J. Biomedical Science and Engineering, 2010, 3, 763-775 JBiSE
doi:10.4236/jbise.2010.38102 Published Online August 2010 (http://www.SciRP.org/journal/jbise/).
Published Online August 2010 in SciRes. http:// www. scirp. org/journal/jbise
MRI monitoring of lesions created at temperature below the
boiling point and of lesions created above the boiling point
using high intensity focused ultrasound*
Christakis Damianou1,2, Kl eanthis Ioannides3, Ve ne diktos Hadjisavvas 4, Nikos Mylonas4,
Andreas Couppis4, Demetris Iosif1, Panayiotis A. Kyriacou4
1Frederick University Cyprus, Limassol, Cyprus;
2MEDSONIC LTD, Limassol, Cyprus;
3Polikliniki Ygia, Limassol, Cyprus;
4City University, London, UK.
Email: cdamianou@cytanet.com.cy
Received 11 April 2010; revised 26 May 2010; accepted 31 May 2010.
Magnetic Resonance Imaging (MRI) was utilized to
monitor lesions created at temperature below the
boiling point and lesions created at temperature
above the boiling point using High Intensity Focused
Ultrasound (HIFU) in freshly excised kidney, liver
and brain and in vivo rabbit kidney and brain.
T2-weighted fast spin echo (FSE) was proven as an
excellent MRI sequence that can detect lesions with
temperature above the boiling point in kidney. This
advantage is attributed to the significant difference in
signal intensity between the cavity and the thermal
lesion. In liver the MRI sequence of Proton Density
is recommended to detect lesions above boiling. In
brain T1-W FSE was the optimum pulse sequence to
detect lesions of either type. In order to monitor the
temperature elevation during a HIFU exposure,
T1-weighted fast spoiled gradient (FSPGR) was used.
The shape of the focal temperature distribution was
uniform with the absence of boiling, whereas with an
exposure affected by boiling, the temperature distri-
bution could be of irregular shape, demonstrating the
drastic effects taking place during boiling. In order to
confirm that boiling occurred, the temperature was
estimated using the widely used method of Proton
Resonance Frequency (PRF) shift.
Keywords: Ultrasound; Kidney; Brain; Liver; MRI; Lesion
This paper is a continuation of previous papers by Damia-
nou [1] and Damianou et al. [2]. The first paper de-
scribes methods for ablating kidney tissue using high
intensity focused ultrasound (HIFU), whereas the second
paper deals with magnetic resonance imaging (MRI)
guidance of HIFU for the case of kidney. This paper
goes one step further by evaluating lesions created at
temperatures above the boiling point during HIFU ex-
posures using MRI. The evaluation was performed in
kidney, liver and brain.
We have chosen to explore kidney, liver and brain be-
cause there is currently a lot of ongoing research either
in animal models or in humans for these 3 tissues. In the
area of kidney ablation with HIFU Watkin et al. [3] used
a large animal model and proved the feasibility of this
treatment method. Recently Roberts et al. [4] have per-
formed ablations in the normal rabbit kidneys and they
suggested that the mechanical effects of ultrasound, can
be used to homogenize tissue.
HIFU ablation of renal tumours in humans remains in
the early stages of clinical trials. In the early 1990s, Val-
lancien et al. [5] reported the first clinical feasibility
study in kidney using extracorporeal HIFU. Susani et al
[6], Wu et al. [7], and Marberger et al. [8] conducted
clinical trials in patients with renal tumours and proved
that HIFU may have a place in the treatment of renal
Hacker et al. [9] performed also ablation of 43 kid-
neys (porcine and human), using an experimental hand-
held extracorporeal technology. Finally Klingler et al.
[10] use laparoscopic methods to treat kidney tumors.
Small animal models [11,12] were used to establish
the feasibility of HIFU to create lesions in liver tissue.
The thresholds for liver tissue destruction at varying ex-
posure parameters were established in the 70s and 80s
*This work was supported by the Research Promotion Foundation
(RPF) of Cyprus under the contract ERYAN/2004/1, ΑΝΑΒ
ΘΜΙΣΗ/ ΠΑΓΙΟ/0308/05, and ΕΠΙΧΕΙΡΗΣΕΙΣ/ ΕΦΑΡΜ /0308/01.
764 C. Damianou et al. / J. Biomedical Science and Engineering 3 (2010) 763-775
Copyright © 2010 SciRes. JBiSE
Basic research in the area of liver continued during
the 1990s; for example the histological effects of HIFU
[14], the effects of blood perfusion on during liver abla-
tion [15], and the relationship between tissue depth and
the required intensity levels [16].
Several tumour models have been used to predict the
effects of HIFU on liver tumors in humans (for example
[17,18]) and to destroy VX2 liver tumours in rabbits [19,
20]. Hooded Sarcoma N (HSN) fibrosarcoma has been
also used as a tumour model in rats with some success
[21]. HIFU was also used for the treatment of metastatic
melanoma in liver in a cat [22].
There is an increasing interest of work describing
HIFU in the treatment of liver cancer in human clinical
trials. In the early 1990s, Vallancien et al. [23] re-
ported treatment of liver metastases using HIFU.
The Chongqing group published a study [24] describ-
ing a clinical trial of 68 patients with liver malignancies
using HIFU. Li et al. [25] reported clinical trial of pa-
tients with liver cancer who were treated also with HIFU.
Wu et al. [26] use HIFU in combination with TACE for
the treatment of HCC.
MRI-guided HIFU has generally been used for the
treatment of uterine fibroids using the ExAblate 2000
system (InSightec, Haifa, Israel). However, it is very
likely that soon this system will be utilized for the treat-
ment of kidney, liver and brain tumours. At Imperial
College, London they have recently started a non-ran-
domised clinical trial to assess the safety and efficacy of
the MRI-guided HIFU system of InSightec in the treat-
ment of liver tumours [27].
Thermal ablation of brain in animals with high inten-
sity focused ultrasound (HIFU) was very popular in the
50’s and 60’s (for example [28] and [29]). HIFU was
used in the clinical setting by Fry and Johnson [30] and
showed that HIFU had the potential to treat brain cancer.
Several groups used hyperthermia (heating of several
minutes at 43ºC) to treat brain tumours [31,32]. The
clinical trials were abandoned probably due to the in-
existence of effective imaging modality to guide the
therapy. Especially for the case of brain it is extremely
important to have absolute control of the ablation in or-
der to avoid vital brain tissue damage such as the neu-
rons. Now with the advancement of HIFU technology
guided by MRI, it will be possible to conduct clinical
studies for brain cancer.
In the early nineties several studies by the group of Dr.
Hynynen [33-38] demonstrated the creation of lesions in
animal brain and use MRI successfully for guiding and
monitoring. Therefore, now there is increasing interest
regarding brain ablation.
The combination of HIFU and MRI was first cited by
Jolesz and Jakab [39] who demonstrated that an ultra-
sonic transducer could be used inside a MRI scanner.
The concept of using MRI to monitor the necrosis pro-
duced by HIFU was fully demonstrated in the early
nineties in canine muscle [40-42]. In these studies it was
shown that the contrast between necrotic tissue and
normal tissue was excellent. This was a great enhance-
ment for the HIFU systems because the therapeutic pro-
tocols can be accurately guided. Therefore the interest of
using MRI as the diagnostic modality of guiding HIFU
was increased. Although there are numerous studies re-
garding lesions created with temperature below the boil-
ing point (also known as thermal lesions), there is insuf-
ficient information regarding MRI detection of lesions
created with temperature above the boiling point. This
paper focuses in the detection of lesions above boiling
using MRI.
Several MRI sequences are investigated. For high
quality imaging, this can be used at the end of a thera-
peutic protocol or at some instances of the protocol, the
fast spin echo (FSE) techniques T1, and T2-weighted are
investigated in all 3 tissues under investigation. For fast
imaging, the T1-weighted fast spoiled gradient (FSPGR)
MRI sequence was used. Fast imaging can be used to
monitor the dynamic increase of tissue temperature dur-
ing the application of an ultrasonic exposure. We have
also use the fluid-attenuated inversion recovery (FLAIR)
sequence for brain since this is the predominant se-
quence used in the clinical setting.
In order to prove whether a lesion is created at tem-
perature below boiling or above, it is necessary to esti-
mate the temperature at the focus. The proton reso-
nance frequency (PRF) shift has been proven to be best
pulse sequence for estimating temperature, because with
this sequence the temperature is less dependent on the
physiological changes of tissue during high-temperature
HIFU exposures [43]. The temperature dependence of
the PRF shift was measured to be linear above 50°C [43].
This linearity of the PRF shift above the tissue necrosis
threshold allows the tissue temperature to be estimated
during the therapeutic ultrasound exposures.
The task with high quality imaging was to find an op-
timum technique that can discriminate thermal from
boiling lesions. Discriminating between lesion and nor-
mal tissue involves two types of tissues. Discriminating
between thermal and boiling lesions involves three types
of layers (normal tissue, lesion and cavity). Therefore
the signal intensity vs. MR parameters needs to be
evaluated for the above layers in order to optimise the
contrast among tissue of interest, lesion and cavity. The
other main task was to monitor the temperature elevation
using a fast MRI technique in order to observe the shape
C. Damianou et al. / J. Biomedical Science and Engineering 3 (2010) 763-775 765
Copyright © 2010 SciRes. JBiSE
of the beam during HIFU exposures.
The growth of vapor bubbles due to boiling occurs
due to the temperature induced by HIFU. Boiling is dif-
ferent form cavitation which occurs due to pressure os-
cillations induced by HIFU. Vapor bubbles created by
boiling can grow rapidly to a size of few millimetres
[44]. This growth can be explosive due to the super
heating caused by HIFU. Therefore the cavities pro-
duced of few mms can be easily monitored by MRI.
There are two reasons for studying the MRI appear-
ance of boiling lesions that could be of importance:
1) Since boiling provides enhanced heating, it could
be possible for large tumours (for example giant fi-
broadenomas) to use this type of heating (especially in
the center of the tumor) in order to accelerate the abla-
2) since the rabbit model is used extensively by many
researchers in MRI animal experiments (for example
[33-37]), then it is very useful for them to know the ap-
pearance of boiling lesions.
2.1. HIFU/MRI System
Figure 1 shows the block diagram of the HIFU/MRI
system which includes the following subsystems:
1) HIFU system, 2) MR imaging, 3) Positioning de-
vice (robot) and associate drivers, 4) Temperature meas-
urement, 5) Cavitation detection, 6) MRI compatible
camera, 7) Software.
2.1.1. HIFU System
The HIFU system consists of a signal generator (HP
33120A, Agilent technologies, Englewood, CO, USA), a
RF amplifier (250 W, AR, Souderton, PA, USA), and a
spherically shaped bowl transducer made from piezo-
electric ceramic of low magnetic susceptibility (Etalon,
Lebanon, IN, USA). The transducer used for the kidney
and liver ablation operates with frequency of 4 MHz,
and the transducer for brain ablation operates at 1 MHz.
The transducer is rigidly mounted on the MRI-compati-
ble positioning system (MEDSONIC LTD, Limassol,
Cyprus) which is described shortly.
2.1.2. MRI Imaging
The 3-d positioning device and the transducer were
placed inside a MRI scanner (Signa 1.5 T, by General
Electric, Fairfield, CT, USA). The spinal coil (USA in-
struments, Cleveland, OH, USA) was used to acquire the
MRI signal for the case of kidney and liver, whereas a
brain coil (USA instruments, Cleveland, OH, USA) was
used to acquire the MRI signal from the brain tissue.
2.1.3. Po sitioni n g De vice/Drivers
The robot has been developed initially for three de-
grees-of-freedom, but it can be easily developed for 5
degrees of motion. Since the positioning device is placed
on the table of the MRI scanner its height should be
around 55 cm (bore diameter of the MRI scanner). The
length of the positioning device is 45 cm and its width
30 cm. The weight of the positioning device is only 6 kg
and therefore it can be considered portable. Figure 2
shows the schematic the positioning device illustrating
the 3 stages, transducer, and coupling method. The posi-
tioning device operates by means of 3 piezoelectric mo-
tors (USR60-S3N, Shinsei Kogyo Corp., Tokyo, Japan).
More details of this positioning device can be found in
[31]. The movement of the positioning device is moni-
tored by an MRI compatible camera (not shown in the
figure) which is placed on a non-magnetic holder. More-
over, the positioning system includes optoelectronic en-
coders for providing signals indicating the relative posi-
tions of the movable elements in the positioning system.
The resolution of all 3 axes of the positioning device is
0.1 mm.
2.1.4. MRI Compa tibl e Cam era
In order to monitor the condition of the animal or humans
Figure 1. HIFU system under MRI guidance showing the vari-
ous functionalities of the HIFU/MRI system.
Figure 2. Schematic of the robot showing all of its stages.
766 C. Damianou et al. / J. Biomedical Science and Engineering 3 (2010) 763-775
Copyright © 2010 SciRes. JBiSE
(future use), a MRI compatible camera (MRC Systems
GmbH, Heidelberg, Germany) was mounted on the sys-
tem. The camera was interfaced by means of a video
card. With the aid of the MRI compatible camera, the
researcher can monitor the welfare of the animal.
2.1.5. Temper ature Measu rement
Temperature was measured in few experiments in order
to confirm that the temperature estimated using the PRF
method was accurate enough. Temperature is measured
using a data acquisition system (HP 34970A, Agilent
technologies, Englewood, CO, USA). Temperature is
sensed using a 50-μm diameter T-type copper-costantan
thermocouple (Physitemp Instruments, Inc. New Jersey,
USA) which is MRI compatible. The thermocouple is
placed in the tissue by means of a catheter. The thermo-
couple measures the temperature at the focus. This is
achieved by applying low-intensity (low enough not to
cause tissue damage) and during the application of ul-
trasound the transducer is scanned accordingly in order
to detect the maximum temperature. This establishes
positioning of the thermocouple in the focus of the
transducer. The temperature error of the thermocouple is
in the order of 0.1ºC.
2.1.6. C a vitation Detector
From a scientific point of view it will be useful to sepa-
rate lesions developed based on thermal and based on
cavitation mechanisms. In this system we use a passive
MRI compatible cavitation detector (Etalon, Lebanon,
IN, USA), which is placed perpendicularly to the beam
of the HIFU transducer (method described in [45]).
Since the HIFU protocol is applied inside the magnet of
an MRI scanner, the detector must be MRI compatible.
The diameter of the detector was 1 cm, its radius of cur-
vature was 10 cm and operated between 1 and 13 MHz
(centre frequency is 7 MHz). The detector was me-
chanically coupled to the HIFU transducer. The voltage
from the detector was fed to an analogue to digital (A/D)
card (CS1250, A/D 12 bit, 50 MHz, from GAGE, La-
chine, Canada). The A/D card was synchronised to re-
ceive the signal when the HIFU transducer was activated.
The received signal was stored in a PC (Dell Inc. Round
Rock, Texas, USA). The signal was then displayed using
EXCEL (Microsoft Corporation, Redmond, WA USA) in
order to visualize whether cavitation occurs or not.
2.1.7. So ft ware
A user-friendly program written in MATLAB (The
Mathworks Inc., Natick, MA) has been developed in
order to control the system. The software serves the fol-
lowing main tasks: 1) Displaying of MR images, 2)
transducer movement (the user may move the robotic
arm in a specific direction or customize the automatic
movement of the robotic arm in any rectangular forma-
tion by specifying the pattern, the step and the number of
steps), 3) messaging (starting time, treatment time left
etc), 4) Patient data (age, weight, etc), 5) Display of mo-
tor position, and 6) Display of the contents of an MR
compatible camera, 7) Cavitation detection window, 8)
temperature measurement, and 9) MR estimated tem-
perature using the PRF method.
2.2. In Vitro Expe riments
The tissue was placed on top of an absorbing material in
order to shield adjacent tissue form stray radiation from
the bottom. The transducer was placed on the arm of the
positioning device and was immersed in the water tank,
thus providing good acoustical coupling between tissue
and transducer. Any bubbles that may have collected
under the face of the transducer face were removed in
order to eliminate any reflections. In all experiments the
tissues used (kidney, liver and brain) were extracted
from freshly killed lamb, and the experiment was con-
ducted in the same day. Totally 22 kidneys, 8 liver and
16 brains were ablated for investigating various issues.
2.3. In Vivo Experim ents
For the in vivo experiments, adult rabbits from Cyprus
were used weighting approximately 3.5-4 kg. Totally 8
rabbits were used in the experiments. The rabbits were
anaesthetized using a mixture of 500 mg of ketamine
(100 mg/mL, Aveco, Ford Dodge, IA), 160 mg of xy-
lazine (20 mg/mL, Loyd Laboratories, Shenandoah, IA),
and 20 mg of acepromazine (10 mg/mL, Aveco, Ford
Dodge, IA) at a dose of 1 mL/kg.
Presence of the skull in the ultrasonic path not only
distorts the field by reflection, but may also destroys the
underlying tissue in contact with it by absorbing ultra-
sonic energy and dissipating it as heat. A craniotomy
adequate in extent to permit unimpeded passage of the
cone of sound was imperative. The extent of the crani-
otomy depends on the solid angle of radiation and the
depth of the target from the cranial surface. The larger
the angle and the deeper the target, the larger the size of
craniotomy needed. For the transducer used and a target
depth of 1 cm, a circular craniotomy of 3 cm in diameter
was adequate. The animal experiments protocol was
approved by the national body in Cyprus responsible for
animal studies (Ministry of Agriculture, Animal Ser-
2.4. HIFU Parameters
The in situ spatial average intensity was estimated based
on the applied power and the half-power width of the
beam of the transducer. The details of the intensity esti-
mation can be found in Damianou 2003 [1]. In order to
create large lesions, a grid pattern of 3 × 3 or 4 × 4 over-
C. Damianou et al. / J. Biomedical Science and Engineering 3 (2010) 763-775 767
Copyright © 2010 SciRes. JBiSE
lapping lesions was used. The spacing between succes-
sive transducer movements was 2 mm, which creates
overlapping lesions for the intensity and pulse duration
used. In all the exposure, unless stated otherwise the
ultrasound was turn on for 5 s. The delay between suc-
cessive ultrasound firings was 10 s for the scanned abla-
tion. The intensity indicated through this paper is in situ
spatial average intensity.
2.5. MRI Parameters
The various MRI parameters used for the various pulse
sequences are listed in Tabl e 1. The Region of Interest
(ROI) was circular with diameter of nearly 2 mm.
The temperature change (T) has been estimated using
the equation stated in Chung et al. 1996 [46] which is as
 (1)
is the temperature-dependent phase shift
which is the phase acquired before and during tempera-
ture elevation and which accumulates during the echo
time TE using the gradient-echo pulse sequence FSPGR.
The other terms are
which is the gyromagnetic ratio of
proton, 42.58 MHz/T,
is the average proton resonance
frequency coefficient and B0 is the flux density of the
static magnetic field. The measured temperature eleva-
tion can be added to the base-line temperature to obtain
the absolute temperature. The average temperature coef-
ficient for the frequency shift was taken from the study
of Vykhodtseva et al. [47].
Figure 3 shows an MRI image of 4 lesions in vitro kid-
ney (plane perpendicular to the beam) resulting from
intensities ranging from 1000 to 2500 W/cm2 using
T1-weighted FSE (Figure 3(a)), T2-weighted FSE (Fig-
ure 3(b)) and T1-weigthed FSPGR (Figure 3(c)) with
pulse duration of 5 s. The MRI parameters used are
shown in Ta b l e 1 (row 1, 2 and 4). The MRI estimated
maximum temperature at the focus was 55ºC for the
intensity of 1000 W/cm2, 83ºC for 1500 W/cm2, 105ºC
for 2000 W/cm2 and 123ºC for 2500 W/cm2. The
temperature measured using the thermocouple for the
1000 W/cm2 was 53ºC, which is very close to the tem-
perature estimated using the PRF method.
Note that with T2-weighted FSE (Figure 3(b)) white
spots (cavity) within the dark thermal lesion are seen for
intensities higher than 2000 W/cm2. T1-W FSE and
FSPGR show some indication of these cavities, but the
resolution is weaker than T2-w FSE.
Figure 4 shows MRI images (in a plane parallel to the
transducer beam) of 3 lesions in kidney in vitro at inten-
sities of 1000, 2000 and 3000 W/cm2. Again T2-
weighted FSE shows cavities (due to boiling) within the
thermal lesion. T1-W FSE and T1-W FSPGR fail to pro-
vide good resolution in this axis.
Having observed that T2-W FSE was probably a suc-
cessful MRI sequence to detect boiling lesions, this
pulse sequence was investigated further by evaluating
the Contrast to Noise Ratio (CNR) vs. Echo Time (TE).
Figure 5 shows the plot of CNR vs. TE for kidney tissue,
lesion and cavity of the lesion of Figure 4(b) (intensity
of 3000 W/cm2) demonstrating that good contrast be-
tween lesion and cavity is achieved using T2-weighted
FSE between 20 and 50 ms.
Figure 6 shows a T2-weighted FSE image of 3 lesions
in rabbit kidney in vivo using different intensities (1000,
2000 and 2500 W/cm2) for a 5 s pulse. The estimated
temperature was 52ºC, 95ºC and 115ºC.
The lesion created using 2500 W/cm2 appears to have
a white spot (cavity). These temperatures are lower than
the corresponding temperatures in vitro (Figure 4) for
the same intensity due to the removal of heat due to
blood flow or possibly due to reflection from various
Figure 7(a) shows an MR image of a lesion acquired
using T1-weighted FSPGR (for MRI parameters see
Table 1, row 4). Figure 7(b) shows the photograph of the
kidney showing cavity within the large thermal lesion.
Note that this large lesion was created using a matrix
of 4 × 4 single lesions with intensity of 1500 W/cm2. Out
of these 16 lesions, one lesion was created possibly due
to the mechanism of cavitation, which results to tissue
evaporation or boiling. The estimated temperature during
Table 1. Parameters used for the various MRI pulse sequences.
Series NAME TR
(ms) TE (ms) Slice thickness
(mm) Matrix FOV
(cm) NEX BW
(KHz) ETL Other
1 T1-weighted
FSE 500 9.2 3, (gap 0.3 mm) 256 × 25616 1 31.25 8 -
2 T2-weighted
FSE 2500 8,16,32,48,64,80 3, (gap 0.3 mm) 256 × 25616 1 31.25 8 -
3 PD 2500 7.2 3, (gap 0.3 mm) 256 × 25616 1 31.25 8 -
T1-weighted 50 2.7 3, (gap 0.3 mm) 256 × 25616 1 62.50 - Flip angle 50
5 FLAIR 8000 80 3, (gap 0.3 mm) 256 × 25616 1 6.9 8 Inversion Time
768 C. Damianou et al. / J. Biomedical Science and Engineering 3 (2010) 763-775
Copyright © 2010 SciRes. JBiSE
(a) (b)
Figure 3. MR images (in a plane perpendicular to the beam) of
four lesions (intensities 1000, 1500, 2000 and 2500 W/cm2) in
kidney in vitro using. (a) T1-weighted FSE; (b) T2-weighted
FSE; (c) T1-weighted FSPGR. With intensities above 2000 W/cm2
the lesions exhibit boiling activity. The discrimination between
boiling and non-boiling lesion is best monitored using T2-W
(a) (b)
Figure 4. MR images (in a plane parallel to the beam) of 3
lesions in kidney in vitro using. (a) T1-weighted FSE; (b)
T2-weighted FSE; (c) T1-weighted FSPGR. With intensities
above 2000 W/cm2 the lesions exhibit boiling activity. The dis-
crimination between boiling and thermal lesion is best moni-
tored using T2-W FSE.
Figure 5. CNR vs. TE for the lesion, kidney and cavity for the
lesion of Figure 8(b) with intensity of 3000 W/cm2.
Figure 6. MRI image using T2-weighted FSE of 3 lesions in
rabbit kidney in vivo at different intensities (1000, 2000, and
2500 W/cm2) for a 5 s pulse. The lesion with intensity of
2500 W/cm2 is affected by tissue boiling.
the creation of this lesion was 120ºC, whereas the tem-
perature for the rest of the lesions (non-boiling) varied
from 80 to 85ºC.
This bubbly lesion exhibits low-signal lesion (dark
spot) and lies inside the large lesion (white spot within
the kidney tissue). At the location of the boiling lesion,
the passive cavitation detector confirmed the occurrence
of cavitation since broadband emission was detected (see
Figure 7(c) which shows the frequency spectrum of the
HIFU transducer). Although with this intensity bubbly
lesions should not be produced, the high temperature
estimated for this one lesion, should be attributed to
cavitation. Cavitation was possibly initiated by bubbles
that are trapped in the in vitro tissue due to the absence
of blood in the vasculature.
Figure 8 demonstrates the temperature increase in vi-
tro kidney using HIFU and MRI monitoring. The MRI
images were acquired using the dynamic sequence T1-
weighted FSPGR. Each image was acquired in 5 s. In
Figure 8(a) ultrasound was OFF. In the next 5 images
020406080100120 140160
TE (ms)
C. Damianou et al. / J. Biomedical Science and Engineering 3 (2010) 763-775 769
Copyright © 2010 SciRes. JBiSE
(a) (b)
Frequency (MHz)
Spectral Power (mW
Figure 7. (a) MR image (in a plane perpendicular to the beam)
of large lesion in kidney in vitro using T1-weighted FSPGR
(TR = 50 ms), showing one cavitation lesion; (b) Photograph
of the kidney showing the cavitation lesion within the large
thermal lesion; (c) Frequency spectrum of the HIFU transducer
exhibiting cavitation activity.
the applied spatial average intensity was 1000 W/cm2
(for 25 s), and in the last 2 images ultrasound is turned
OFF. Due to the heating a dark spot is observed (see
arrows). The estimated maximum temperature was 55ºC.
Figure 9 shows the corresponding temperature in-
crease using T1-weighted FSPGR influenced by boiling.
In Figure 9(a) ultrasound was OFF. In the next 5 images
the applied spatial average intensity was 3500 W/cm2
(for 25 s). Note that compared to Figure 7 where the
focal beam is circular, the focal beam in this figure is
distorted, which is attributed to the occurrence of boiling
confirmed also by the temperature of 112ºC measured.
Figure 10 shows MRI image of lesion in liver. The
image was acquired using Proton density. The lesion was
created using low intensity (1000 W/cm2) for long time
(30 s). These ultrasonic parameters produce temperature
elevation which is above the boiling point (120ºC). Thus,
the evaporation of tissue causes a cavity that follows the
shape of the beam.
Figure 11 shows the MRI image a large lesion in
lamb brain in vitro created by scanning the transducer
with a 4 × 4 grid using 2000 W/cm2 using T1-w FSE
resulting to a large bubbly lesion. The maximum esti-
mated temperature for this lesion was 110ºC. Note that
in some location no lesion was created due to poor ul-
trasound penetration due to the air bubbles possibly
trapped in the blood vessels. This image shows once
more the excellent contrast between normal brain and
lesions (in this case boiling lesions).
Figure 12 shows ablation in rabbit in vivo using a 4 ×
4 grid with intensity of 2000 W/cm2 for 20 s. This large
lesion was created using thermal mechanisms and there-
fore the lesion appears bright. The maximum estimated
temperature for this exposure was 90ºC. The contrast of
thermal lesions is definitely much better than the case of
boiling lesions. Unlike the in vitro case of Figure 11
where with this level of intensity boiling lesions were
created, in this in vivo example in none of the 16 lesions
a boiling lesion was produced. This proves that in vitro
brain includes bubbles which are responsible for pro-
ducing boiling lesions possibly due to enhanced heating
or cavitation. Both of these mechanisms possibly pro-
duced temperatures above boiling and therefore bubbly
lesions were created.
The methodology applied for kidney which evaluates
T1-W FSE, T2-W FSE and FSPGR (Figures 3 and 5)
was also applied for liver. In brain in addition to these 3
pulse sequences, FLAIR was also evaluated because this
MRI sequence is widely used clinically for the case of
brain. It was found that for liver the best pulse sequence
to evaluate boiling lesions was T2-w FSE with minimum
TE (i.e. proton density). For brain T1-W FSE was the best
pulse sequence to monitor both boiling and non-boiling
lesions. Table 2 summarizes the recommended MRI
sequence according to our experience for monitoring
boiling lesions.
So far it was concluded by several studies (for example
Quesson et al. [48], Salomir et al. [49], McDannold et al.
[50]) that MRI-guidance of HIFU serves mainly 4 pur-
poses: 1) localisation of the focus, 2) imaging of the
temperature elevation, 3) imaging at the end of the
treatment protocol, in order to evaluate the necrosis and
Table 2. Recommended pulse sequences for discriminating
between (a) normal and thermal lesions; and (b) normal tissue,
lesions created with temperature below boiling and lesions cre-
ated with temperature above boiling.
Tissue type Contrast between
lesion below boiling
and tissue
Contrast between lesion
below boiling, tissue,
and lesion above boiling
Kidney T1-W FSE, T2-W
Liver T1-W FSE, T2-W
Brain T1-W FSE T1-W FSE
770 C. Damianou et al. / J. Biomedical Science and Engineering 3 (2010) 763-775
Copyright © 2010 SciRes. JBiSE
Figure 8. Temperature evolution in vitro kidney using T1- weighted FSPGR (thermal
mechanism). Each image was acquired in 5 s. (a) Ultrasound is OFF; (b)-(f) Applied
spatial average intensity: 1000 W/cm2 (for 25 s); (g)-(h) Ultrasound is OFF.
C. Damianou et al. / J. Biomedical Science and Engineering 3 (2010) 763-775 771
Copyright © 2010 SciRes. JBiSE
Figure 9. Temperature elevation in vitro kidney using T1- weighted FSPGR (boiling). (a) Ultrasound is OFF; (b)-(f) applied spatial
average intensity: 3500 W/cm2 (for 25 s).
Figure 10. MR image using proton density of lesion in liver
created under the influence of boiling. The lesion was created
using low intensity (1000 W/cm2) for long time (30 s).
Figure 11. MRI image using T1-w FSE of large boiling lesion
created in vitro using 2000 W/cm2 for 20 s.
772 C. Damianou et al. / J. Biomedical Science and Engineering 3 (2010) 763-775
Copyright © 2010 SciRes. JBiSE
Figure 12. MRI image using T1-w FSE of large thermal lesion
created in vivo using 2000 W/cm2 for 20 s.
possibly re-planning the treatment protocol in the event
of incomplete coverage of the target and 4) follow-up
imaging, evaluating the effectiveness of HIFU ablation,
several days after the treatment. The main goal in this
paper was to develop methods for detecting lesions cre-
ated at temperature below boiling point and lesions cre-
ated above boiling using MRI techniques.
This paper enhances the role of MRI guidance in
HIFU because it provides techniques to discriminate
between non-boiling and boiling lesions.
In kidney the best MRI sequence to detect boiling le-
sions was T2-W FSE. T2-W FSE was evaluated further by
plotting the CNR vs. TE for the three regions of interest
(kidney, lesion and cavity). It was concluded that be-
tween TE’s of 20 and 50 ms, the signal difference, and
hence the contrast between the three regions of interest
(kidney, lesion and cavity) is maximum (Figure 5). Ac-
cording to Figure 5, the cavity, which appears inside the
lesion, has stronger signal and decays slower.
T1-weighted FSE and T1-weighted FSPGR do not
consistently show boiling lesions in kidney, and even
when cavities are visible the contrast is not very good.
In liver T2-W FSE with low TE (i.e. Proton density) is
recommended as an MRI sequence to detect non-boiling
lesions and boiling lesions.
Previous literature [36,37] demonstrated that lesions
in the brain can be monitored with excellent contrast
using T1-W FSE. However in the previous studies only
thermal lesions were shown. In this paper we have ex-
plored extensively the use of MRI to image both lesions
created under thermal mechanisms and under boiling.
With T1-w FSE the signal intensity of the brain tissue is
homogeneous and therefore the contrast with thermal
lesions or with bubbly lesions is excellent. In this study
it was concluded that in brain T1-weighted FSE was the
optimum MRI sequence not only to detect non-boiling
lesions, but also boiling lesions.
It was observed that bubbly lesions appear darker than
thermal lesions. Bubbly lesions appear dark, due to the
air spaces resulting from cavities caused by boiling or
sometimes by cavitation. In the brain tissue in vitro it
was demonstrated that non-degassed excised tissue is a
good model for easily initiating enhanced heating or
cavitation since air is trapped in the vasculature of the
brain. This model of cavitation might not be of any sig-
nificant for clinical use since in live tissue cavitation will
not occur with such exposure; however, this model of
initiating cavitation is very useful for the purpose of
studying the MRI appearance of bubbly lesions. The
exposure of using 2000 W/cm2 in brain results to tem-
peratures of around 90ºC (Figure 12) in vivo. However
with the same exposure in vitro the resulting estimated
maximum temperature was 110ºC. This is probably due
to enhanced heating or cavitation which results to tissue
boiling. The enhanced heating due to bubbles was also
noted in the studies by Fry et al. [51], by ter Haar [52],
and by Chavier et al. [53].
This is the first paper demonstrating creation of large
lesions in brain formed by scanning the transducer in
grid formation. Both thermal and bubbly lesions were
monitored successfully using T1-w FSE with excellent
contrast, proving the potential of HIFU to treat reliably
tumours in the brain in the future. The proposed system
effectively creates large lesion in brain and on the same
time these lesions are effectively monitored using MRI
with strong confidence on the margins of these lesions
especially using T1-w FSE.
Boiling bubbles scatter and reflect ultrasound. These
reflections result in shielding the HIFU focus and thus
increased prefocal heating is observed. Therefore boiling
bubbles, similar to cavitation bubbles, result to distortion
of the lesion from a cigar shape into a tadpole shape
(Figure 4). In addition to this the growth of the lesion is
shifted towards the transducer. This phenomenon which
is attributed to bubbles, either due to ultrasound-induced
cavitation or to boiling has been observed also in the
study by ter Haar [53], and by Chavier et al. [52]. For
the scanned lesions the focus is not shifted towards the
transducer, because while boiling begins in a very short
time in both single and scanned lesions, it cannot distort
the lesion for scanned exposures because the focus of the
transducer moves away from the boiling site. This
speculation is also supported in the paper by Khokhlova
C. Damianou et al. / J. Biomedical Science and Engineering 3 (2010) 763-775 773
Copyright © 2010 SciRes. JBiSE
This paper also showed by means of MRI images that
focal beam is distorted during the occurrence of boiling.
Initially low intensity was used (1000 W/cm2) and the
temperature elevation was monitored using T1-weigthed
FSPGR. The shortest acquisition time that we could
achieve with our system was 5 s. Modifying any MRI
parameter to decrease the time resulted to poor contrast
or low signal to noise ratio (SNR). If we had achieved
faster acquisitions (less than 5 s), then it would have
possible to see more drastic changes. A decrease in the
signal (black spot) demonstrated the increase in the
temperature (also observed by Hynynen at al [54]). The
shape of the black spot was circular. The black spot in-
creased gradually with increased temperature, and the
shape remains circular at all times (Figure 8). When the
intensity was increased to 3500 W/cm2, the shape of the
black spot was irregular indicating that boiling occurred
(Figure 9).
[1] Damianou, C. (2003) In vitro and in vivo ablation of
porcine renal tissues using high intensity focused ultra-
sound. Ultrasound in Medicine & Biology, 29(9), 1321-
[2] Damianou, C., Pavlou, M., Velev, O., Kyriakou, K. and
Trimikliniotis, M. (2004) High intensity focused ultra-
sound ablation of kidney guided by ΜRI. Ultrasound in
Medicine & Biology, 30(3), 397-404.
[3] Watkin, N.A., Morris, S.B., Rivens, I.H. and Ter Haar,
G.R. (1997) Highintensity focused ultrasound ablation of
the kidney in a large animal model. Journal of En-
dourology, 11(3 ), 191-196.
[4] Roberts, W.W., Hall, T.L., Ives, K., Wolf, J.S., Jr., Fowlkes,
J.B. and Cain, C.A. (2006) Pulsed cavitational ultrasound:
A noninvasive technology for controlled tissue ablation
(histotripsy) in the rabbit kidney. Journal of Urology,
175(2), 734-738.
[5] Vallancien, G., Chartier-Kastler, E., Harouni, M., Chopin,
D. and Bougaran, J. (1993) Focused extracorporeal py-
rotherapy: Experimental study and feasibility in man.
Seminars in Urology, 11(1 ), 7-9.
[6] Susani, M., Madersbacher, S., Kratzik, C., Vingers, L. and
Marberger, M. (1993) Morphology of tissue destruction
induced by focused ultrasound. European Urology, 23(1),
[7] Wu, F., Wang, Z.B., Chen, W.Z., Bai, J., Zhu, H. and
Qiao, T.Y. (2003) Preliminary experience using high in-
tensity focused ultrasound for the treatment of patients
with advanced stage renal malignancy. Journal of Urol-
ogy, 170(6), 2237-2240.
[8] Marberger, M., Schatzl, G., Cranston, D. and Kennedy,
J.E. (2005) Extracorporeal ablation of renal tumours with
high-intensity focused ultrasound. British Journal of
Urology International, 95(2), 52-55.
[9] Hacker, A., Michel, M.S., Marlinghaus, E., Kohrmann,
K.U. and Alken, P. (2006) Extracorporeally induced ab-
lation of renal tissue by high-intensity focused ultrasound.
British Journal of Urology International, 97(4), 779-785.
[10] Klingler, C., Susani, M., Seip, R., Mauermann, J., Sanghvi,
N. and Marberger, M. (2008) A novel approach to energy
ablative therapy of small renal tumours: Laparoscopic
high-intensity focused ultrasound. European Urology,
53(4), 810-818.
[11] Taylor, K.J. and Connolly, C.C. (1969) Differing hepatic
lesions caused by the same dose of ultrasound. Journal of
Pathology, 98(4), 291-293.
[12] Linke, C.A., Carstensen, E.L., Frizzell, L.A., Elbadawi,
A. and Fridd, C.W. (1973) Localized tissue destruction
by high-intensity focused ultrasound. Archives of Surgery,
107(6), 887-891.
[13] Frizzell, L.A. (1988) Threshold dosages for damage to
mammalian liver by high intensity focused ultrasound.
IEEE Transactions on Ultrasonics, Ferroelectrics and
Frequency Control, 35(5), 578-581.
[14] Ter Haar, G. and Robertson, D. (1993) Tissue destruction
with focused ultrasound in vivo. European Urology, 23(1),
[15] Chen, L., Ter Haar, G., Hill, C.R., Dworkin, M., Carno-
chan, P., Young, H. and Bensted, J.P. (1993) Effect of
blood perfusion on the ablation of liver parenchyma with
high-intensity focused ultrasound. Physics in Medicine
and Biology, 38(11), 1661-1673.
[16] Sibille, A., Prat, F., Chapelon, J.Y., Abou el Fadil, F.,
Henry, L., Theillere, Y., Ponchon, T. and Cathignol, D.
(1993) Extracorporeal ablation of liver tissue by high-
intensity focused ultrasound. Oncology, 50(2), 375-379.
[17] Moore, W.E., Lopez, R.-M., Mathews, D.E., Sheets, P.W.,
Etchison, M.R., Hurwitz, A.S., Chalian, A.A., Fry, F.J.,
Vane, D.W. and Grosfeld, J.L. (1989) Evaluation of high-
intensity therapeutic ultrasound irradiation in the treat-
ment of experimental hepatoma. Journal of Pediatric
Surgery, 24(1), 30-33.
[18] Yang, R., Reilly, C.R., Rescorla, F.J., Faught, P.R., Sanghvi,
N.T., Fry, F.J., Franklin, T.D., Jr., Lumeng, L. and Gros-
feld, J.L. (1991) Highintensity focused ultrasound in the
treatment of experimental liver cancer. Archives of Sur-
gery, 126(8), 1002-1010.
[19] Sibille, A., Prat, F., Chapelon, J.Y., Abou el Fadil, F.,
Henry, L., Theilliere, Y., Ponchon, T. and Cathignol, D.
(1993) Characterization of extracorporeal ablation of
normal and tumor-bearing liver tissue by high intensity
focused ultrasound. Ultrasound in Medicine & Biology,
19(9), 803-813.
[20] Prat, F., Centarti, M., Sibille, A., Abou el Fadil, F.A.,
Henry, L., Chapelon, J.Y. and Cathignol, D. (1995) Ex-
tracorporeal high-intensity focused ultrasound for VX2
liver tumors in the rabbit. Hepatology, 21(3), 832-836.
[21] Ter Haar, G., Rivens, I., Chen, L. and Riddler, S. (1991)
High intensity focused ultrasound for the treatment of rat
tumours. Physics in Medicine and Biology, 36(11), 1495-
[22] Ter Haar, G., Clarke, R.L., Vaughan, M.G. and Hill, C.R.
(1991) Trackless surgery using focused ultrasound: Tech-
nique and case report. Minimally Invasive Therapy and
Allied Tech nolo gies , 1(1), 13-19.
[23] Vallancien, G., Harouni, M., Veillon, B., Mombet, A., Pra-
potnich, D., Brisset, J.M. and Bougaran, J. (1992) Fo-
cused extracorporeal pyrotherapy: Feasibility study in
man. Journal of Endourology, 6, 173-181.
774 C. Damianou et al. / J. Biomedical Science and Engineering 3 (2010) 763-775
Copyright © 2010 SciRes. JBiSE
[24] Wu, F., Chen, W. and Bai, J. (1999) Effect of high-inten-
sity focused ultrasound on patients with hepatocellular
cancer–preliminary report. Chines e Journal of Ult rasonog,
8, 213-216.
[25] Li, C.X., Xu, G.L., Jiang, Z.Y., Li, J.J., Luo, G.Y., Shan,
H.B., Zhang, R. and Li, Y. (2004) Analysis of clinical ef-
fect of high-intensity focused ultrasound on liver cancer.
World Journal of Gastroenterology, 10(15), 2201-2204.
[26] Wu, F., Wang, Z.B., Chen, W.Z., Zou, J.Z., Bai, J., Zhu,
H., Li, K.Q., Jin, C.B., Xie, F.L. and Su, H.B. (2005)
Advanced hepatocellular carcinoma: Treatment with high-
intensity focused ultrasound ablation combined with
transcatheter arterial embolization. Radiology, 235(5), 659-
[27] Gedroyc, W.M. (2006) Magnetic resonance guided fo-
cused ultrasound (MRgFUS) treatment of liver tumours.
In: Coussios, C.C., Ed., 6th International Symposium on
Therapeutic Ultrasound, Oxford, 30 August-2 September
2006, 539-547.
[28] Fry, W., Mosberg, W., Barnard, J. and Fry, F. (1954) Pro-
duction of focal destructive lesions in the central nervous
system with ultrasound. Journal of Neurosurgery, 11(2 ),
[29] Lele, P.P. (1962) A simple method for production of
trackless focal lesions with focused ultrasound. Journal
of Physiology, 160(3), 494-512.
[30] Fry, F. and Johnson, L.K. (1978) Tumor irradiation with
intense ultrasound. Ultrasound in Medicine & Biology,
4(4), 337-341.
[31] Britt, R.H., Lyons, B.E., Pounds, D.W. and Prionas, S.D.
(1983) Feasibility of ultrasound hyperthermia in the treat-
ment of malignant brain tumors. Medical Instrumentation,
7(2), 172-177.
[32] Guthkelch, A.N., Carter, L.P., Cassady, J.R., Hynynen,
K.H., Iacono, R.P., Johnson, P.C., Obbens, E.A., Roemer,
R.B., Seeger, J.F. and Shimm, D.S. (1991)Treatment of
malignant brain tumors with focused ultrasound hyper-
thermia and radiation: results of a phase I trial. Journal of
Neuro-Oncology, 10(3), 271-284.
[33] Vykhodtseva, N.I., Hynynen, K. and Damianou, C. (1994)
Pulse duration and peak intensity during focused ultra-
sound surgery: Theoretical and experimental effects in
rabbit brain in vivo. Ultrasound in Medicine & Biology,
20(9), 987-1000.
[34] Hynynen, K., Vykhodtseva, N.I., Chung, A.H., Sorren-
tino, V., Colucci, V. and Jolesz, F.A. (1997) Thermal ef-
fects of focused ultrasound on the brain: Determination
with MR imaging. Radiology, 204(1), 247-253.
[35] Vykhodtseva, N., Sorrentino, V., Jolesz, F.A., Bronson,
R.T. and Hynynen, K. (2000) MRI detection of the ther-
mal effects of focused ultrasound on the brain. Ultra-
sound in Medicine & Biology, 26(5), 871-880.
[36] Hynynen, K., McDannold, N., Vykhodtseva, N. and Jolesz,
F. (2001) Noninvasive MR imaging-guided focal opening
of the blood-brain-barrier in rabbits. Radiology, 220(3),
[37] Hynynen, K., McDannold, N., Martin, H., Jolesz, F., Vy-
khodtseva, N. (2003) The threshold for brain damage in
rabbits induced by bursts of ultrasound in the presence of
an ultrasound contrast agent (optison). Ultrasound in
Medicine & Biology, 29(3), 473-481.
[38] Hynynen, K., McDannold, N., Sheikov, N., Jolesz, F.,
Vykhodtseva, N. (2005) Local and reversible blood-brain
barrier disruption by noninvasive focused ultrasound at
frequencies suitable for trans-skull sonications. NeuroI-
mage, 24(1), 12-20.
[39] Jolesz, F.A. and Jakab, P.D. (1991) Acoustic pressure
wave generation within a magnetic resonance imaging
system: Potential medical applications. Magnetic Reso-
nance Imaging, 1(5), 609-613.
[40] Hynynen, K., Darkazanli, A., Damianou, C., Unger, E.
and Schenck, J.F. (1992) MRI-guided ultrasonic hyper-
thermia. 1992 Radiological Society of North America
Annual Meeting, September 1992.
[41] Hynynen, K., Damianou, C., Darkazanli, A., Unger, E.
and Schenck, J.F. (1993) The feasibility of using MRI to
monitor and guide noninvasive ultrasound surgery. Ul-
trasound in Medicine & Biology, 19(1), 91-92.
[42] Hynynen, K., Darkazanli, A., Damianou, C.A., Unger, E.
and Schenck, J.F. (1993) Tissue thermometry during ul-
trasound exposure. European Urology, 23(1), 12-16.
[43] Hynynen, K., Freund, W.R., Cline, H.E., Chung, A.H.,
Watkins, R.D., Vetro, J.P. and Jolesz, F.A. (1996) A clini-
cal, noninvasive, MR imaging-monitored ultrasound sur-
gery method. Radiographics, 16(1), 185-195.
[44] Khokhlova, V., Bailey, M., Reed, J., Cunitz, B., Kacz-
kowski, P. and Crum, L. (2006) Effects of nonlinear
propagation, cavitation, and boiling in lesion formation
by high intensity focused ultrasound, in a gel phantom.
Journal of the Acoustical Society of America, 119(3), 1834-
[45] Coussios, C., Farny, C., Ter Haar, G. and Roy, R. (2007)
Role of acoustic cavitation in the delivery and monitor-
ing of cancer treatment by high-intensity focused ultra-
sound (HIFU). I nter national Journal of Hyp erthermia, 23(2),
[46] Chung, A.H., Hynynen, K., Colucci, V., Oshio, K., Cline,
H.E. and Jolesz, F.A. (1996) Optimization of spoiled gra-
dient-echo phase imaging for in vivo localization of a
focused ultrasound beam. Magnetic Resonance in Medi-
cine, 36(5), 745-752.
[47] Vykhodtseva, V., Sorrentino, V., Jolesz, F., Bronson, R.
and Hynynen, K. (2000) MRI detection of the thermal
effects of focused ultrasound on the brain. Ultrasound in
Medicine & Biology, 26(5), 871-880.
[48] Quesson, B., Zwart, J. and Moonen, C. (2000) Magnetic
resonance temperature imaging for guidance of ther-
motherapy. Journal of Magnetic Resonance Imaging, 12(4),
[49] Salomir, R., Palussière, J., Vimeux, F.C., de Zwart, J.A.,
Quesson, B., Gauchet, M., Lelong, P., Pergrale, J., Grenier,
N. and Moonen, C.T.W (2000) Local hyperthermia with
MR-guided focused ultrasound: Spiral trajectory of the
focal point optimized for temperature uniformity in the
target region. Journal of Magnetic Resonance Imaging,
12(4), 571-583.
[50] McDannold, N., Hynynen, K. and Jolesz, F. (2000) MRI
monitoring of the thermal ablation of tissue: Effects of
long exposure times. Journal of Magnetic Resonance
Imaging, 13(3), 421-427.
[51] Fry, F.J., Kossoff, G., Eggleton, R.C. and Dunn, F. (1970)
Threshold ultrasound dosages for structural changes in
C. Damianou et al. / J. Biomedical Science and Engineering 3 (2010) 763-775 775
Copyright © 2010 SciRes. JBiSE
the mammalian brain. Journal of the Acoustical Society
of America , 48(6), 1413-1417.
[52] ter Haar, G. (1995) Ultrasound focal beam surgery. Ul-
trasound in Medicine & Biology, 21(9), 1089-1100.
[53] Chavrier, F., Chapelon, Y., Gelet, A. and Cathignol, D.
(2000) Modelling of high-intensity focused ultrasound-
induced lesions in the presence of cavitation bubbles.
Journal of the Acoustical Society of America, 108(1),
[54] Hynynen, K., Damianou, C.A., Colucci, V., Unger, E.,
Cline, H.H. and Jolesz, F.A. (1995) MR monitoring of
focused ultrasonic surgery of renal cortex: Experimental
and simulation studies. Journal of Magnetic Resonance
Imaging, 5(3), 259-266.