Ultrasound in Med. & Biol., Vol. 35, No. 7, pp.

10921097, 2009
Copyright 2009 World Federation for Ultrasound in Medicine & Biology
Printed in the USA. All rights reserved
0301-5629/09/$see front matter

doi:10.1016/j.ultrasmedbio.2009.01.004

Original Contribution
QUANTIFICATION OF MUSCLE VOLUME BY ECHOGRAPHY: COMPARISON
WITH MRI DATA ON SUBJECTS IN LONG-TERM BED REST
PHILIPPE ARBEILLE,* PASCALINE KERBECI,* ARNAUD CAPRI,* CHARLES DANNAUD,*
SCOTT W. TRAPPE,y and TODD A. TRAPPEy
* UMPS-Unite de Medecine et Physiologie Spatiales, Department Medecine Nucleaire et Ultrasons,CHU Trousseau, Tours,
France; and y Human Performance Laboratory, Ball State University, Muncie, IN, USA
(Received 28 January 2008; revised 23 December 2008; in final form 9 January 2009)

AbstractThe objective of this study was to determine the accuracy of an echographic method for measuring the
change in leg muscle volume against the gold standard, magnetic resonance imaging (MRI). Leg muscle volume
was measured using an echographic scanner, which consisted of two metallic rails on which a probe holder moved
via an electric engine. Ten to 20 transverse muscle views were collected along the area scanned, and the muscle
cross-sectional area (CSA, cm2) was measured on each of them. The integration of all the CSAs along the scanned
area provided the muscle volume (cm3). Echographic results were compared with MRI data on 24 subjects undergoing 60 d of bed rest (8 control Con, 8 with exercise countermeasures Ex and 8 with nutrition countermeasures Nut). The vastus intermedius (VI) and the vastus medialis (VM) volumes decreased significantly and
similarly in both Con and Nut (VI, 17%; VM, 21%; p , 0.02). In the Ex group, the VI and VM did not change
significantly. The correlation coefficient between the muscle volume change measured with the echographic and
MRI methods was 0.78. The present study confirms that the echographic scanner is sufficiently accurate for assessing muscle volume changes and detects the effect of exercise countermeasures on muscle volume during long-term
bed rest. (E-mail: arbeille@med.univ-tours.fr) 2009 World Federation for Ultrasound in Medicine & Biology.
Key Words: Muscle, Atrophy, Echography, MRI, Bed rest.

was accurate and could be used in the aforementioned

populations more feasibly than the current approaches.
Because of its size, low cost and relative ease of use,
echography could serve as such a tool to accurately
measure human skeletal muscle volume. However, 3-D
echographs do not have a sufficient scanning angle for
insonating and measuring large parts of human muscles
(i.e., muscle volume). Thus, we developed a mechanical
probe holder that could move a 2-D probe along the
muscles and measure muscle volume by scanning a dedicated part of the muscle. The system was validated on
subjects in which muscle atrophy was induced by putting
them in head-down bed rest (HDBR) for two months. The
study was designed to mimic the undesirable effects of
microgravity on the main physiological systems,
including the cardiovascular and musculoskeletal
systems, and to test exercise and nutritional countermeasures designed to prevent these effects.
Loss of muscle mass and function has been one of the
hallmarks of exposure to real and simulated microgravity
(Adams et al. 2003; Alkner et al. 2004a, 2004b; Fitts et al.
2000, 2001; Tesch et al. 2005; Trappe et al. 2001a).

INTRODUCTION
Presently, muscle volume in humans is estimated by relatively simple and inaccurate anthropometric measurements (e.g., measuring the thigh perimeter at different
levels) or with sophisticated, cumbersome and expensive
magnetic resonance imaging (MRI) (Trappe et al. 2001b)
or computed tomography (Radegran et al. 1999) methodologies. Quantifying muscle volume loss is critical for the
treatment of sarcopenia (the age-related loss of muscle
mass) and for assessing the recovery of muscle atrophy
induced by immobilization after trauma or surgery. Accurate measurements of muscle volume are also important
for humans travelling on long-duration space missions,
such as on the International Space Station or future trips
to the moon or other planets with reduced gravity. Our
objective was to develop a new tool for assessing physiological and pathological changes in muscle volume that

Address correspondence to: Pr Philippe Arbeille, Med Nucl Ultrasons, CHU Trousseau, 37044 Tours, France. E-mail: arbeille@med.
univ-tours.fr
1092

Comparison of muscle volume by echography vs. MRI d P. ARBEILLE et al.

Muscle volume of the knee extensors decreases significantly after short and long-duration space flights (8 to
17 d: 5 to 15%; 112 to 196 d: 10%), despite exercise countermeasures and dietary control (caloric intake requirements). However, because of technical constraints,
measurements (usually MRI) are restricted to before and
after flight, with no real time indication of the efficacy
of the prescribed countermeasures (LeBlanc et al. 2000).
Ground-based studies show the knee extensor muscle
volume loss approaches 20% over several months and
can be partially or completely prevented with exercise
countermeasures (Akima et al. 2001; Alkner et al.
2004a, 2004b; Blottner et al. 2006; Gallager et al. 2005;
Tesch et al. 2004; Trappe et al. 2004, 2007).
The objective of this study was to determine the
accuracy of an echographic method for measuring the
change in leg muscle volume against the gold standard,
MRI. Echographic and MRI measurements were performed on the same subjects before and at the end of
two months of HDBR. Some of the subjects used exercise
or nutritional countermeasures designed to reduce or
prevent muscle atrophy.
MATERIALS AND METHODS
Principle of the ultrasound muscle scanner
A conventional echographic probe (5 to 10 MHz
sector array, Challenge 2000, Esaote, Firenze, Italy) was
fixed on a probe holder fabricated by the investigators
that moved continuously along two parallel metallic rails
(25 cm long and 6 cm apart) via a small electric engine
(Model: ESCAP, reference:26N58-216E1, brushed DC,
12v, 5.7w; by PORTESCAP Cie, La Chaux-de-Fonds
Canton de Fribourg, Switzerland) (Fig. 1). The DC motor
uses a rotor, of which the active part simply consists of
a self-supporting cylindrical copper winding without an
iron core. This design results in characteristics such as
low rotor inertia, low friction, low starting voltage,
absence of iron loss, very high efficiency, good heat dissipation and a linear speed torque curve. The two rails are
parallel to the long axis of the thigh, with the probe holder
being oriented to acquire transverse cross-sectional views
of the thigh muscles during the scanning. The engine
provided sufficient power to move the probe (and its
cable) continuously along the rails at various speeds
from 15cm/s. The location and orientation of the two
rails along the limb were adjusted according to the
muscles to be investigated.
Ultrasound muscle scanning and data processing
procedure
The ultrasound muscle scanning device was in
contact with the upper pole of the patella, which served
as a reference for measurements at different time points

1093

Fig. 1. The motorized probe holder. The probe holder moves

along the two metallic rails, which allows for the imaging of
20 cm of the muscle. The probe holder is in contact with the
upper pole of the patella, which serves as a reference for
measurements at different sessions.

and allowed the scan area to start 6 cm proximal to the

patella. Moreover, the device was placed to visualize the
muscles of interest in the center of the image (vastus intermedius [VI] and vastus medialis [VM]). The maximal
length covered by the scanner was 19.6 cm, over which
20 echographic slices were collected. Due to the shape
of the scanner and the localization of the muscles, the
entire muscle belly was not necessarily covered during
the scan. The light weight (300 g) of the device minimized
compression of muscles during the scanning. The same
scanner speed was applied for each subject at the preHDBR and HDBR day 55 measurement sessions.
The video sequence was stored and then processed
via software (VirtualDub-MPEG2, 1.6.11, FREE Software Foundation, Cambridge, MA, USA), which extracted
1020 consecutive equidistant transaxial images (JPEG
format). A MATLAB 6.5 R13 program (The MathWorks,
Natick, MA, USA) was used to analyze the series of axial
images. The outline of each muscle cross-sectional area
(CSA, cm2) on the first echographic images was traced
manually. The muscle CSAs on the remaining images
were automatically traced and calculated using adapted
software written with MATLAB 6.5 R13 (Fig. 2). The
volume of each muscle (cm3) was obtained by summing
the CSAs over the distance scanned by the probe.
Comparison of the echography and MRI data during the
WISE Bedrest Study
A 60-d head-down tilt (6 ) bed rest study (WISE
HDBR) was completed at the MEDES Institute at Rangueil University Hospital, Toulouse, France in 2005.
The study was approved by the French ethical committee
for protection of the human research subjects (CCPPRB),
and each subject was informed about the study before

1094

Ultrasound in Medicine and Biology

Volume 35, Number 7, 2009

Fig. 2. Six of the 20 consecutive transaxial echographic views of the VI. Each view is separated by 1 cm. The first muscle
CSA is outlined manually (view 4), and on the remaining images the muscle is outlined automatically.

providing signed consent. The study population consisted

of all women and was randomly assigned to one of three
groups: a control group (Con, n 5 8, 34 6 1 y, 163 6 2
cm, 55.6 6 1.4 kg), an exercise countermeasure group
(Ex, n 5 8, 33 6 1 y, 165 6 3 cm, 58.1 6 2.2 kg) or
a nutrition countermeasure group (Nut, n 5 8, 29 6 1 y,
170 6 2 cm, 61.2 6 1.6 kg).
The Ex group performed resistance and aerobic exercise throughout the bed rest period, which has been
described in detail previously (Trappe et al. 2007). Briefly,
resistance exercise was performed on an inertial ergometer
in the 6 head-down tilt position and was focused on the
thigh and calf muscles using the supine squat (SS) and
calf press (CP) exercises, respectively. A total of 19
sessions were scheduled approximately every third day
beginning on day 2 of bed rest (~2 to 3 sessions/week).
Each session consisted of 10 min of light supine cycling
and submaximal SS and CP repetitions for warm up, followed by SS exercise (4 sets of 7 maximal concentric and
eccentric repetitions) and CP exercise (4 sets of 14
maximal concentric and eccentric repetitions). Aerobic
exercise was completed in the supine position on a vertically-oriented treadmill within a lower body negative
pressure (LBNP) device as previously described (Macias
et al. 2007). A total of 29 sessions were scheduled 34
d per week beginning on the first day of bed rest. Each
session consisted of 40 min of exercise followed by 10
min of resting LBNP. The Nut group consumed
a leucine-enriched high-protein diet throughout the bed
rest period (Trappe et al. 2007).
The muscle volume changes as measured by the
ultrasound scanner on the VI 1 VM were compared

with those measured by MRI on VI 1 VL 1 VM on the

same population (Trappe et al. 2007). Ultrasound
measurements were made on 22 of the 24 subjects; therefore, comparisons were made on the 22 individuals that
had both measurements. The MRI included the three
main knee extensor muscles, whereas the ultrasound
scanner could cover only two of them (VI and VM)
because of the limited size of the ultrasound beam. MRI
was completed in a 1.5-T scanner (Integra; Philips
Medical Systems, Best, The Netherlands). After a scout
scan to establish the location of the tibial notch and the
imaging position on the legs, serial interleaved images 8
mm thick were taken for the thigh. The MR images
were analyzed with National Institutes of Health image
software (Image J, version 1.34; NIH, Bethesda, MD,
USA) using manual planimetry. A detailed description
of the manual planimetry measurements for the thigh
muscle has been presented previously (Alkner and Tesch
2004a, 2004b; Trappe et al. 2001b). CSA (cm2) of the
muscle(s) of interest from each slice was determined and
the muscle volume (cm3) was calculated by multiplying
the CSA by the appropriate slice thickness (Trappe et al.
2007).
The ultrasound scanner and MRI measurements were
made before HDBR and on day 55 (ultrasound) and 57
(MRI) of HDBR.
Statistical analysis
Comparison of the absolute values for thigh muscle
volume (pre and HDBR55) was completed with a twoway (group and time) analysis of variance (ANOVA)
with repeated measures. Comparison of the percent

Comparison of muscle volume by echography vs. MRI d P. ARBEILLE et al.

1095

Table 1. VI and VM muscle volume (cm3) measured in the supine position before and after 55 d of HDBR in the exercise (Ex),
control (Con) and nutrition (Nut) groups using the echographic scanner
Vastus intermedius
Group
Ex
Nut
Con

Vastus medialis

Pre HDBR

HDBR 55

Pre-HDBR

HDBR 55

67.88 6 13.74
53.59 6 12.32
60.91 6 16.83

68.17 6 14.51
45.17 6 14.09*
50.52 6 13.00*

64.95 6 14.95
67.58 6 21.35
70.91 6 05.47

62.08 6 14.46
52.75 6 14.96*
56.57 6 11.16*

Values in cm3. Mean values are 6 SD.

*p , 0.02 vs. pre-HDT

change in thigh muscle volume (from pre to HDBR55)

was completed with a one-way (group) ANOVA. Significance was accepted at p , 0.05. Values are presented as
mean 6 SD. The correlation between ultrasound and MRI
changes were expressed using the coefficient of correlation (r).
RESULTS
The system was tested in vitro on different soft materials (silicone cylinders and cubes, and pieces of beefsteak), and the error on the volume (100 to 500 cm3)
was less than 3%. The volume of each piece was determined by measuring the water displacement caused by
immersion and compared with the volume measured on
the echographic views collected with the ultrasound
scanner.
The VI and VM were investigated with the lowest
part of the probe support in contact with the upper pole
of the patella, thus the movement of the probe allowed
to scan an area of 20 cm long, starting at 6 cm from the
patella. The time for one acquisition and data processing
was less than 20 min for one muscle.
Table 1 shows the VI and VM muscle volume data
(cm3) before and after 55 d of bed rest in the Con, Ex
and Nut groups. From prebed rest to HDBR day 55,
the VI and VM volumes decreased significantly and similarly in both Con and Nut (VI, 17%; p , 0.02; VM,
21%; p , 0.02). In the Ex group, the VI and VM did
not change significantly (Fig. 3).
Comparison of the data obtained with echography
and MRI are presented in Figs. 4 and 5. The correlation
between the percent change in volume measured with
the echographic scanner (VI 1 VM) and with MRI (VI
1 VM 1 VL) was 0.78 (Fig. 5).

acquisition run (1.5 cm/s) was sufficient to collect the

images for the muscle volume calculation. Automatic
muscle edge detection significantly shortened the time
required for measuring all of the muscle CSAs and ultimately determining the muscle volume.
Because of technical and anatomical reasons, the
echographic method investigated only a portion of the
thigh muscle. The ultrasound probe must remain in
contact with the thigh skin while moving along the rails
of the motorized probe holder. Because most muscles
are not rectilinear structures, it is difficult to remain
centered on each muscle while using a probe holder
moving along two linear rails. Thus, we decided to limit
the length of muscle investigated to the linear part of the
muscle and the area were the skin surface remained flat.
The MRI system does not have such a problem because
the detectors are not in contact with the skin and thus
can follow the muscles even if the skin surface is tortuous.
However, using the patella as the reference starting point
for the ultrasound method, the same part of each thigh
muscle could be easily investigated several weeks or
months later. Despite the fact that the muscle volume
investigated by ultrasound was relatively small compared
with the MRI, the evaluation of the percent change in
volume at various time points was possible and the results
of the ultrasound scanner closely followed those provided
by MRI.
At the end of the HDBR period, the vastus volumes
of the Con and Nut groups were significantly and similarly
Ex

Con

Nut

5
-5
-15

DISCUSSION
The light probe holder (300 g) was easy to place and
maintain on the thigh of the subjects during the scanning
operation. One or two runs of the motorized probe holder
were performed at high speed (5 cm/s) to confirm the
muscle of interest was scanned appropriately, and one

-25
%VI

%VM

-35

Fig. 3. Percent changes in the muscle volume of the VI (%VI)

and VM (%VM) in the exercise (Ex), control (Con) and nutrition
(Nut) groups using the echographic scanner (p , 0.05 Ex vs. Con
and Nut). Values are mean 6 SD.

1096

Ultrasound in Medicine and Biology

Ex

Con

Nut

-10
-20
-30
-40

%(VI+VM) US
%(VI+VM+VL) MRI

Fig. 4. Percent change of thigh muscle as measured by echography (VI 1 VM) and by MRI (VI 1 VM 1 VL). MRI data are
reproduced with permission (Trappe et al. 2007).

decreased by both methods compared with pre-HDBR.

Similarly, both methods showed that the vastus volumes
were unchanged in the Ex group over the HDBR period.
The correlation coefficient (0.78) confirms the accuracy
of the ultrasound method compared with the MRI gold
standard. Nevertheless, there was a higher variability of
the echographic measurements compared with the MRI.
This difference may be related to the fact that the echographic volumes were much smaller than the MRI and
that the contour of the muscles on the MR images were
more clearly delineated than on the echographic images.
The MRI measurements used to quantify muscle
atrophy induced by exposure to microgravity are accessible only before and after space flight. In addition,
previous studies of HDBR, also based on MRI measurements, support the current results. That is, thigh muscle
mass and function are significantly decreased with simulated space flight exposure and can be partially or
completely prevented with exercise countermeasures
(Adams et al 2003; Akima et al. 2001; Alkner et al.
2004a, 2004b; Blottner et al. 2006; Gallager et al. 2005;
Tesch et al. 2004; Trappe et al. 2004, 2007). However,
the techniques used in-flight are limited to insufficient
anthropometric measurements (Fitts et al 2000; 2001).
The ergonomics of the echographic scanner and the quick
data processing makes the echographic method very well
adapted for quantifying muscle mass, and thus monitoring
the efficacy of the in-flight countermeasures to combat
muscle atrophy, during future space flight missions.
10
%(VI+VM) Ultrasound
-40

-30

-20

-10

10

-10
-20
-30
y = 0,7162x - 7,2049
R = 0,78

%(VI+VM+VL)
MRI

-40

Fig. 5. Comparison between the percent volume change from

pre to the end of HDBR, as measured by the echographic scanner
(VI 1 VM; x-axis) and MRI (VI 1 VM 1 VL; y-axis).

Volume 35, Number 7, 2009

The findings from the current study reinforce the

potential benefits of using the echographic method, which
is easy to set up and use, easy to access and accurate
enough for assessing muscle atrophy across time.
Conventional echographic practice requires several
months of training, whereas the probe holder of the ultrasound scanner can be manipulated by a person not highly
trained in clinical ultrasound. The operator needs to know
the main anatomical references from which the scanner
will be positioned and must be familiar with the ultrasound image of the relevant transverse sections. The
only issue is to be sure the operator places the probe holder
centered on the muscle axis such that the transverse view
of the muscle will completely capture the muscle of
interest. In such a case, a second high-speed image capture
could be repeated after repositioning of the probe holder.
CONCLUSION
The muscle volume data measured by the
echographic scanner confirmed the MRI data. That is,
the echographic method tracked the muscle volume loss
because of the bed rest, the prevention of muscle loss
with the exercise countermeasure or the lack of influence
of the nutrition countermeasure. The ultrasound scanner
allowed the accurate evaluation of the volume change of
various muscles by an easy and semi-automated method.
AcknowledgmentsThe study WISE-2005 was sponsored by the European Space Agency (ESA), the National Aeronautics and Space Administration of the USA (NASA), the Canadian Space Agency (CSA) and the
French space agency Centre National dEtudes Spatiales (CNES). The
authors would like to thank the radiology staff at Rangueil Hospital (Virginie Lorite, Ghylene Enrique, Christine Mazars, Muriel Rogawski) for
their efforts with this investigation and the sonographers of the Dept
Med Nucl et Ultrasons (M. Porcher, V. Moreau, J. L. Boulay) from
CHU Trousseau in Tours. The MRI phase of the investigation was supported by a NASA grant (NNJ04HF72G) to S. and T. Trappe, and the
ultrasound phase by CNES grant (4800000326) to P. Arbeille.

REFERENCES
Adams GR, Caiozzo VJ, Baldwin KM. Skeletal muscle unweighting:
Spaceflight and ground-based models. J Appl Physiol 2003;95:
21852201.
Akima H, Kubo K, Imai M, Kanehisa H, Suzuki Y, Gunji A, Fukunaga T.
Inactivity and muscle: Effect of resistance training during bed rest on
muscle size in the lower limb. Acta Physiol Scand 2001;172:
269278.
Alkner BA, Tesch PA. Efficacy of a gravity-independent resistance exercise device as a countermeasure to muscle atrophy during 29-day bed
rest. Acta Physiol Scand 2004a;181:345357.
Alkner BA, Tesch PA. Knee extensor and plantar flexor muscle size and
function following 90 days of bed rest with or without resistance
exercise. Eur J Appl Physiol 2004b;93:294305.
Blottner D, Salanova M, Puttmann B, Schiffl G, Felsenberg D,
Buehring B, Rittweger J. Human skeletal muscle structure and function preserved by vibration muscle exercise following 55 days of bed
rest. Eur J Appl Physiol 2006;97:261271.
Fitts RH, Riley DR, Widrick JJ. Physiology of a microgravity environment. Invited review: Microgravity and skeletal muscle. J Appl Physiol 2000;89:823839.

Comparison of muscle volume by echography vs. MRI d P. ARBEILLE et al.

Fitts RH, Riley DR, Widrick JJ. Functional and structural adaptations of
skeletal muscle to microgravity. J Exper Biol 2001;204:32013208.
Gallagher P, Trappe S, Harber M, Creer A, Mazzetti S, Trappe T,
Alkner B, Tesch P. Effects of 84-days of bedrest and resistance
training on single muscle fibre myosin heavy chain distribution in
human vastus lateralis and soleus muscles. Acta Physiol Scand
2005;185:6169.
LeBlanc A, Lin C, Shackelford L, Sinitsyn V, Evans H, Belichenko O,
Schenkman B, Kozlovskaya I, Oganov V, Bakulin A, Hedrick T,
Feeback D. Muscle volume, MRI relaxation times (T2), and body
composition after spaceflight. J Appl Physiol 2000;89:21582164.
Macias BR, Cao P, Watenpaugh DE, Hargens AR. LBNP treadmill exercise maintains spine function and muscle strength in identical twins
during 28-day simulated microgravity. J Appl Physiol 2007;102(6):
22742278.
Radegran G, Blomstrand E, Saltin B. Peak muscle perfusion and oxygen
uptake in humans: Importance of precise estimates of muscle mass.
J Appl Physiol 1999;87:23752380.

1097

Tesch PA, Trieschmann JT, Ekberg A. Hypertrophy of chronically

unloaded muscle subjected to resistance exercise. J Appl Physiol
2004;96:14511458.
Tesch PA, Berg HE, Bring D, Evans HJ, Leblanc AD. Effects of 17-day
spaceflight on knee extensor muscle function and size. Eur J Appl
Physiol 2005;93:463468.
Trappe SW, Trappe TA, Lee GA, Widrick JJ, Costill DL, Fitts RH.
Comparison of a space shuttle flight (STS-78) and bed rest on human
muscle function. J Appl Physiol 2001a;91:5764.
Trappe TA, Lindquist DM, Carrithers JA. Muscle specific atrophy of the
quadriceps femoris with aging. J Appl Physiol 2001b;90:20702074.
Trappe S, Trappe T, Gallagher P, Harber M, Alkner B, Tesch P. Human
single muscle fibre function with 84 day bed-rest and resistance exercise. J Physiol 2004;557(2):501513.
Trappe TA, Burd NA, Louis ES, Lee GA, Trappe SW. Influence of
concurrent exercise or nutrition countermeasures on thigh and calf
muscle size and function during 60 days of bed rest in women.
Acta Physiol (Oxf) 2007;191(2):147149.