Vol 63- No. 4Nov 2007
GUIDE DOG FOUNDATION
For The Blind, Inc.
106th Infantry DivisionAssociation, Inc.
Thank you for helping the
Guide Dog Foundation
for the Blind, Inc.
provide the vital gift of
Second Sight to people
who are blind, and for
sponsoring the dog you
have named "THE CUB."
Grete Bide and Wells B. Jones
CAE. CFRE - Chief Executive Officer
Guide Dog Foundation for
the Blind Inc.
A quarterly publication 61. the 106th InfantryDivish,6 tvwcihtioh, bu. .4 lumprrtfit Organization - St Paul, df \shit,
Alma: John P. Kline, Editor
Paid Membership November 1,418
Membership includes CUB magazine subscription
Life Vets/Associates ... $75 Auxiliary $15
Annual Vets/Associates... $10 Auxiliary $2
Annual Dues payable by June 30 each year
Payable to "Intah Infantry Division Association"
in care of Treasurer. - Sec address below.
PresidentGifford B. Doxsee
1st Vice-PresEdward Christianson
2nd Vice-PresHarry Martin, Jr.
Adjutant: Joseph P. Maloney
1120 Warren Ave, Arnold PA 15068-4048 Tel: 724-335-6104
Treasurer: Lyle Beeth
2004 Golf Manor Blvd, Valrico, FL 33594-7288
Tel: 813-689-9621 Fax: 813-655-8952
Toll Free Number 1-888-644-4337
Note: Addresses given only for the primary person
Chaplain: Dr. Duncan Trueman/ Rev Ewell Black. Jr.
29 Overhill Lane, Warwick. NY 10990
Tel/Fax 845-986-6376 dttrucinanki yahoo.com
Memorials: Dr. John G. Rbbb / Frank Trautman
238 Devore Dr., Meadville, PA 16355
CUB Editor John P. Kline
1987 - 2(107 (Retiring after this issue ‑
See more details in this CUB)
11 Harold Drive, Burnsville, MN 55337-2786
HistorianJohn Schaffner/William McWhorter
Atterbury Memorial Representative Philip Cox
Resolutions ChairmanReverend Ewell Black
Orderof the Golden Lion.... John Swett/Joseph Massey
Nominating Committee ChairmanSy Litchenfeld
Mini-ReunionsHarry F. Martin, Jr./George Call
ADA LiaisonJoseph Maloney/Gifford Doxsee
Membership ChairmanJohn Kline
1st Response ChairmanEllsworth Schanerberger
Board o/ Directors
34 Brae Burn Drive, Purchase. NY 33138
Ethelbarn4ao1.com Website: www.mayrsolin.com
Murray Stein(423/1) (Exec Comm) .. 2008)
7614 Charing Crossing Lane. Delray Beach, FL33446
Dr. Duncan Trueman(424/AT) (2008)
29 Overhill Lane, Warwick, NY 10990
Tel/Fax 845-986-6376 email@example.com
Newton Weiss(423/HQ 3Bn) . . .(2008)
400 Morse Avenue. Gibbstown. NJ 08027-1066
Geo Call(424/B) (2009)
105 Mt. Lebanon Rd. glen Gardner. NJ 08826-3018
Walter C. Greve423/11() I Rn(20(19)
13929 E Manna 1)r ,4,11-1 Aurora. CO 80014
SeymOur Lichtenfeld 422/1 (2009)
19450 NE 21st Ct. North Miami Beach FL 33179
Martin L. Weide 423/1 (Exec Comm) . (2009)
1309 Paseo Valle Vista Covina, CA 91724
Rev. Ewell C. BlackJr. (422/A) (20101
2000 E-W Conn - Apt 212 Austell, GA 3011)6
Edward Christianson(331st MED/C . (201 0)
303 Harper Hollow Lane Winchester. VA 22603
Gifford B. Doxsee(423/1-1Q 3 Bn)(2010)
1 Canterbury Drive Athens. OH 45701-3708
Dr. Ralph Nelson(422/CN).. .....(2010)
10437 Prestwick NE, Albuquerque NM 87111
Dr. John G. Robb(4221D) (2011
238 Devore Drive, Meadeville, PA 16335 Tel: 814-333-6364 firstname.lastname@example.org John (Jack) M. Roberts(592/C). (2011)
1059 Alter Rd. Bloomfields Hills, MI 48304
Tel: 248-338-2667 j inr8 204(Zi1aol.com
John R. Schaffner(589/A)(2011)
1811 Miller Rd. Cockeysville. MD 2103(1-1401
Frank S. Trautman(422/D) (2011)
9 Mcadowcresti Drive Parkersburg, WV 26101
President's View .
Dr. Gifford B. Doxsee
106th Infantry DivisionAssociation
423rd Combat Infantry Regiment
1 Canterbury Drive, Athens, OH 45701
As your new president, I wish to extend my sincere thanks to all who made this honor possible. I shall do my best to serve the 106th
Infantry Division Association to the best of my ability.
Thanks also for making the 2007 Reunion in Kansas City an enjoyable and successful
occasion. Kansas City turned out to be a beautiful city, adorned with numerous parks and fountains. Our hotel accommodations were excellent.
The trips to the Truman Library, the dinner theater, and the Kansas City tour were were all memorable. The presence of Ted Day - Aimed Forces ReunionsDirector, added to the occasion as he and his staff have done so much in recent years to facilitate the reunion planning.
Dr. Duncan Trueman, our Chaplain, delivered another inspiring address at the Memorial Service. I am most grateful to all who helped ease me into the role of Acting President, most
especially to our out going Adjutant, Marion Ray, without whose advice and counsel my role would have been formidable at best.
After extensive discussions at the general business meeting, two important decisions were made.
Following up on John Schaffner's Committee on the future of the organization, the majority view was that we should await the attendance at the 2008 Reunion in Louisville before reaching a conclusion on the future reunions.
Remarks on the location for the 2008 reunion brought out that Louisville seemed to indicate a better choice than San Antonio because its location is much closer to the population center of the attendees; that it offers more options for sightseeing: and that its climate in September is more pleasant than in Texas.
The Galt House in Louisville has recently been renovated and upgraded, and I can now report that the contract with the Galt House for next September has now been signed.
The newly elected officers for the coming year are as follows: President, Gifford Doxsee; First Vice President, Edward Christianson; Second Vice President, Harry Martin, Jr.; Adjutant Joseph Maloney; Treasurer, Lyle Beeth.
Newly elected board members are:
John Robb, Jack Roberts, John Schaffner, and Frank Trautman.
Frank, who - by the way, has recently moved from his longtime home in Parkersburg, West Virginia, to 600 Morningside Drive, Zionsville, IN 46007-1903.
His new home telephone number is 317-732-4371. He and I have both been active for more than 10 years in the Mid-Ohio Valley Chapter, American Ex-Prisoners of War, and he is already greatly missed by our members. 0
The CUB of the Golden Lion 1
Past President's Message .
From Past-President Murray Stein2007-2008
106th Infantry DivisionAssociation
423rd Combat Infiintry Regiment, "1" Company
7614 Charing Cross Lane, Delray Beach, FL 33446
Phone: 561-499-7763 greg0803,1c) omcast.net
Unfortunately due to a severe illness, I was unable to attend our reunion in Kansas City, I am doing well and already looking forward to Louisville in 2008.
Our new president Gifford Doxsee, his officers and Board of Directors were
elected in Kansas City and our Association is in good hands for this coming year.
As your president for the past two years, I experienced many interesting and hurtful moments: the passing of our incoming president, Irwin Smoler, the loss of our friend and Treasurer, Richard Rigatti, Don Herndon's serious accident and the loss of more than one-hundred of our Brothers.
I was privileged to be surrounded by so many good men - especially our officers, Adjutant Marion Ray; Editor John Kline; Vice-President "Chic" Wente; Chaplain Duncan Trueman; Historian John Schaffner, Treasurer Lyle Beeth. Also many other good friend including but not limited to Jack Roberts,Sy Lichtenfeld,Dr., John Robb,Dr. Ralph nelson. Reverend Ewell Blackand numerous others. I was pleased and impressed in getting to know better our new president Gifford Doxsee.The Association is in good hands, I have been truly blessed by my association with these beautiful people as well as to have been honored to serve as president of the 106th Infantry DivisionAssociation.
During the past two years I was also involved in the wreath laying in Saint Vith, Belgium, The Camp AtterburyMuseum and James West, Associate member. Our contact with the Indianapolis archives and a number of conversations with friends of the 106th.
We know our Association is in good hands with President, Dr. Gifford Doxsee. I look forward to seeing you all in Louisville, Kentucky next year.
NEWS about our guide dog: "The CUB."
The cover of this CUB magazine shows the original photo of the dog we furnished to the Guide Dog Foundation for the Blind.
He is of course much more grown by now.
I would like to recite two letters obtained during this period:
One from the Department of Veteran's Affairs in West Palm Beach, Florida and the other from the "GUIDE DOG FOUNDATION."
First from the VA:
A letter to our Treasurer, dated March 21, 2007:
"Dear Mr. Beeth and esteemed members of the 106th,
On behalf of the Department of Veterans Affairs and the Guide Dog Foundation for the Blind, let me take this opportunity to thank the members of your group for your very generous donation of $6,000 for the Disable Vet Dog Program (DVDP).
(continued next pti Lte
The CUB of the Golden Lion
Past President's Message . .
Mr. Murray Steinleft specific instructions that the dog purchased by your members should be named, "The CUB."
A copy of this letter (and the check) is being forwarded to the Guide Dog Foundation so that every effort might be made to recognize and honor your members in this manner.
Because of your efforts, a blinded veteran from the West Palm Beach area will be provided a new chance at independence and an imnproved quaility of life.
This program began in 2005 and has placed 12 dogs so far, with blind veterans in the Florida area.
Chief, Blind Rehabilitation Service
West Palm Beach VAMC
Now the letter received from the Guide Dog Foundation for the Blind, Inc.
October 3, 2007 Dear Mr. Stein:
It's time time for a progress report on "The CUB." He's growing by leaps and bounds! He's almost full adult height, although he'll be filling out until at least the age of 2 years.
Now that the public excursions are allowed, the CUB is learning about "The Big World" beyond the front door. Wearing the bright puppy coat that identifies him as a future guide dog, he and his puppy waker, Terry, visit public places to gain exposure to sights and sounds.
Altough a six month old puppy is still very much an adolescent, Terry is working hard to teach "The CUB" good manners, both in public and at home.
The last four months have been a period of great growth and change for him. We hope that he'll continue to mature into a good candidate for formal guid dog training.
We will write agin when he is ten month months old. Until then, best regards from every one here!
Director of Canine care
Thanks again for your support and trust furing my term of office. I salute you all
Past-President 2005 - 2007 Fiscal Years. 106th Infantry DivisionAssociation
The CUB of the Golden Lion
Chaplain's Message .
"OUR HEARTS WERE
TOUCHED WITH FIRE"
Great pride should move through our souls as we come together this morning for this memorial service. Not just because our thoughts are focused upon a war long past but also because we think of so many brave comrades who fought those battles with us.
Sometimes our battles were fought against guns and steel and armor. Sometimes they were fought against imprisonment and hunger and cruel indignity. Always our battles were fought against bitter cold, against pain and suffering. Whenever we think of those times and those places and those comrades, we ought to be filled with pride.
On the other hand, whenever we think of American soldiers at any place or at any time, we ought to be filled with pride.
There's such sadness in our hearts as we meet today. We're sad to think of those companions who have passed on during this last year. They are the ones we come today to honor and memorialize.
And we're also sad because for still another year our minds are filled with images of suffering and struggle of this present generation of young Americans ... images that flash up on our TV screens each and every day ... images of youngsters hoping to survive, hoping to become old veterans like us some day.
So we're sad not just at the loss of our comrades we have known for years; we're also
sad because the very unique bond that has always existed between us as combat veterans, now begins to include our own grandchildren.
We had hoped that would never be so.
Tim Dyas of the 101st Airborneexpressed at this way in his poem:
Combat soldiers on different sides, as the war's memories become just that, come closer
together and in a circle, shut out those who were never there.
Not done with the intent, but only because men who faced each other in battle remain different
from all others in their lives. Thus the Blue and the Gray were united because of Antietam
and Gettysburg,the khaki and field gray entered the circle because of the Argonne and
Verdun. Then Guadalcanal and Normandybrought more within the circle. And later army
.fatigues and black pajamas moved inside,
Those inside, mostly ghosts now, pray that no more will ever enter,
so that the circle shall become ever smaller.
Ever smaller was not to be!
Do you recognize yourself in some of those images that you constantly see? Those soldiers look somewhat different, but inside they are the same as you and I. Their uniforms have changed. Their weapons are more deadly. The vehicles are different from our canvas tops. But, inside they are the same in every respect, and in one respect in particular: "Their hearts are touched with fire!-
Chaplain Dr. Duncan Trueman424/AT
106th Infantry DivisionAssociation
729 Overhill Lane Warwick, NY 10990
Tel/Fax: 845-986-6376 email@example.com
The CUB of the Golden Lion
Now that phrase is not mine, It was coined by a man who knew the meaning of combat, a man three times wounded in the Civil War. He later became a great jurist on the United States Supreme Court. His name was Oliver Wendell Holmes.
He never forgot the soldiers he had known on the field of battle.
Whether they fought for the North or for the South did not matter. They were part of that inner circle known only to those who have been there. He held them all in the same high regard for they had all fought for a cause in which they believed.
He saw all of them as heroes and he said of his generation ... all of them:
"In our youth are hearts were touched with fire,
It was given to us to learn at the outset that life
is a profound and passionate thing."
I believe that fire is a distinctly American quality, totally unlike the fanatical beliefs of enemies we once faced, and totally unlike the fanatical beliefs of enemies encountered today. This fire that Holmes spoke is in the American heart and it burns not out of hatred or intolerance, but purely out of love for home, country and freedom.
It was in the hearts of those fought for freedom at Lexington and Concord. It was in the hearts of those who defended us once again during the War of 1812. It in the hearts of both North and South during the Civil War. In was in the hearts of the Rough-Riders. It was in the hearts of the doughboys at the Argonne and at Verdun.
It was in the hearts of our generation who fought at Pearl Harbor,at Anzio,in the Solomons, and Normandy, in the Ardennes, at Iwo Jima. It was still there in Koreaand in Vietnam; yes and in every battle and skirmish were our troops have gone. And today we observe it day after day in Afghanistanand Iraq. "Hearts that are touched with fire."
On June 12, 1944 Ernie Pylewrote these words about the D-Day invasion: "It seems to me a pure miracle that we ever took the beaches. We did it with every advantage on the enemy's side and every disadvantage on ours ... We sit and talk and call it a miracle that our men ever got on at all, or were able to stay on."
Years later it was Stephen Ambrosewho was able to evaluate the same landing in a more pragmatic way. He wrote: "It was not a miracle. It was infantry. As is almost always the case in war, it was up to the Infantry."
The exact same judgment was made about the Battle of the Bulge. Eisenhower, Bradley, Marshall, and most historians since, have reiterated the same refrain. "It was not the high command, the tactics developed by the generals that turned defeat into victory ... it was the ordinary soldier, the infantryman, who fought like hell against the odds.
When officer leadership was lacking, the enemy often lost efficiency, could not make decisions, floundered indecisively. Not so with the American infantryman. If a leader was needed, someone became one. Their hearts were touched by fire.
General Ike once said: "What counts is not the size of the dog in the fight, but the size of the fight in the dog." Outnumbered as we were, we were the far smaller dog, though we probably knew it not, our hearts were touched with fire.
What does that mean ... fire? To wives, children, grandchildren and others, let me say: Look around at the men here. Look right. Look left. Wives, look at your husbands, children look at your fathers, grandchildren at your grandfathers ...
You cannot by the wildest stretch of your imagination imagine what your father, your husband or your grandfather did on that bitter-cold and deadly field of battle 63 years ago.
You cannot imagine some of the deeds performed, some of the risks run, the sacrifices made, the pain endured, the grief experienced. You cannot imagine what most have tried to forget. But through it all the fire in their hearts was what helped them fight, helped them win, and helped them survive.
The CUB of the Golden Lion
Chaplain's Message .
In MRI, the negative contrast generated by susceptibility effects is widely used to visualize objects and to study processes. In this context, the relaxation effects of systems of infinite cylinders or microscopic spheres have been studied to model susceptibility effects in a capillary bed (1, 2) and a suspension of blood cells (3, 4). Until now, little attention has been given to the appearance of a small individual object of deviating susceptibility. However, insights into mesoscopic susceptibility deviations would be relevant to the presentation of, e.g., iron deposits (ferritin), embolic metal fragments (5), paramagnetic markers on interventional devices (6), seeds used in brachytherapy (7), clusters of magnetically labeled cells (8, 9), and air bubbles.
In this work, the susceptibility artifact of a subvoxel object in gradient-echo imaging, assuming an otherwise homogeneous environment, is studied. We introduce the dephased-volume model, which simplifies the effect of a subvoxel susceptibility deviation to a localized suppression of signal. The size and positional accuracy of the dephased volume in relation to scan parameters and object composition are determined by 3D time-domain simulation of the imaging process and by in vitro experiments.
The magnetic volume susceptibility χ of a material (henceforth termed “susceptibility”) is a dimensionless quantitative measure of the material's response to an applied magnetic field. Susceptibility differences are a major source of B0 inhomogeneities. These field inhomogeneities ΔBz(r) lead to slice distortion and signal displacements in the frequency-encoding direction, intensity distortions and, in gradient-echo imaging, signal loss due to intravoxel dephasing and echo shift from the center of the acquisition window (10, 11).
The phase dispersion at the echo time (TE) of the magnetization M(r) that is mapped into the same voxel changes the signal I according to:
Define a subvoxel object as one that will fit in a sphere with radius d, the in-plane pixel size. Then, for a subvoxel object with volume V, with a susceptibility difference with the surrounding tissue Δχ ≪ 1, the inhomogeneous part of the field disturbance for ρd > 1 corresponds to the field of a magnetic dipole moment m = Δχ VB0/μ0 (12), given by:
using a polar coordinate frame. Here, ρd denotes the radial coordinate normalized to the pixel size. From Eqs.  and  it is directly apparent that the intravoxel dephasing at TE is determined by:
Physically, K is equal to the phase at the TE for ρd = 1, in the equatorial plane (z = 0).
The premise of the dephased-volume model is that intravoxel dephasing caused by the phase accumulation at the TE (see Eq. ) overshadows geometrical distortions. This condition is largely met for gradient-echo imaging of a subvoxel object in a thick imaging slice. In that case, the dispersion of the frequency distribution is primarily attributable to the voxel's extent in the slice direction (13, 14), and in-plane variations can be neglected. All of the elongated voxels will contain on-resonance spins, because ΔBz(r) falls off rapidly to zero. Thus, any magnetization with a phase accumulation at TE, ΔϕTE, such that |ΔϕTE| = |γ ΔBz(r) TE| > 2π will find some magnetization somewhere in the slice direction that cancels its signal, and is part of the dephased volume (Fig. 1). Since the dephased volume will contain the near field, ρd < 1, of the object, the exact shape of ΔBz(r) near the object is unimportant. Then, since only regions with |ΔϕTE| < 2π contribute to the voxel signal, geometric distortions are limited to |Δr| < 2π/γ TE GR, with GR being the readout gradient strength. This limit is less than a pixel for full-echo acquisitions. In summary, total signal loss is expected to scale with K, to be insensitive to in-plane resolution, and to be independent of geometrical distortions. In addition, the position of the artifact is expected to accurately reflect object position.
Simulations were performed to study the susceptibility artifact in the absence of system nonidealities, and to study the dependence on parameters that cannot easily be varied. A 3D supersampled time-domain simulation program, similar to that described in Ref. 15, was implemented. The field disturbance was introduced using the analytical expression of Eq. .
Proton imaging, γ = 2π · 42.576 106 rad/s/T, at 1.5 T, was simulated using the following scan parameters: field of view (FOV) = 256 mm, imaging matrix = 256, a full echo centered at 9.2 ms, GR = 5.2 mT/m, slice thickness = 20 mm, and a slice-selection pulse with a bandwidth of 2000 Hz. In the default slice orientation, the readout direction was parallel to B0. The object was a sphere with a radius of 0.5 mm, with Δχ = 860 ppm, giving Δχ V = 450 · 10–6 mm3.
Starting from these settings, the parameters were systematically varied: TE ranged from 1.5 to 25 ms for a full echo, and from 0.5 to 9.2 ms for a 62.5% fractional echo, increasing GR when necessary. Three slice orientations were examined: a coronal slice (B0 in-plane) with readout respectively parallel or perpendicular to B0, and an axial slice (B0 through-plane). For these orientations the effects of readout gradient polarity and fractional echo were studied. Slice thickness was varied from 5 to 50 mm, FOV from 128 to 512 mm, and asymmetric echo fraction from 62.5% to 100%. GR was varied between 1.7 and 10.4 mT/m. In addition, a fully phase-encoded image was simulated (i.e., with no geometric distortion).
The effect of increasing the susceptibility of the sphere was examined, with Δχ V ranging from 100 to 4000 · 10–6 mm3. Finally, the dephasing parameter K was varied over four decades (0.01–100), assuming an infinitesimally small object.
In Vitro Experiments
Imaging was performed on a 1.5-T system (Intera; Philips Medical Systems, Best, The Netherlands), using a 22-cm circular coil for signal reception. A small Dy2O3 ring marker from a 1.7-mm diameter catheter, Δχ V = 450 · 10–6 mm3, was embedded in agar. Scan parameters were systematically varied as described in the Simulations section, if this was permitted by scanner software and hardware. The slice was centered at the marker. We used a flip angle of 10° and a default repetition time (TR) of 20 ms. In addition, the orientation of the imaging slice was varied in steps of 15° from coronal (B0 in-plane) to axial (B0 through-plane).
In the images, integrated signal loss and artifact center of mass were evaluated. In the simulations, the signal of the undisturbed object was known, and in the in vitro experiments the average signal of an undisturbed area was used as a reference. Integrated signal loss, Vloss, was obtained by summing the signal loss over the region of interest (ROI). The center of mass of the artifact was calculated by integrating the product of position and loss over the ROI; only pixels with a negative signal change were included in the calculation.
The relation between Vloss and K was determined by linear regression through the origin. To this end we only used data with K < 300, obtained in the simulations that varied Δχ V, K, and TE, and in the experiments that varied TE, all using full echoes. Coefficients of proportionality are given ± 95% confidence limits.
Simulations and experiments showed that artifact size increases with TE (Fig. 2a): Vloss was nearly linearly related to TE, and thus to the dephasing parameter K. For TE longer than 10 ms, Vloss is smaller than expected from a linear relation. Simulated and experimentally obtained Vloss correlated well.
Figure 3 shows how the artifact changes with gradient polarity and using fractional echo for three different orientations. For the full-echo acquisitions, the images are highly similar to the projections of the dephased volume in Fig. 1. For fractional-echo acquisitions, however, the shape of the artifact changes by echo shifting of the signal from regions where the gradient of the field disturbance has the same polarity as the readout gradient. This one-sided signal loss slightly shifted the center of mass of the artifact and contributed to Vloss, which was up to 20% larger for fractional-echo acquisitions.
Experimentally, Vloss was independent of the orientation of the readout gradient within 3%, except that for fractional echo a 7% increase was caused by swapping readout and phase-encoding directions. Likewise, for the settings examined, Vloss was independent of the readout gradient strength (3% variation for full echoes). Although displacements within the artifact were apparent at reduced gradient strengths (Fig. 3), the shift of the center of mass remained well within 0.5 mm. Vloss was hardly reduced for higher in-plane resolution (12% variation from 0.5 to 2 mm).
Signal loss was slightly larger for a transverse slice than for coronal/sagittal slices (Fig. 2). A peak is apparent for intermediate angles. This peak is related to an increased overlap of the poles and the toroid, which contain phases of opposite sign, resulting in increased intravoxel dephasing.
The shape of the artifact varied with slice thickness (Fig. 4): for slices over 10 mm, the shape of the artifact resembled the projection of the dephased volume in Fig. 1. Poles and toroid were unrecognizable when the thick-slice assumption was violated (e.g., for a 2.5-mm slice). Vloss increased until the slice thickness was, in this case, about 20 mm and then leveled off (Fig. 2c). Thus, for thin slices the dephased volume is only partly contained in the imaging slice, and hence increasing the slice thickness will pick up more dephasing until the dephased volume is completely contained in the slice. Increasing the slice thickness beyond this point will not improve the detectable effect of the object.
The simulation allowed us to evaluate the extent of the artifact for a range of Δχ V, i.e., the magnetic content of the susceptibility deviation. Figure 2d shows that Vloss is nearly linear in Δχ V, and thus in the dephasing parameter K, until the dephased volume extends outside the imaging slice.
Down to K ≈ 10, the shape of the artifact could be clearly recognized. For K ≈ 1, the artifact still extended to neighboring pixels. Smaller K values reduced the intensity in only one pixel.
The relation between Vloss and K for the data obtained in simulations is given by Vloss = (3.4 ± 0.2) K d3, r = 0.998 (N = 14). The experimental results obtained by varying TE were characterized by Vloss = (3.1 ± 0.3) K d3, r = 0.997 (N = 6).
Susceptibility-based negative contrast provides a way to amplify the effect of objects that are too small to be detected easily using MR—for example, clusters of cells (8, 9), lymph nodes (16), and interventional devices (6). For these applications, reliable negative contrast and positional accuracy are desired.
In this study, the dephased-volume model is used to describe the artifact of a subvoxel susceptibility deviation in gradient-echo imaging using a thick slice. In this model the artifact is viewed simply as a localized removal of the signal from the vicinity of the object. The size of the artifact was demonstrated in simulations and in in vitro experiments to scale with the dephasing parameter K: Vloss, the volume of dephased signal (in mm3) was nearly linearly proportional to K d3. In addition, the signal loss was well localized, and shown to be within 0.5 mm (0.5 d) of the object for practical settings. The shape of the artifact, for full-echo acquisitions, resembles the projection of the dephased-volume in the slice direction (Fig. 1).
Vloss being proportional to K, the most important parameters are the para-/diamagnetic content of the susceptibility deviation, Δχ V, and the TE. The increase of artifact size with TE in gradient-echo sequences has long been recognized (13, 14, 17). Neither phase-encoding direction nor readout gradient strength (scan parameters that have been shown to strongly influence artifact appearance in spin-echo imaging (10, 18)) had an effect on the size of the artifact in this case. Likewise, the dependence of signal loss on in-plane resolution, which has a strong influence in nearly isotropic 3D gradient-echo scans, is almost absent. The independence of Vloss on these parameters bears on the thick slice condition that makes intravoxel dephasing dominant over geometrical distortion.
The exact shape of the object with deviating susceptibility appears to be immaterial when |χi| ≪ 1, as long as the susceptibility deviation is small compared to the pixel size. This is supported by the close correspondence between the simulations and in vitro experiments that were performed using a 0.5-mm sphere and a thin 1.7-mm diameter ring.
Three parameters, which are not included in K, may significantly change Vloss: c.q. slice thickness, fractional-echo acquisition, and slice orientation. When the thick-slice condition is violated, the dephased-volume model no longer applies. Phase dispersion in-plane becomes important, and the dephased volume will only be partly contained in the imaging slice. Fractional-echo acquisitions change the shape and slightly displace the center of mass of the artifact. The signal that is preserved at the zeros of the dipole field for full-echo acquisitions is shifted outside the acquisition window by the nonzero local gradient (11). Similar effects are expected for half-matrix acquisition in the phase-encoding direction.
The orientation of the imaging slice affects the artifact shape, which can be imagined as taking different projections of the dephased volume (Fig. 1). In going from coronal to transverse, the toroid becomes wider and finally becomes a circle. The poles will increasingly overlap the toroid, and because poles and toroid are of opposite polarity, this overlap will result in a slightly increased signal loss.
Visualization methods that employ negative contrast benefit from a high background signal. Therefore, scan parameters should be optimized so as to yield the maximum possible signal from the tissues surrounding the object. Thus, although TR, flip angle, readout gradient strength, gradient moment nulling, etc., do not influence the overall signal loss in mm3 that relates to the extent of the region void of signal, they do affect the conspicuity of the object by changing the signal yield of background tissue.
A quantitative relation between object composition, i.e., Δχ V, scan parameters, and artifact size, can be used to narrow down the possible sources of artifacts encountered (5). In addition, such a relation allows one to design susceptibility markers that provide the desired artifact size.
The constant of proportionality relating Kd3 and Vloss has been determined to be between 2.8 and 3.6. This value is considerably larger than the volume bounded by the |ΔϕTE| = 2π surfaces used in introducing the dephased volume, which is 8√3/27 Kd3 ≈ 0.51 Kd3. Clearly, magnetization with |ΔϕTE| < 2π contributes significantly to the signal loss. On the other hand, Vloss is significantly smaller than if the magnetization in a dipole field were summed over all of space—that is, as if the image consisted of one big voxel (4, 19): 8π2√3/27 Kd3 ≈ 5.06 Kd3. Integration of the complex signal change over the ROI in the simulations obtained by varying Δχ V and TE resulted in Vloss = (4.9 ± 0.1) Kd3. The difference between Vloss on magnitude images and the loss predicted by Yablonskiy and Haacke (4) is related to the spatial separation of positive and negative frequencies in the dipole field. This separation reduces intravoxel phase dispersion, especially in slices parallel to B0. Note that for slices at an angle of 45–60° with B0, frequency separation is reduced and Vloss comes close to the value predicted in Ref. 4. For long TEs and high Δχ V (see Fig. 2a and d), Vloss is noticeably less than proportional to Kd3; apart from the spatial separation of frequencies, the slice in those cases will not completely encompass the dephased volume. The nearly linear dependence of Vloss on K, with the constant of proportionality determined, allows one to estimate Δχ V from the artifact. However, because the object volume is not easily determined, one cannot distinguish between objects with the same Δχ V product.