Digital Fluoroscopy- Part III

Digital Fluoroscopy - Part III




The next improvement that will be seen will be flat panel imagers composed of silicon pixel detectors (SPD). These are direct detectors. Eventually all television camera devices will be obsolete.

Video System


The video system with conventional fluoroscopy uses a 525-line system.
Such a system is adequate for DF, but the higher spatial resolution can be obtained with a 1000-line system.




Interlaced Versus Progressive Mode


The method used for conventional fluoroscopy is the interlaced mode, where two fields of 2621/2 lines were read in 1/60th s (17 ms) to form a 525-line video frame in 1/30th s (33 ms).
In DF, the camera tube reads in progressive mode.
When reading the video signal in the progressive mode, the electron beam of the television camera tube sweeps the target assembly continuously from top to bottom in 33 ms.
What will occur is two interlaced fields with be combined to create the full image.
The video image is similar to form on the television monitor. There is no interlace of one field with another. This produces a sharper image with less flicker on the image screen.


Signal-to-Noise Ratio (SNR)


All analog electronic devices are inherently noisy.
Because of heated filaments and voltage differences, there is always a very small electric current flowing in any circuit. This is called background electronic noise. This electronic noise obscures the electronic signal. Because conventional television camera tubes have an SNR of about 200:1, the maximum output signal will be 200 times greater than the background electronic noise.
A SNR of 5:1 is minimally visible. A SNR og 200:1 is not sufficient for DF because the video signal is rarely at maximum and lower signals become even more lost in the noise.
This is very true for subtraction techniques. Image contrast resolution is degraded a lot by a system with a low SNR. When comparing the dynamic ranges where the SNR is 200:1, this is around 2^8 and the 1000:1 is 2^10.
The second set of number units are much more compatible with the computer systems of today.




Computer


Minicomputers and microprocessors are used with DF. The capacity of the computer is an important factor in determining image quality, the manner and speed of image acquisition, and image processing and manipulation.


Important characteristics of a DF system are:
Image matrix size
The system dynamic range
Image acquisition rate
The output signal from the television camera tube is transmitted by cable to an analog-to-digital converter (ADC).
The ADC accepts the continuously varying television camera output signal, the analog signal, and digitizes it.



To be compatible with the computer, the ADC must have the same dynamic range as the DF system. An 8-bit ADC would convert the analog signal into values between 0 and 255.
A 10-BIT ADC would be more precise, with an analog-to-digital conversion range from 0 to 2^10 or 0 to 1023. The output of the ADC is then transferred to main memory and manipulated so that a digital image in matrix form is stored. The dynamic range of each pixel, the number of pixels, and the method of storage determine the speed with which the image can be acquired, processed, and transferred to an output device.
The output of the ADC is then transferred to the main memory and manipulated so that a digital image in matrix form is stored. The dynamic range of each pixel, the number of pixels, and the method of storage determine the speed with which the image can be acquired, processed, and transferred to an output device.

If the image storage is in primary memory, which is usually the case, then data acquisition and transfer can be as rapid as 30 images per second. In general, if the image matrix is doubled, (ex: 512 to 1024), the image acquisition rate will be reduced by a factor of four.
A representative system may be able to acquire 30 images per second in a 512 x 512 matrix mode. If a higher spatial resolution image is required and a 1024 x 1024 mode is requested, then only 8 images per second can be acquired. This is limited by the time required to conduct and transfer the amount of data from one segment of memory to another.




Image Formation


The principle advantage of DF exams are the image subtraction techniques that can be performed and the ability to visualize vascular markings with a venous injection of contrast material. A disadvantage is an area beam must be used to perform this, which reduces image contrast because of the scattered radiation. Image contrast will then have to be enhanced electronically. Image contrast is improved by subtraction techniques that provide instantaneous viewing of the subtracted image, during the passage of a bolus of contrast medium.
Digital fluoroscopy provides better contrast resolution by postprocessing image subtraction.


Temporal subtraction and energy subtraction are two methods that are used in DF.
Each have their own advantages and disadvantages.


Temporal Subtraction:


A single kVp setting is used;

Normal x-ray beam filtration is adequate;
Contrast resolution of 1 mm at 1% is achieved;
Simple arithmetic image subtraction is necessary;
Motion artifacts are a problem;
Total subtraction of common structures is achieved;
Subtraction possibilities are limited by number of images.



Energy Subtraction:


Rapid voltage switching is required;
X-ray beam filter switching is preferred;
Higher x-ray intensity is required for comparable contrast resolution;
Complex image subtraction is necessary;
Motion artifacts are greatly reduced;
Some residual bone may survive subtraction;
Many more types of subtraction images are possible.


Temporal Subtraction techniques are more frequently used because of the high-voltage generator limitations in the energy subtraction mode. When the two techniques are combined, the process is called hybrid subtraction. Image contrast is enhanced still further by hybrid subtraction because of reduced patient motion between subtracted images.



Temporal Subtraction


Temporal subtraction refers to a number of computer-assisted techniques whereby an image is obtained at one time is subtracted from an image obtained at a later time. If, during the intervening period, contrast material was introduced into the vessels, the subtracted image will only contain vessels filled with the contrast media. There are two methods that are commonly used: mask mode and time-interval difference mode (TID).

Mask Mode


The patient is positioned under normal fluoroscopic control to ensure that the region of anatomy under investigation is within the FOV of the image intensifier. A power injector is armed and readied to deliver 30 to 50 ml of contrast material at the rate of approximately 15 to 20 ml/s through a venous entry. If an arterial entry is utilized, 10 to 25 ml of diluted contrast material at 10 to 12 ml/s is typically used. The imaging apparatus is changed from the fluoroscopic mode o the DF mode. This requires an increase in x-ray tube current of 20 to 100 times the fluoroscopic mode and the activation of a program of pulse image acquisition. Mask mode results in successive subtraction images of contrast filled vessels. The injection is fired and after a delay of 4 to 10 s., before the bolus of contrast medium reaches the anatomic site, an initial x-ray pulsed exposure is made. The image obtained is stored in primary memory and displayed on the video monitor. This is the mask image. The mask image is followed by a series of additional images that are stored in adjacent memory locations, mask image is subtracted from each with the result stored in primary memory. At the same time the subtracted image is displayed on video monitor B. The digital subtraction of the static object allows a better analysis of the opacified arteries especially in their distal parts. The subtracted images appear in real time and are then stored in memory. After the exam, each subtracted image can be recalled for closer examination. Each image is obtained from a 33-ms x-ray pulse. The time required for one video frame is 33 ms. Because the video system is slow to respond and the video noise may be high, several video frames (usually 4-8 frames on average) may be summed in memory to make each image. This process is called image integration. Although this improves contrast resolution, it also increases patient dose because more image frames are acquired.

In mask-mode DF, the imaging sequence after acquisition of the mask can be manually controlled or pre-programmed. If pre-programmed, the computer controls the data acquisition according to the demands of the exam itself. If the computer capacity for acquiring images is sufficient, any combination of multiple delays and varying image acquisition rates is possible

Re-masking


If on subsequent examination the initial mask image is inadequate because of patient motion, improper technique or for some other reason, later images may be as the mask image.
A typical exam may require a total of 30 images in addition to the mask image.


Time-Interval Difference Mode


Some exams call for each subtracted image to be made from a different mask and follow-up frame.
In a cardiac study, for example, image acquisition begins 5 s after injection at the rate of 15 images per second for 4 s. A total of 60 images are obtained in such a study. These images are identified as frame numbers 1 through 60. Each image is stored in a separate memory address as it is acquired. If a TID of four images (268 ms) is selected, the first image to appear will be obtained when frame one is subtracted from frame five. The second image will contain the subtraction of frame two from frame six, the third will contain the subtraction of frame three from frame seven, and so on.

TID mode produces subtracted images from progressive masks and following frames.
In real time, the images observed convey the flow of the contrast medium dynamically.
Subsequent closer examination of each TID image shows it to be relatively free of motion artifacts but with less contrast than mask-mode imaging. As a result, TID imaging is principally applied in cardiac monitoring.

Misregistration


If patient motion occurs between the mask image and a subsequent image, the subtracted image will contain misregistration artifacts. The anatomy is not registered in the same pixel of the image matrix. This type of artifact can frequently be eliminated by reregistration of the mask, by shifting the mask by one or more pixels so that the superimposition of images is again obtained.



Energy Subtraction


Temporal subtraction techniques take advantage of changing contrast media during the time of the examination and require no special demands on the high-voltage generator.
Energy subtraction uses two different x-ray beams alternately to provide a subtraction image resulting from differences in photoelectric interaction. It is based on the abrupt change in photoelectric absorption at the K-edge of the contrast media compared with that of soft tissue and bone.This energy corresponds to the binding energy of the two K-shell electrons of iodine.
lWhen the incident x-ray energy is sufficient to overcome the K-shell electron binding energy of iodine, there is an abrupt and large increase in absorption. Graphically, this increase is known as the K absorption edge. Energy subtraction has a disadvantage of requiring some method of providing an alternating x-ray beam of two different emission spectra.
Two methods have been created, by alternately pulsing the x-ray beam at 70 kVp and then at 90 kVp, and introducing unlike metal filters into the beam alternately on a flywheel.




Hybrid Subtraction


Some DF systems are capable of combining temporal and energy subtraction techniques into what is called hybrid subtraction. Image acquisition follows the mask-mode procedure as described before. However, the mask and each subsequent image are formed by an energy subtraction technique. If patient motion can be controlled, hybrid imaging can theoretically produce the highest-quality DF images.


Patient Dose


One potential advantage to DF is reduced patient dose. DF images appear to be continuous, but in fact they are discrete. Most DF x-ray beams are pulsed to fill one or more 33-ms video frame, therefore, the fluoroscopic dose rate is lower than that for continuous analog fluoroscopy even though the mA setting may be higher. Static images with DF are also made with lower dose per frame than with 100-mm spot film camera. Both the television camera tube and a CCD have a higher sensitivity than the spot film. Digital spot images are so easy to acquire that it is possible to make more exposures than is necessary.



Picture Archiving and Communication System


Digital images come from nuclear medicine, digital ultrasound, digital subtraction angiography, CT, and MRI. Analog images, such as conventional radiographs, can be digitized using a various devices of today. Such film digitizers are based on laser beam technology. A picture archiving and communication system (PACS) when fully implemented, allows not only the acquisition but also the interpretation (soft copy) and storage of each medical image in digital form without resorting to film (hard copy). The projected efficiencies of time and cost are enormous.
The four principal components of a PACS are the image acquisition system, the display system, the network, and the storage system.

Display System


The heart of a PACS display system is either the cathode ray tube monitor of a video workstation or a solid-state flat panel display. To replace film viewing, each must be high resolution, at least 2048 X 2048. Current image matrices used with most digitally acquired images range from 256 X 256 to 1024 X 1024, which is considerably less than that required to equal the spatial resolution of film. PACS display stations are equipped with a keyboard and mouse control for the various image-processing modes. Some relaxation of the spatial resolution requirements of the workstation is allowed because of electronic image processing modes that are available. The simplest example is the ability to pan and zoom and change magnification of the displayed image. Image processing is possible because of the digital nature of the image and the interactive nature of the workstation.

Subtraction of one image from another emphasizes vascular structures.


Edge enhancement is effective for fractures and small, high-contrast objects.


Windowing is useful for amplifying soft tissue differences.


High-lightening can be effective in identifying diffuse nonfocal disease.


Pan, scroll, and zoom allow for careful visualization of precise regions of the image.


For these image-processing to be effective each workstation must be microprocessor-controlled and interact with each imaging device and the central computer. To provide this type of interaction to occur, a network is required.


Network


A network is used to describe the manner in which many computers can be connected to interact with one another. In radiology, in addition to general workstations, the network may consist of various types of images, PACS workstations, remote PACS workstations, a departmental mainframe, and a hospital mainframe. Clients are interconnected, usually by cable in a building, by telephone or cable television lines among buildings, and by microwave or satellite transmission to remote facilities.


The name teleradiology has been given to the process of remote transmission and viewing of images. There has been developed a standard imaging and interface format called DICOM, for Digital Imaging and Communications in Medicine. The network begins at the imaging system, where data are acquired in digital form. The images reconstructed from the data are processed at the console of the imaging system or transmitted to a PACS workstation for processing.
At any time, such images can be transferred to other clients within or outside the hospital.
Instead of running films up to surgery for viewing on a viewbox, one simply transfers the image electronically to the PACS workstation in surgery. When a radiologist is not immediately available for image interpretation, the image can be transferred to a PACS workstation in the radiologist’s home. Time is essential when considering image manipulation, and therefore fast computers and networks with high bandwidth are required for this task.
These requirements are relaxed for the information management and database portion of PACS, which is the Radiology Information System (RIS). From the RIS workstation, any number of coded diagnostic reports can be initiated and transferred to a secretarial workstation for report generation. The secretarial workstation in turn can communicate with the main hospital computer for patient identification, billing, accounting, and interaction with other departments.
These terminals can be utilized for scheduling of patients, technologists and radiologists and for analysis of departmental statistics. Finally, at the completion of an examination, PACS allows for more efficient image archiving.


Storage System


One motivation for PACS is archiving. This saves on films lost, spatial storage issues and film costs. With PACS, a film file room is replaced by a magnetic or optical memory device.
The future of PACS systems depends on the continuing development of the optical disk.
Optical disks can accommodate tens of GB of data and images and, when stored in a jukebox, can accommodate terabytes (TB).


END- Digital Fluoroscopy-Part III


Disclainer: "Capital Medical Equipment, Inc. and Sharon Moore will not be held liable for written materials within this blog or any future blogs. This blog was created as an educational tool and for no other purpose. It will be prohibitted by law to copy, reprint or publish this material without written permission of the author or Capital Medical Equipment, Inc."