rotating gantry/C-arm. To digitize images generated by this x-ray source, a 1k × 1k-pixel x-ray detector from Varian (Palo Alto, CA, USA; www.varian.com) is placed in the other end of this arm. To perform x-ray imaging, the animal is first sedated and then placed on a motorized table in the middle of the C-arm’s rotation.

“As the gantry is rotated,” says Charaminsinau, “the x-ray point source generates a cone of x-rays that are digitized to 12-bit precision by the x-ray detector at up to 1000 angular steps. The table is then indexed through the gantry and the procedure repeated.” This data acquisition process takes about 20 s.

To control the x-ray source and imager, the I/O interface and display, the motion of the gantry, and the x-y-z position of the table within, Charaminsinau and his colleagues used LabVIEW software from National Instruments (Austin, TX, USA; www.ni.com). Initially developed on a PC, the software was then ported to a single-board RIO board from National Instruments. “Because this board integrates a processor, FPGA, and I/O on a single PCB,” says Charaminsinau, “it can be used to control every aspect of the Fidex scanner.”

Although the single-board RIO is used to control the scanner, it is not used to perform image reconstruction. Rather, after images are digitized they are transferred under I/O control from the FlexRIO board to a host PC that incorporates a GeForce graphics card from Nvidia (Santa Clara, CA, USA; www.nvidia.com).

To reconstruct 3-D images, Animage uses the COBRA software package from Exxim. According to Exxim, the company has enhanced the standard Feldkamp 3-D reconstruction algorithm traditionally used in image reconstruction with an iterative algorithm that is more tolerant against input data imperfections.

Using this algorithm ported to run across the multiple processors of the GeForce, COBRA can start reconstruction immediately after obtaining the first projection from the Fidex scanner. This allows data acquisition and reconstruction processes to run in parallel. At NIWeek 2009, Animage showed the first version of the scanner, a video of which can be seen at http://zone.ni.com/wv/app/doc/p/id/wv-1689/upvisited/y.

More Vision Systems Issue Articles

Vision Systems Articles Archives

Webcasts

Tips and techniques for improving OCR/OCV

Optical character recognition (OCR) and optical character verification (OCV) technologies offer many advantages in automation applications. They also involve some inherent challenges, which this we...

Making accurate measurements with machine vision software

In this webcast viewers will learn how to use machine vision to solve metrology tasks in factory automation and offline inspection applications. We will show how to choose the proper calibration an...

Choosing components for intelligent transportation systems

Intelligent transportation systems depend on the interoperation of sophisticated OEM components. This expert webcast will provide guidance on how to choose such components—including illumination sy...
July 22, 2014

3D Machine Vision - the key to automation

When it comes to the automation industry, machine vision continues to grow increasingly important. Without machine vision, automation is not possible. This webcast, sponsored by MVTec and presented...
July 17, 2014

Archives

Click here to view archived Vision Systems Design articles