M e d i c a l P hy s i c s a n d I n f o r m a t i c s • O r i g i n a l R e s e a r c h

Ogura et al.
Importance of Fractional b Value for Calculating ADC

Medical Physics and Informatics

Original Research

Importance of Fractional b Value

for Calculating Apparent Diffusion
Coefficient in DWI
Akio Ogura1 OBJECTIVE. The purpose of this study is to examine how the fractional b value affects
Isamu Hatano1,2 the calculation of apparent diffusion coefficient (ADC) using DWI. The fractional b value is
Kohki Osakabe1,3 the point of intersection between the fast and slow components of biexponential decay in DWI.
Natsumi Yamaguchi1,4 SUBJECTS AND METHODS. Human brains were imaged using multiple b values on
Daisuke Koyama1,5 echo-planar DWI. The ADCs of white matter, gray matter, and thalamus were calculated us-
ing the combination of b values by two-point and multipoint methods, and the characteristics
Haruyuki Watanabe1
of each ADC value were compared.
American Journal of Roentgenology 2016.207:1239-1243.

Ogura A, Hatano I, Osakabe K, Yamaguchi N,
Koyama D, Watanabe H
RESULTS. When the two selected points for calculation were smaller than the fractional
b value (b = 1700 s/mm2), the ADC value was 0.0007–0.0008 mm2/s, but when the two points
used for calculation were greater than the fractional b value, the ADC value was 0.0003–
0.0004 mm2/s. When a range of b values was included in the fast and slow components by
use of the multipoint method, the ADC value showed a statistically significant increase as the
number of multiple b values increased.
CONCLUSION. The ADC value fluctuated when the b values used for calculation were
higher than the fractional b value. Therefore, it is important to determine the fractional b val-
ue of the target tissue.

WI is a very useful tool in clini- ADC is calculated by using two or more b

Keywords: apparent diffusion coefficient, biexponential,
DWI, fractional b value, multipoint method
DOI:10.2214/AJR.15.15945
Received November 25, 2015; accepted after revision
June 2, 2016.
1
Graduate School of Radiological Technology,
Gunma ­Prefectural College of Health Sciences,
323-1, Kamioki-machi, Maebashi, Gunma 371-0052,
­Japan. ­Address ­correspondence to A. Ogura
(a-ogura@mbox.kyoto-inet.or.jp).
2
Department of Radiology, Jichi Medical University
Hospital, Tochigi, Japan.
3
Department of Radiology, Gunma Saiseikai Maebashi
Hospital, Gunma, Japan.
4
Department of Radiology, Tokyo Metropolitan Tama
Medical Center, Tokyo, Japan.
5
Department of Radiology, National Hospital ­Organization
Matsumoto Medical Center, Nagano, Japan.
AJR 2016; 207:1239–1243
0361–803X/16/2076–1239
© American Roentgen Ray Society
D cal diagnosis. First, the use of
the apparent diffusion coeffi-
cient (ADC) in the detection of
acute ischemic stroke was a significant de-
velopment [1, 2]. Thereafter, owing to the
improvement in technical aspects of calcu-
lating ADC, which allowed its application to
body diffusion, ADC was used in the diag-
nosis of malignant tumors and inflammatory
diseases [3–5].
In addition, ADC values calculated from
DWI can be used for the differentiation of tu-
mors. The clinical use of ADC now includes
improved tissue characterization for moni-
toring treatment response after chemother-
apy or radiation and differentiation of post-
therapeutic changes from residual active
tumor. Many studies have been conducted
using ADC to distinguish malignant tumors
from benign tumors, particularly in the field
of pharmaceutical drug development, and to
predict therapeutic efficacy using MRI [3–
25]. It has been reported in multicenter tri-
als that the ADC fluctuates according to the
b values used for calculation and other imag-
ing parameters [26–31].
values derived from DW images. The ADC
is calculated on the basis of the difference in
the signal intensity on DW images at two b
values according to monoexponential decay.
The utility of the multipoint method for cal-
culating ADC from three or more b values
has also been reported [32, 33].
However, the diffusion decay curve of
the human body is not monoexponential. A
biexponential fit should be a basic require-
ment, considering that DWI of the human
body requires restricted diffusion [26, 34–
38]. In this case, the ADC value is known
to vary along with b values used for calcu-
lating the ADC [26, 27]. The biexponential
model is identical to nongaussian diffusion
[39]. Tumors are graded through analysis of
nongaussian diffusion [40, 41]. In contrast, it
has been reported that if the same b values
are used for calculation, the ADC values are
the same, independent of the vendor mod-
el or static magnetic field intensity [42, 43].
In other words, b values are very important
factors for calculating the ADC. The diffu-
sion decay curve of biexponential decay is
shown as follows: the point of intersection of

AJR:207, December 2016 1239


Fig. 1—22-year-old man. MR image shows how
American Journal of Roentgenology 2016.207:1239-1243.
5-mm-diameter ROIs were set at locations of white
matter, gray matter, and thalamus.
the fast and slow components of biexponen-
tial decay is called the fractional b value. For
biexponential diffusion decay, the fractional
value may be considered as the b value for
ADC calculation.
The purpose of this study was to under-
stand how the ADC value was influenced by
the association between fractional b values
and the b values used for calculating the ADC.
Subjects and Methods
Imaging and Analysis
The institutional ethics review board of Gunma
Prefectural College of Health Sciences approved
this prospective study, and written informed con-
sent was obtained from all volunteers recruited. We
prospectively examined eight healthy volunteers
(four men and four women; age range, 22–55 years).
These eight volunteers responded to an open call
for participants from among students and teach-
ers at the university. Thus, the subjects may have
been younger than typical patients. The volunteers
underwent echo-planar DWI of the brain, using a
head coil with a 1.5-T clinical MRI system (Inge-
nia, Philips Healthcare), with multiple b values on
slices parallel to the anterior commissure–posteri-
or commissure line. The imaging parameters were
TR/TE of 3500/103, FOV of 300 mm, slice thick-
ness of 8 mm, phase-encoding matrix of 120, sensi-
tivity encoding factor of 2, and two excitations. The
b values of the motion-probing gradient were the
following 16 points: 0, 300, 500, 700, 1000, 1300,
1500, 1800, 2000, 2300, 2500, 2800, 3000, 3500,
4000, and 5000 s/mm2. ROIs (5 mm in diameter)
1240 AJR:207, December 2016
Ogura et al.
1 intensity of white
Signal Intensity (Sb/S0)
0.1
White Matter
Gray Matter
Thalamus
0.01
0 1000 2000 3000
b Value (s/mm2)
encompassing the white matter, gray matter, and
thalamus were set, and the signal intensity of the
DW image for each b value was measured on a con-
sole by using the copy-and-paste function (Fig. 1).
Calculation of the Apparent Diffusion
Coefficient and Data Analysis
The ADC values were calculated using the
two-point and multipoint methods. The calcula-
tion with the two-point method was as follows:
ADC = –ln(SIb1 / SIb2) / (b1 – b2),
where SI is signal intensity and b is b value.
The least square method was used for calculat-
ing ADC by the multipoint method as follows:
N N
N ∑ i = 1 XiYi – ∑ Ni= 1 Xi∑ i = 1 Yi
ADC = N ,
N ∑ i = 1 Xi2 – (∑ Ni= 1 Xi)2
where N is multi b, X is b value, and Y is sig-
nal ­intensity.
In white matter images, difference in ADCs re-
sulting from the change in the maximum b value
were compared using the two-point method. In ad-
dition, the difference in ADCs resulting from the
change in both the maximum and minimum b val-
ue were compared. Also, the ADC values calcu-
lated using the two-point and three-point meth-
ods with a b value of 1700 s/mm2 were performed
with the values calculated for white matter imag-
es. In addition, the ADC values calculated using
the two-point, three-point, and multipoint meth-
ods with a b value of 5000 s/mm2 were compared
with those calculated for white matter images.
Statistical Analysis
Exact Wilcoxon signed rank tests were used to
compare ADC values for the eight subjects. Stat-
Mate software (version V, GraphPad) was used for
all statistical analyses. A p < 0.05 was considered
statistically significant.
Fig. 2—Graph of signal
matter, gray matter, and
thalamus with multiple
b values on brain MR
images. Diffusion decay
was biexponential,
which had fractional b
value of approximately
1700 s/mm2 .
4000 5000
Results
The signal intensities of white matter, gray
matter, and thalamus for multiple b values of
brain images are shown in Figure 2. The dif-
fusion decay was biexponential, which had
a fractional b value of 1700 s/mm2 in white
matter, gray matter, and thalamus. The ADC
values calculated using the two-point method
with a b value of 0 s/mm2 and each b value for
the white matter images are shown in Figure
3. With an increase in the maximum b value,
(1) the ADC value decreased. The ADC values
for white matter changed along with a change
in both maximum and minimum b values ac-
cording to the two-point method, as shown in
Figure 4. When the two points of choice were
smaller than the fractional b value (b = 1700
(2) s/mm2), the ADC value was 0.0007–0.0008
mm2/s, but when the two points of choice
were greater than the fractional b value,
the  ADC value was 0.0003–0.0004 mm2/s.
The ADC values of the white matter calcu-
lated with b values of 0 and 1500 s/mm2 and
with b values of 0, 1000, and 1500 s/mm2 are
shown in Figure 5; no statistically significant
difference was found for either calculation.
In addition, the ADC value showed a sta-
tistically significant increase when the b
values were set according to the multipoint
method with b values of 0 and 5000 s/mm2;
b values of 0, 3000, and 5000 s/mm2; and
b values of 0, 1000, 2000, 3000, 4000, and
5000 s/mm2, as shown in Figure 6 (p < 0.05).
Discussion
The ADC values calculated from DW im-
ages can be used for differentiating between
tumors. Its clinical use now includes im-
proved tissue characterization for monitoring
treatment response after chemotherapy or ra-
diation and for differentiating posttherapeutic
 Importance of Fractional b Value for Calculating ADC

0.0010 0.0008

0.0009 0.0007

ADC Value (mm2/s)

0.0008 0.0006
0.0007
ADC Value (mm2/s)

0.0005
0.0006 0.0004
0.0005
0.0003
0.0004
0.0002
0.0003
0.0001
0.0002
0
0.0001

0
00

00

00

00

00

00

00

00

00

00
50

70
10

20

30

40

50

20

25

30

40

50
0

d
0 and 0 and 0 and 0 and 0 and 0 and 0 and 0 and

d
an

an
an

an

an

an

an

an

an

an

an

an
0

0
300 500 1000 1500 2000 3000 4000 5000

30

30

0
00

00

00

00

00

00

00

00

00
30
10

10

10

10

18

20

20

20

20
b Value (s/mm2) b Value (s/mm2)

Fig. 3—Graph of apparent diffusion coefficient (ADC) values for white matter
calculated using two-point method, with b value of 0 and each b value specified.
Vertical lines and whiskers denote 95% CIs.
American Journal of Roentgenology 2016.207:1239-1243.
Fig. 4—Graph showing how apparent diffusion coefficient (ADC) values for white
matter changed along with changes in both maximum and minimum b values using
two-point method. When two points of choice were smaller than fractional b value
(b = 1700 s/mm2), ADC value was 0.0007–0.0008 mm2 /s, but when two points of

choice were greater than fractional b value, ADC value was 0.0003–0.0004 mm2 /s.
Vertical lines and whiskers denote 95% CIs.

changes from residual active tumors. Howev-
er, there are several factors that negatively af-
fect the precision of ADC calculation. We ex-
amined the effect of using different b values
for ADC calculation and the difference in the
values obtained using the two-point and mul-
tipoint methods. The diffusion decay curve of
the human body is biexponential, not monoex-
ponential, owing to restricted diffusion. The
biexponential diffusion decay is considered to
be identical to a nongaussian distribution.
For nongaussian distributions, it has been
already reported that the ADC value varies
according to the b value used for calculation,
but previous studies did not explain the as-
sociation with the fractional b value [27, 43,
44]. In this study, we found an association
typical reports of ADC values for adult brain
gray matter, (i.e., 0.0007–0.0008 mm2/s  in
the fast component and 0.0003–0.0004
mm2/s in the slow component) [45–47]. In
this study, the fractional b value of normal
white matter, gray matter, and thalamus was
1700 mm2/s, but the values in other organs
(e.g., prostate and breast) may be different.
In addition, as for both the two-point and
the multipoint method, the aforementioned
principle is similar, but the multipoint meth-
od is associated with less error if the b val-
ue is in the same component [32, 48]. When
a physician measures or refers to the ADC,
we suggest that it is necessary to check the
b values used for calculation. What the ADC
actually expresses is different depending on
itation of this study is that it is difficult to
identify fractional b values. Because tissues
typically have fractional b values, it is neces-
sary to publish this information. Therefore, a
method to automatically and easily identify
fractional b values is currently under study.
Conclusion
In calculation of the ADC, the relation-
ship between the b value used for calcula-
tion and the fractional b value is very impor-
tant. To have a clinically significant meaning,
the ADC should be calculated from b values
0.0009
0.00085
ADC Value (mm2/s)

between the fractional b value and b values
used for calculation of the ADC.
The ADC value of the fast component
was obtained using b values smaller than the
fractional b value, and the ADC value of the
slow component was obtained using b values
greater than the fractional b value. Moreover,
when the two b values were higher and lower
than the fractional value, the ADC had clini-
cally nonsensical mixed values with fast and
slow components. In this study, when the two
points of choice were smaller than the frac-
tional b value (b  = 1700 s/mm2), the ADC
value was 0.0007–0.0008 mm2/s, but when
the two points of choice were greater than
the fractional b value, the ADC value was
0.0003–0.0004 mm2/s, as shown in Figures
3 and 4. These values are similar to those in
whether the b values used for calculation are 0.0008
higher or lower than the fractional value.
Therefore, it is important to determine the 0.00075
fractional b value of the target tissue and to
0.0007
use the b values in the fast and slow compo-
nents. Ogura et al. [49] proposed that com- 0.00065
parison of computed DWI changes using b
values higher and lower than the fractional 0.0006
0 and 1000 0,1000,
b value could be used for calculation of the and 1500
computed MRI (a calculation from DWI of b Values Used for
two b values). Additionally, Jambor et al. [50] ADC Calculation (s/mm2)
recommended that b values be divided into
three clustered distributions for DWI of the Fig. 5—Graph of apparent diffusion coefficient
prostate. The multipoint method is associat- (ADC) values of white matter using b values of 0 and
ed with less measurement error, but the im- 1500 s/mm2 and using b values of 0, 1000, and 1500
aging time is long. However, all b values in s/mm2 . No statistically significant difference was
found. Boxes indicate upper and lower quartiles, lines
the multipoint method should be smaller or within boxes indicate medians, and vertical lines and
larger than the fractional b value. The lim- whiskers indicate maximum and minimum values.

AJR:207, December 2016 1241


0.0006
0.00055
ADC Value (mm2/s)
0.0005
0.00045
0.0004
0.00035
0.0003
0 and 5000 0,3000, 0,1000, 2000, 3000,
and 5000
b Values Used for ADC Calculation (s/mm2)
across fractional values. Therefore, it is neces-
sary to know the fractional b value of the tis-
sue being analyzed before ADC calculation.
American Journal of Roentgenology 2016.207:1239-1243.
References
1. Moseley ME, Kucharczyk J, Mintorovitch J, et al.
Diffusion-weighted MR imaging of acute stroke:
correlation with T2-weighted and magnetic sus-
ceptibility-enhanced MR imaging in cats. AJNR
1990; 11:423–429
2. Lutsep HL, Albers GW, DeCrespigny A, Kamat
GN, Marks MP, Moseley ME. Clinical utility of
diffusion-weighted magnetic resonance imaging
in the assessment of ischemic stroke. Ann Neurol
1997; 41:574–580
3. Kul S, Cansu A, Alhan E, Dinc H, Gunes G, Reis
A. Contribution of diffusion-weighted imaging to
dynamic contrast-enhanced MRI in the character-
ization of breast tumors. AJR 2011; 196:210–217
4. Yoshikawa T, Kawamitsu H, Mitchell DG, et al.
ADC measurement of abdominal organs and le-
sions using parallel imaging technique. AJR 2006;
187:1521–1530
5. Taouli B, Vilgrain V, Dumont E, Daire JL, Fan B,
Menu Y. Evaluation of liver diffusion isotropy and
characterization of focal hepatic lesions with two
single-shot echo-planar MR imaging sequences:
prospective study in 66 patients. Radiology 2003;
226:71–78
6. Guo Y, Gao YG, An NY, et al. Differentiation of
clinically benign and malignant breast lesions us-
ing diffusion-weighted imaging. J Magn Reson
Imaging 2002; 16:172–178
7. Theilmann RJ, Borders R, Trouard TP, et al.
Changes in water mobility measured by diffusion
MRI predict response of metastatic breast cancer
to chemotherapy. Neoplasia 2004; 6:831–837
8. Mardor Y, Pfeffer R, Spiegelmann R, et al. Early
detection of response to radiation therapy in patients
with brain malignancies using conventional and
high b-value diffusion-weighted magnetic reso-
1242 AJR:207, December 2016
Ogura et al.
Fig. 6—Graph of
apparent diffusion
coefficient (ADC) values
showing statistically
significant increase
when b values were set
according to multipoint
method at b values of 0
and 5000 s/mm2; b values
of 0, 3000, and 5000
s/mm2; and b values of 0,
1000, 2000, 3000, 4000,
and 5000 s/mm2 (p < 0.05).
Boxes indicate upper
and lower quartiles, lines
within boxes indicate
medians, and vertical
4000, and 5000
lines and whiskers
indicate maximum and
minimum values.
nance imaging. J Clin Oncol 2003; 21:1094–1100
9. Rechichi G, Galimberti S, Signorelli M, et al. En-
dometrial cancer: correlation of apparent diffu-
sion coefficient with tumor grade, depth of myo-
metrial invasion, and presence of lymph node
metastases. AJR 2011; 197:256–262
10. Higano S, Yun X, Kumabe T, et al. Malignant as-
trocystic tumors: clinical importance of apparent
diffusion coefficient in prediction of grade and
prognosis. Radiology 2006; 241:839–846
11. Bulakbasi N, Guvenc I, Onguro O, Erdogan E,
Tayfun C, Ucoz T. The added value of the appar-
ent diffusion coefficient calculation to magnetic
resonance imaging in the differentiation and grad-
ing of malignant brain tumors. J Comput Assist
Tomogr 2004; 28:735–746
12. Yerli H, Agildere AM, Aydin E, et al. Value of ap-
parent diffusion coefficient calculation in the dif-
ferential diagnosis of parotid gland tumors. Acta
Radiol 1987; 48:980–987
13. Matsushima N, Maeda M, Takamura M, Takeda
K. Apparent diffusion coefficients of benign and
malignant salivary gland tumors: comparison to
histopathological findings. J Neuroradiol 2007;
34:183–189
14. Desouza NM, Reinsberg SA, Scurr ED, Brewster
JM, Payne GS. Magnetic resonance imaging in
prostate cancer: the value of apparent diffusion
coefficients for identifying malignant nodules.
Br J Radiol 2007; 80:90–95
15. Eida S, Sumi M, Sakihama N, Takahashi H,
­Nakamura T. Apparent diffusion coefficient map-
ping of salivary gland tumors: prediction of the be-
nignancy and malignancy. AJNR 2007; 28:116–121
16. Nakayama T, Yoshimitsu K, Irie H, et al. Diffu-
sion-weighted echo-planar MR imaging and ADC
mapping in the differential diagnosis of ovarian
cystic masses: usefulness of detecting keratinoid
substances in mature cystic teratomas. J Magn
­Reson Imaging 2005; 22:271–278
17. Sinha S, Lucas-Quesada FA, Sinha U, DeBruhl N,
Bassett LW. In vivo diffusion-weighted MRI of
the breast: potential for lesion characterization.
J Magn Reson Imaging 2002; 15:693–704
18. Pickles MD, Gibbs P, Lowry M, Turnbull LW.
Diffusion changes precede size reduction in neo-
adjuvant treatment of breast cancer. Magn Reson
Imaging 2006; 24:843–847
19. Khalid L, Carone M, Dumrongpisutikul N, et al.
Imaging characteristics of oligodendrogliomas
that predict grade. AJNR 2012; 33:852–857
20. Sandrasegaran K, Patel AA, Ramaswamy R, et al.
Characterization of adrenal masses with diffu-
sion-weighted imaging. AJR 2011; 197:132–138
21. Razek AA, Sadek AG, Kombar OR, Elmahdy TE,
Nada N. Role of apparent diffusion coefficient
value in differentiation between malignant and
benign solitary thyroid nodule. AJNR 2008;
29:563–568
22. Sumi M, Ichikawa Y, Nakamura T. Diagnostic
ability of apparent diffusion coefficients for lym-
phomas and carcinomas in the pharynx. Eur
­Radiol 2007; 17:2631–2637
23. Hambrock T, Somford DM, Huisman HJ, et al.
Relationship between apparent diffusion coeffi-
cients at 3.0-T MR imaging and Gleason grade in
peripheral zone prostate cancer. Radiology 2011;
259:453–461
24. Peng Y, Jiang Y, Yang C, et al. Quantitative analy-
sis of multiparametric prostate MR images: dif-
ferentiation between prostate cancer and normal
tissue and correlation with Gleason score—a
computer-aided diagnosis development study.
­Radiology 2013; 267:787–796
25. Ei Khouli PH, Jacobs MA, Mezban SD, et al. Dif-
fusion-weighted imaging improves the diagnostic
accuracy of conventional 3.0T breast MR imag-
ing. Radiology 2010; 256:64–73
26. Ogura A, Hayakawa K, Miyati T, Maeda F. Imag-
ing parameter effects apparent diffusion coeffi-
cient determination of magnetic resonance imag-
ing. Eur J Radiol 2011; 77:185–188
27. Iima M, Le Bihan D. Clinical intravoxel incoher-
ent motion and diffusion MR imaging: past, pres-
ent, and future. Radiology 2016; 278:13–32
28. Kolff-Gart AS, Pouwels PJ, Noij DP, et al. Diffu-
sion-weighted imaging of the head and neck in
healthy subjects: reproducibility of ADC values in
different MRI systems and repeat sessions. AJNR
2015; 36:384–390
29. Malyarenko DI, Newitt D, Wilmes LJ, et al. Dem-
onstration of nonlinearity bias in the measurement
of the apparent diffusion coefficient in multicenter
trials. Magn Reson Med 2016; 75:1312–1323
30. Chung AW, Thomas DL, Ordidge RJ, Clark CA.
Diffusion tensor parameters and principal eigenvec-
tor coherence: relation to b-value intervals and field
strength. Magn Reson Imaging 2013; 31:742–747
31. Malkyarenko DI, Chenevert TL. Practical esti-
 Importance of Fractional b Value for Calculating ADC
mate of gradient nonlinearity for implementation
of apparent diffusion coefficient bias correction.
J Magn Reson Imaging 2014; 40:1487–1495
32. Park SY, Kim CK, Park BK, Kwon GY. Compari-
son of apparent diffusion coefficient calculation
between two point and multipoint b value analysis
in prostate cancer and benign prostate tissue at
3 T: preliminary experience. AJR 2014; 203:[web]
W287–W294
33. Corona-Villalobos CP, Pan L, Halappa VG, et al.
Agreement and reproducibility of apparent diffu-
sion coefficient measurements of dual-b value and
multi-b-value diffusion-weighted magnetic reso-
nance imaging at 1.5 Tesla in phantom and in soft
tissues of the abdomen. J Comput Assist Tomogr
2013; 37:46–51
34. Schwarcz A, Bogner P, Meric P, et al. The exis-
tence of biexponential signal decay in magnetic
resonance diffusion-weighted imaging appears to
be independent of compartmentalization. Magn
Reson Med 2004; 51:278–285
American Journal of Roentgenology 2016.207:1239-1243.
35. Niendorf T, Dijkhuizen RM, Norris DG, van
Lookeren Campagne M, Nicolay K. Biexponential
diffusion attenuation in various states of brain tis-
sues: implications for diffusion-weighted imag-
ing. Magn Reson Med 1996; 36:847–857
36. Mulkern RV, Vajapeyam S, Robertson RL, Caruso
PA, Rivkin MJ, Maier SE. Biexponential apparent
diffusion coefficient parametrization in adult vs
newborn brain. Magn Reson Imaging 2001;
19:659–668
AJR:207, December 2016 1243
37. Chin CL, Wehrli FW, Hwang SN, Takahashi M,
Hackney DB. Biexponential diffusion attenuation
in the rat spinal cord: computer simulations based
on anatomic images of axonal architecture. Magn
Reson Med 2002; 47:455–460
38. Mulkern RV, Gudbjartsson H, Westin CF, et al.
Multi-component apparent diffusion coefficients
in human brain. NMR Biomed 1999; 12:51–62
39. Hall MG, Bongers A, Sved P, Watson G, Bourne
RM. Assessment of non-gaussian diffusion with
singly and doubly stretched biexponential models of
diffusion-weighted MRI (DWI) signal attenuation
in prostate tissue. NMR Biomed 2015; 28:486–495
40. Yan R, Haopeng P, Xiaoyuan F, et al. Non-gauss-
ian diffusion MR imaging of glioma: compari-
sons of multiple diffusion parameters and correla-
tion with histologic grade and MIB-1 (Ki-67
labeling) index. Neuroradiology 2016; 58:121–132
41. Bai Y, Lin Y, Tian J, et al. Grading of gliomas by
using monoexponential, biexponential, and
stretched exponential diffusion-weighted MR im-
aging and diffusion kurtosis MR imaging.
­Radiology 2016; 278:496–504
42. Ogura A, Tamura T, Ozaki M, et al. Apparent dif-
fusion coefficient is not dependent on magnetic
resonance system and field strength under fixed
imaging parameters in brain. J Comput Assist
­Tomogr 2015; 39:760–765
43. Dale BM, Braithwaite AC, Boll DT, Merkle EM.
Field strength and diffusion encoding technique
affect the apparent diffusion coefficient measure-
ments in diffusion-weighted imaging of the abdo-
men. Invest Radiol 2010; 45:104–108
44. Schmidt H, Gatidis S, Schwenzer NF, Martirosian
P. Impact of measurement parameters on apparent
diffusion coefficient quantification in diffusion-
weighted-magnetic resonance imaging. Invest
­Radiol 2015; 50:46–56
45. Dhital B, Labadie C, Stallmach F, Möller HE,
Turner R. Temperature dependence of water dif-
fusion pools in brain white matter. Neuroimage
2016; 127:135–143
46. Helenius J, Soinne L, Perkiö J, et al. Diffusion-
weighted MR imaging in normal human brains in
various age groups. AJNR 2002; 23:194–199
47. Tanner SF, Ramenghi LA, Ridgway JP, et al.
Quantitative comparison of intrabrain diffusion in
adults and preterm and term neonates and infant.
AJR 2000; 174:1643–1649
48. Kim SY, Lee SS, Park B, et al. Reproducibility of
measurement of apparent diffusion coefficients of
malignant hepatic tumors: effect of DWI tech-
niques and calculation methods. J Magn Reson
Imaging 2012; 36:1131–1138
49. Ogura A, Koyama D, Hayashi N, Hatano I,
­Osakabe K, Yamaguchi N. Optimal b values for
generation of computed high-b-value DW images.
AJR 2016; 206:713–718
50. Jambor I, Merisaari H, Aronen HJ, et al. Optimi-
zation of b-value distribution for biexponential
diffusion-weighted MR imaging of normal pros-
tate. J Magn Reson Imaging 2014; 39:1213–1222