In today’s health care environment, time truly has become money. With capitated contracts and managed care, all aspects of an imaging department need to be scrutinized for cost effectiveness.  Within the MRI suite an easy way to improve throughput and productivity is to minimize the time a patient spends in the scanner.  Decreasing patient scan time can be done through various methods. Certainly, the easiest is to eliminate sequences.  Unfortunately, eliminating sequences could decrease pathologic information, thereby compromising patient care.

We have focused our efforts on decreasing the number of NEXs (time) and then post-processing the images with the Image Enhancement System (IES).  Remember, scan time is proportional to the number of NEXs, therefore any decrease in NEXs has a corresponding decrease in scan time.  IES is an aftermarket hardware and software package designed to improve image quality through a variety of proprietary means.

To test the efficacy of the IES system we scanned patients with our routine protocols and then with fewer NEXs. For example, when the routine called for 4 NEXs, the new sequence was reduced to 2 or 3 NEXs.  After acquisition, the new images (fewer NEXs) were post-processed using the IES system.  The post-processed enhanced images were then compared to the original images and evaluated by two radiologists.  The radiologists compared the images side-by-side to determine which of the two images had the highest image quality. In an effort to produce a more accurate result, they were not told which image was which.

After all the images were evaluated, 56% of the post-processed images (those with fewer NEXs) were found to be better than the originals (those with more NEXs).  Of the images that were not better, 25% were scanned at 1 NEX versus 2 NEXs and then post-processed.  Using this data, we have chosen to modify only those sequences with originally more than 2 NEXs.  Since the conclusion of the evaluation process, we have been able to modify most of our protocols and decrease our examination times by approximately 5 minutes per patient.  With an average patient load of ten patients per day we are now able to scan an additional patient each day, without extending operating hours or increasing operational costs.

Obviously, everybody’s time savings and subsequent cost savings will differ; yet, the reality remains that post-processing MR images, improves both image quality and patient throughput.  This post-processing system (IES) is an easy and cost effective method to improve MRI efficacy.  Lastly, while this paper has focused on the ability of post-processors to minimize time, it is important to remember that they have a variety of other applications that further enhance their utility.

To evaluate the clinical utility of improving image quality using an external image enhancement system on a high field MRI scanner.

25 patients who underwent MR imaging on a high field MR scanner (GE Signa 1.5 T Horizon) for either MRA or high resolution fast spin echo (FSE) brain imaging had their studies post-processed off-line using the FDA cleared Image Enhancement System (IES, Hayward CA). Comparison of 256 x 256 matrix imaging versus 512 x 256 or 314 x 512 or 512 x 512 matrix keeping the same imaging time (NEX) were performed on enhanced and unenhanced images.  Pre- and post-enhanced MRA images were also compared.

The Image Enhancement System improved MR image quality by increasing the conspicuity of flow-related enhancement in MR angiography and allowed for an increase in matrix size when using high resolution FSE.

These improvements were obtained without the need to increase the NEX (time) by using a two-step image analysis post-processing procedure.  The first step analyzes images in terms of location, orientation and curvature of structures.  
The second step uses an advanced filtering process to make the image sharper while reducing image noise at the same time.
This post-processing procedure is carried out utilizing an external CPU, which can interface to a variety of MR scanners.  
No lesion boundary was lost due to the post-processing, however decreased conspicuity of motion artifacts was achieved.

Post-processing of MR images can improve image quality by removing unwanted image noise, increasing the signal-to-noise ratio, and enhancing image detail by increasing the conspicuity of flow related enhancement on MRA scans.  The clinical utility of the IES system allows for an increase in matrix size and spatial resolution, without a need to increase the NEX or without paying a penalty in time.

Since their inception, post-processing computers have helped improve the quality of MRI images.  The emphasis however, has been on making good images better.  We used the Image Enhancement System (IES, Hayward CA) to determine whether or not image quality on images exhibiting motion artifacts could be improved.  The primary purpose was to ascertain if post-processing might improve image quality enough to eliminate the need for rescanning those patients with motion.


 METHODS

A random sample of images, comprising a range of motion from completely useless to minimal physiologic motion was selected to be in the study.  The rationale for including a range in types of motion artifacts was that post-processing might work best within a given range thereby improving efficacy of the MR technique.  Unidentified pre- and post-processed images were given to two radiologists who were asked to rate their overall quality on a scale of one to ten.  A score of one corresponded to the worst possible image quality and ten the best.

Once the radiologists rated the films, they were analyzed for improvement in image quality.  Of the images rated, an average of 44% demonstrated an increase in image quality after post-processing.  The percentage increased to 56% when the worst images (those rated initially below a 3) were dropped from the analysis and increased to 60% when both the worst and best (initially rated over 8) images were dropped.  An added benefit of using the IES was that the image quality is improved on all types of examinations and sequences.

The use of the IES post-processing system improves the overall image quality of MR images with motion artifacts.  As a result, we have experienced a drop in the number of repeat sequences required due to patient motion.  Consequently, IES post-processing has allowed our MRI department to improve efficiency and throughput without affecting our examination times or staffing.  Any MRI department should be able to recognize similar benefits using a post-processing enhancement system.  The only difficulty may be in determining which images are suitable for enhancement and which should be rescanned.  After a brief acclimation period this problem should be easily overcome.  In summary, it would appear that a post-processing enhancement system is an easy way of improving both image quality and patient throughput, without sacrificing patient care.

Fifty patients who underwent MR imaging on a high field MR scanner (GE Signa 1.5 T Horizon) for either MRA, high resolution 2D, 3D or single shot fast spin echo (FSE) imaging had their studies post-processed off-line using the FDA cleared Image Enhancement System (IES, Hayward, CA).  The clinical applications of the noise reduction algorithm in craniospinal, musculoskeletal and body MRI will be presented.

Using an Image Enhancement System allowed for an increase in matrix size, smaller fields- of-view, reduced slice thickness and faster scan times.  The conspicuity of flow related enhancement on MRA scans was also increased. There was no lesion boundary loss, and decreased conspicuity of motion artifacts was achieved.  These enhancements were obtained by using a two-step image analysis post-processing procedure.

Post-processing MR images can improve quality by removing unwanted noise thereby increasing the signal-to-noise ratio.  The clinical utility of the system allows for an increase in matrix size and spatial resolution, without a need to increase the NEX and pay a penalty in time.

To demonstrate the clinical benefits of using an external image enhancement system in high field spinal MRI with high performance gradients.

MRI of the spine often suffers from noisy images because of 1) patient size affecting surface coil loading, 2) patient motion, 3) newer high resolution and volumetric acquisitions with faster gradients.  Moreover, utilization of the body coil and not the surface coil due to patient size or the need for flexion and extension views places a demand on signal-to-noise.  This exhibit demonstrates the indications for using a noise reduction algorithm and presents clinical examples where utilization of the FDA cleared Image Enhancement System (IES, Hayward, CA) aided or made a difference in the clinical diagnosis.  The IES utilizes a two-step image analysis post-processing procedure, which makes the images sharper and reduces image noise, without loss in lesion boundary.

IES increased the signal-to-noise ratio and conspicuity of lesions due to better edge definition, when the body coil had to be used for cervical spine MRI on patients with severe kyphosis or large size.  Flexion and extension views were optimally performed using the combinations of IES and body coil.  IES enhanced the diagnostic confidence in pediatric spine MRI, where patient motion and sedation issues becomes a time limiting factor.  3DFSE imaging of the craniocervical junction with reprojection images benefited from IES by increased conspicuity of pathology and normal anatomy.

IES was instrumental in optimizing spine MRI quality of pediatric studies, large patients, volumetric acquisitions, reprojection images and in situations where a surface coil could not be used due to technical or anatomical limitations.

Song, I.S.M., Naple, Glover, G.H., Pelc, N.J., Noise reduction in three-dimensional phase contrast MR velocity measurements. JMRI 1993:3:587