Journal of Computer Assisted Tomography:
Diffusion Tensor Imaging of the Normal Foot at 3 T
Elzibak, Alyaa H. MSc*†; Kumbhare, Dinesh A. MSc, MD, FRCPC‡§; Harish, Srinivasan MBBS, FRCPC†∥¶; Noseworthy, Michael D. PhD, PEng*†‡∥¶#
From the *Department of Medical Physics and Applied Radiation Sciences, McMaster University; †Imaging Research Centre, St Joseph’s Healthcare; ‡School of Biomedical Engineering; Departments of §Medicine, Division of Rehabilitation Sciences, ∥Radiology, McMaster University; ¶Department of Diagnostic Imaging, St Joseph’s Healthcare; and #Department of Electrical and Computer Engineering, McMaster University, Hamilton, Ontario, Canada.
Received for publication September 11, 2013; accepted November 1, 2013.
Reprints: Michael D. Noseworthy, PhD, PEng, Electrical and Computer Engineering, McMaster University, 1280 Main St West, Hamilton, Ontario, Canada. L8S 4K1 (e-mail: firstname.lastname@example.org).
Author Elzibak received doctoral postgraduate scholarship from the Natural Sciences and Engineering Research Council of Canada. Author Noseworthy has received lecture honoraria from both GE Healthcare and Bayer Healthcare in the last 12 months.
The other authors report no conflicts of interest.
The objective of this study was to establish normative diffusion tensor imaging (DTI) eigenvalues (λ1,λ2,λ3), apparent diffusion coefficient, and fractional anisotropy in asymptomatic foot muscles.
Ten healthy adults (mean [SD], 25.9 [4.3] years) were examined using a 3-T magnetic resonance imaging scanner. Diffusion tensor imaging indices were evaluated in 5 muscles in the foot: quadratus plantae, abductor hallucis, flexor hallucis brevis, flexor digitorum brevis, and abductor digiti minimi. Signal-to-noise ratio was also measured for each muscle.
In the various foot muscles, λ1 ranged from 1.88 × 10−3to 2.14 × 10−3 mm2/s, λ2 ranged from 1.39 × 10−3 to 1.48 × 10−3 mm2/s, and λ3 ranged from 0.91 × 10−3 to 1.27 × 10−3 mm2/s; apparent diffusion coefficient ranged from 1.48 × 10−3 to 1.55 × 10−3 mm2/s; and fractional anisotropy ranged from 0.21 to 0.40. Statistical differences were seen in some eigenvalues between muscle pairs. Mean signal-to-noise ranged from 47.5 to 69.1 in the various muscles examined.
Assessment of anisotropy of water diffusion in foot muscles was feasible using DTI. The measured DTI metrics in the foot were similar to those in calf and thigh skeletal muscles.
Diffusion tensor imaging (DTI) has become a widely applied technique to characterize the microstructure of anisotropic tissues. The method has been used to noninvasively study various organs, such as the brain,1–3 liver,4 kidney,5,6 heart,7 breast,8 prostate,9 and skeletal muscle.10–13 Diffusion imaging measures the random displacement (Brownian motion) of water molecules, which is isotropic in an unconstrained environment. However, in coherently oriented structures such as muscle fibers, molecular diffusion is hindered by barriers that constrain the free diffusion. This results in anisotropic diffusion, where water molecules move more easily along the axis of the myofibril than perpendicular to it.
Diffusivity of water can be characterized using numerous metrics, including 3 eigenvalues (λ1, λ2, λ3), where λ1 corresponds to diffusion along the principle axis. The average of the eigenvalues is represented by the apparent diffusion coefficient (ADC) or the mean diffusivity. Fractional anisotropy (FA) characterizes the shape of the diffusion ellipsoid and ranges between zero for perfect isotropic diffusion and 1 for anisotropic diffusion, as in an infinite cylinder. In skeletal muscle, an interpretation of the meaning of DTI eigenvalues has been inconclusive. It is generally accepted that λ1 represents diffusion along the muscle fiber axis. Galbán et al14 suggested that λ2 and λ3 might represent diffusion along sheets of fibers and within muscle fiber cross section, respectively.
Since its initial application to examine skeletal muscle in vitro by Basser et al,15 DTI has become a useful tool to probe muscle microstructure in vivo to gain insight about the underlying architecture.16 The technique has been applied to evaluate muscle health and has shown that injured calf muscles are associated with increased eigenvalues and ADC, with a reduction in FA.17 Tracking of muscle fibers in damaged zones further revealed structural abnormalities with injury.17 Diffusion imaging has also been used to evaluate pathological muscle conditions18,19 and muscle disruptions induced by exercise.20 Some DTI studies have demonstrated dependence of diffusivity measures on age21,22 and sex,23 whereas others failed to establish such correlations.13,24 Diffusion tensor imaging metrics have also been shown to depend on muscle fiber architecture and pennation angle.25–27 Changes in diffusion indices have been reported after exercise,28–30 because of ischemic damage,31 and in response to external pressure application.32
Skeletal muscle DTI studies now span a range of muscles located throughout the body, including the thigh,10,18,20 lumbar spine,13 forearm,11 and calf12,14,17,21–34 musculature. However, numerous pathologies are known to affect foot tissue, such as peripheral vascular diseases and complex regional pain syndrome.35,36 These contribute to muscle weakness, disuse, and atrophy, and, in severe cases, could lead to amputation of the affected limb. Although conventional anatomical imaging may aid in diagnosing muscle impairments, an understanding of muscle dysfunction usually requires biopsies or the use of functional imaging methods. The imaging approach is more tolerable by patient populations because biopsies are invasive, painful, and associated with complications. In addition, biopsy of foot muscles is not a routine clinical practice. The importance of examining foot muscles using advanced functional imaging methods has recently been elucidated by Kos et al.37 The researchers explored the use of foot blood oxygen level–dependent magnetic resonance imaging. Using an ischemia-hyperemia paradigm that is widely applied to calf studies, the group demonstrated the feasibility of foot imaging in healthy volunteers.37
We believe that advanced imaging protocols such as functional imaging and DTI could provide valuable clinically relevant information in the evaluation of pathological conditions that manifest with symptoms in the foot. The purpose of this study was to examine the diffusive properties of foot muscles and establish a normative range of values in healthy subjects. To our knowledge, DTI indices (eigenvalues, FA, and ADC) have not been previously investigated in the foot. Assessment of diffusion behavior in an asymptomatic population is necessary before future application of the technique to a group of subjects with disease.
MATERIALS AND METHODS
Ten healthy adult subjects between 19 and 32 years of age (mean [SD] age, 25.9 [4.3] years; height, 173.9 [0.2] cm; weight, 69.9 [12.6] kg; 2 females) took part in the study. The foot muscles of the subject’s dominant leg were imaged. None of the subjects reported having any pathology in the imaged foot. The participants refrained from food and drink intake for at least 4 hours before the study, and none of the subjects performed strenuous exercises in the 24 hours before the imaging session. These measures were imposed in an attempt to standardize our DTI metrics. Some DTI studies require subjects to fast and refrain from intense exercise, whereas others do not report any premeasurement restrictions. All subjects underwent magnetic resonance (MR) screening before taking part in the study, which was approved by our local research ethics board.
Magnetic Resonance Imaging
Magnetic resonance imaging was performed using a GE Signa MR750 3.0-T MRI system and an HD foot/ankle array coil (GE Healthcare, Milwaukee, Wis). The subjects were placed feetfirst and lay in the supine position in the MR scanner. After scout imaging, high contrast T1-weighted coronal (short axis) anatomical images were collected using a fast spin echo sequence with the following parameters: echo time/repetition time, 14.9/785 milliseconds; 5-mm slice thickness; 0-mm gap; field of view, 16 cm; 320 × 320 matrix; 20 slices; and echo train length of 4. Images were acquired starting at the neck of the talus and moving toward the distal end of the foot. The DTI images were subsequently collected from the same anatomical location as that of the T1-weighted images. A dual spin-echo echo planar imaging sequence was used for DTI measurements with 15 diffusion-encoding gradient directions ([1,0,0]; [0.643,0.766,0]; [0.258,0.307,0.916]; [0.258,0.307,0.916]; [0.164,−0.507,0.846]; [−0.796,−0.321,0.513]; [0.761,0.427,0.489]; [−0.506,0.833,0.224]; [0.667,−0.158,0.728]; [0.128,0.959,0.254]; [−0.178,−0.898,−0.403]; [0.255,−0.590,−0.767]; [0.340,−0.736,0.585]; [−0.801,0.329,0.501];[0.336,0.043,−0.941]); 1 b = 0 s/mm2 image; echo time/repetition time, 68.8/4050 milliseconds; b-value, 400 s/mm2; 5-mm slice thickness; 0-mm gap; field of view, 16 cm; frequency encoding along the superior/inferior direction; 64 × 64 matrix; 20 slices; and array spatial sensitivity encoding technique acceleration factor of 2. A shim volume was manually defined in the DTI acquisition. Four DTI data sets with the same prescan values were acquired (4 number of excitations) separately so that each set can be corrected before calculation of the tensor (as described below). In skeletal muscle DTI studies, a range of parameters have been used (b-values from 300–600 s/mm2; 6, 12, or 15 gradient directions; 4–8 averages). The DTI parameters that we have used are in the range of those typically used in muscle DTI studies and have been standardized in our laboratory for skeletal muscle imaging. To determine the signal-to-noise ratio (SNR) for each muscle, an additional identical DTI data set was collected so that the difference method38 could be used in the SNR computation (as described below).
Diffusion tensor imaging indices were measured in 5 muscular regions of interest (ROIs) in the foot: quadratus plantae (QP), abductor hallucis (AH), flexor hallucis brevis (FHB), flexor digitorum brevis (FDB), and abductor digiti minimi (ADM) (Fig. 1). Before tensor analyses, the acquired DTI data sets were corrected for motion and eddy currents using functional magnetic resonance imaging of the brain (FMRIB)’s linear image registration tool, FLIRT,39 which is part of the FMRIB Software Library, FSL (FMRIB Analysis Group).40 The reference image for the registration was set as the b = 0 s/mm2 image of the first number of excitations. After image registration and eddy current correction, the DTI data sets were summed and tensor calculation was carried out using analysis of functional neuroimages (AFNI) software (National Institute of Mental Health).41 The ROIs for each of the investigated muscles were drawn on 5 slices. The SNR was first determined for each muscle ROI. Because parallel imaging was used to acquire the DTI data, SNR was calculated using the difference method as follows38:
where μsum is the mean signal intensity of the ROI in the summed image and σdiff is the standard deviation of the ROI signal in the differenced image. This technique requires the acquisition of 2 separate data sets; hence, 8 DTI scans were performed and the 2 resulting summed images from each of the 4 DTI sets were used to determine the SNR.
Calculation of the diffusion tensor at each voxel of the corrected DTI set was done using AFNI (National Institute of Mental Health)41 resulting in 3 eigenvalues (λ1, λ2, λ3), ADC, and FA, where ADC was computed as follows:
and FA was calculated with the following:
where 〈λ〉 represents the mean of the 3 eigenvalues.
To determine whether there was a significant difference between muscle pairings in terms of eigenvalues, FA, or ADC, a repeated-measures analysis of variance with Tukey honestly significant difference post hoc test was performed (significance was set at P < 0.05). Because 5 muscles were examined in this study, 10 different muscle pairings were possible in total (5 choose 2 = 10) because the order of each muscle pair did not matter (for instance, the QP-AH pairing is the same as the AH-QP pairing). All statistical analyses were performed using GraphPad Prism (Version 5.0 c; GraphPad Software Inc).
Diffusion tensor metrics (λ1, λ2, λ3, ADC, FA) were measured in the foot of 10 healthy volunteers (Table 1). A sample T1-weighted anatomical image and the corresponding b = 0 s/mm2 DTI image from one of the subjects is shown in Figure 2 for 2 slice locations. Figure 3 shows images of the 3 diffusion eigenvalues along with the corresponding T1-weighted anatomical slice for one of the subjects at 1 anatomic level.
Statistical differences were seen in λ1 between the following muscle pairings: QP-AH, QP-FHB, AH-FDB, FHB-FDB, and FDB-ADM (Table 2). None of the muscle pairings showed statistical differences in λ2. However, regarding the third eigenvalue, λ3, statistical differences were seen between QP-FDB, AH-FDB, AH-ADM, FHB-FDB, and FHB-ADM (Table 2). None of the paired muscles showed statistical differences in ADC (Table 2).
The largest FA was measured in the FDB muscle with a mean (SD) of 0.40 (0.07), with all other muscles measuring below 0.30 (Fig. 4). Statistical differences in FA were seen between the following muscle pairs: QP-AH, QP-FHB, QP-FDB, AH-FDB, AH-ADM, FHB-FDB, FHB-ADM, and FDB-ADM (Table 1).
The measured SNR for each of the investigated muscles is given in Table 3. The averaged SNR ranged from 47.5 to 69.1 in the various muscles examined.
The aim of our study was to use DTI to measure the diffusion anisotropy of water in foot muscles of asymptomatic disease–free adult subjects. We characterized diffusion tensor metrics (λ1, λ2, λ3, ADC, FA) in 5 muscles of the foot of healthy volunteers (n = 10). The investigated muscles were as follows: QP, AH, FHB, FDB, and ADM. Although DTI has been previously applied to study various skeletal muscles such as those of the calf,12,14,17,21–34 forearm,11 lumbar spine,13 and thigh,10,18,20 to our knowledge, this is the first study that has extended the technique to examine foot muscles.
Diffusion tensor imaging has been shown to provide useful insight into microstructural muscular abnormalities of the calf and the thigh.17–20 Diffusion tensor imaging–derived metrics have also been shown to alter in the presence of inflammatory markers in the brain,42,43 the liver,4 and the knee joint.44 It is well known that numerous pathologies, including complex regional pain syndrome (CRPS),36 affect foot tissue and result in muscular abnormalities of the affected limb.45,46 There is also evidence suggesting the presence of inflammatory markers in the affected CRPS limb based on blister fluid samples47 and skin biopsies.48 Because of its noninvasive nature and because it has been useful in examining muscular abnormalities and inflammatory changes, DTI of the foot could potentially further our understanding of pathologies that affect this region of the body, including CRPS and peripheral vascular diseases. We hope that this technique may allow earlier diagnosis and allow treatment monitoring with serial scans. Our study was thus performed to examine DTI metrics in foot muscles of an asymptomatic population and demonstrate the feasibility of the technique before future applications to a compromised group of volunteers.
In the current study, we measured λ1 to range from 1.88 × 10−3 to 2.14 × 10−3 mm2/s, λ2 from 1.39 × 10−3 to 1.48 × 10−3 mm2/s, and λ3 from 0.91 × 10−3 to 1.27 × 10−3 mm2/s in the various foot muscles. Our ADC and FA values were observed to range from 1.48 × 10−3 to 1.55 × 10−3 mm2/s and 0.21 to 0.40, respectively. Our measured DTI metrics are similar to those obtained by other groups investigating skeletal muscle DTI.12,14,49 In 1 study, λ1 was observed to vary from 1.93 × 10−3 to 2.29 × 10−3 mm2/s; λ2, from 1.42 × 10−3 to 1.58 × 10−3 mm2/s; and λ3, from 1.19 × 10−3 to 1.40 × 10−3 mm2/s in the muscles of the calf of healthy volunteers.12 Apparent diffusion coefficient and FA ranges of 1.53 × 10−3 to 1.67 × 10−3 mm2/s and 0.18 to 0.33 were reported in that study, respectively.12 Another study investigating diffusive properties of thigh muscles reported λ1 to range from 2.08 × 10−3 to 2.09 × 10−3 mm2/s, λ2 from 1.54 × 10−3 to 1.58 × 10−3 mm2/s, and λ3 from 1.13 × 10−3 to 1.20 × 10−3 mm2/s in the various muscles, with ADC and FA ranges of 1.59 × 10−3 to 1.62 × 10−3 mm2/s and 0.27 to 0.30, respectively.49 These reported DTI values are in agreement with our measured metrics in foot muscles.
We observed variations in some of the DTI metrics between foot muscles. Variation in DTI metrics between muscles (that is, statistical differences between muscle pairs for a given diffusion tensor index) may be a consequence of varying muscle composition or architecture.14 Specifically, statistical differences were seen between QP-AH, QP-FHB, AH-FDB, FHB-FDB, and FDB-ADM for λ1. We did not find any statistical differences in terms of λ2 between any muscle pairs. However, we noted statistical differences in λ3 between QP-FDB, AH-FDB, AH-ADM, FHB-FDB, and FHB-ADM. Previous reports have shown that there are variations in the eigenvalues across calf muscles.12,14 In the study by Sinha et al,12 the largest number of statistically different muscle pairs was seen for λ1, whereas none of the muscles showed differences in terms of λ3. Galbán et al14 reported a different pattern, where they found that the smallest number of pairs of muscles produced statistical differences on the basis of λ1 and that the largest number of statistically different muscle pairs was seen for λ3. The results of the current study show an equal number of statistically different muscle pairs for λ1 and λ3. The variations seen in λ1 between muscle groups have been suggested to stem from the presence of the mitochondria and the sarcoplasmic reticulum, which would result in structural barriers that would vary between muscles.14 Differences seen in λ3 across muscles have been previously explained to be the result of variations in the radius of muscle fibers.14
Signal-to-noise ratio was calculated using the approach for parallel imaging, which requires the acquisition of 2 DTI data sets.38 The same ROIs used for determining the DTI indices were also used in the SNR calculation using the b = 0 images. Our calculated SNR was 49.0 ± 16.0 in the QP muscle, 56.6 ± 25.2 in the AH muscle, 69.1 ± 27.4 in the FHB muscle, 47.5 ± 17.4 in the FDB muscle, and 55.7 ± 23.7 in the ADM muscle. It has been previously shown that the behavior of the DTI metrics depends on the SNR level and the b value used.50 The b value in the current study was 400 s/mm2, and from Figure 3 of the study of Damon,50 it can be seen that, with this b value, for an accuracy of 5% in determining the DTI metrics, the required SNR is approximately 5 for λ1, 8 for λ2,13 for λ3, 9 for ADC, and 20 for FA.50 For an accuracy of 1%, an SNR of 11 is needed for λ1, 18 for λ2, 28 for λ3, 18 for ADC, and 43 for FA.50 The SNR calculated in our study was more than 35 in almost all subjects and muscles examined. Thus, we believe our DTI metrics were reliably determined (between 1% and 5% accuracy) because the SNR levels were higher than minimum requirements, as predicted by Damon.50
In conclusion, we have demonstrated that DTI of the foot is feasible and we have established a normative range of values for the diffusion tensor metrics (λ1, λ2, λ3, ADC, FA) in healthy adult subjects. Our measured DTI indices in the foot were similar to those obtained in calf and thigh skeletal muscles. Future application of the technique to a compromised population could potentially further our understanding of pathologies that affect foot musculature.
The authors thank Norm Konyer for the helpful discussions and Andrew Davis for writing the Bourne-again shell script.
1. Beaulieu C. The basis of anisotropic water diffusion in the nervous system—a technical review. NMR Biomed. 2002; 15: 435–455.
2. Sotak CH. The role of diffusion tensor imaging in the evaluation of ischemic brain injury—a review. NMR Biomed. 2002; 15: 561–569.
3. Bode MK, Ruohonen J, Nieminen MT, et al. Potential of diffusion imaging in brain tumors: a review. Acta Radiol. 2006; 47: 585–594.
4. Taouli B, Chouli M, Martin AJ, et al. Chronic hepatitis: role of diffusion-weighted imaging and diffusion tensor imaging for the diagnosis of liver fibrosis and inflammation. J Magn Reson Imaging. 2008; 28: 89–95.
5. Notohamiprodjo M, Dietrich O, Horger W, et al. Diffusion tensor imaging (DTI) of the kidney at 3 Tesla—feasibility, protocol evaluation and comparison to 1.5 Tesla. Invest Radiol. 2010; 45: 245–254.
6. Sigmund EE, Vivier P-H, Sui D, et al. Intravoxel incoherent motion and diffusion-tensor imaging in renal tissue under hydration and furosemide flow challenges. Radiology. 2012; 263: 758–769.
7. Hsu EW, Muzikant AL, Matulevicius SA, et al. Magnetic resonance myocardial fiber-orientation mapping with direct histological correlation. Am J Physiol. 1998; 274: H1627–H1634.
8. Partridge SC, Murthy RS, Ziadloo A, et al. Diffusion tensor magnetic resonance imaging of the normal breast. Magn Reson Imaging. 2010; 28: 320–328.
9. Gürses B, Kabakci N, Koyanlikaya A, et al. Diffusion tensor imaging of the normal prostate at 3 Tesla. Eur Radiol. 2008; 18: 716–721.
10. Budzik JF, Le Thuc V, Demondion X, et al. In vivo MR tractography of thigh muscles using diffusion imaging: initial results. Eur Radiol. 2007; 17: 3079–3085.
11. Froeling M, Oudeman J, van den Berg S, et al. Reproducibility of diffusion tensor imaging in human forearm muscles at 3.0 T in a clinical setting. Magn Reson Med. 2010; 64: 1182–1190.
12. Sinha S, Sinha U, Edgerton VR. In vivo diffusion tensor imaging of the human calf muscle. J Magn Reson Imaging. 2006; 24: 182–190.
13. Jones GEG, Kumbhare DA, Harish S, et al. Quantitative DTI assessment in human lumbar stabilization muscles at 3 T. J Comput Assist Tomogr. 2013; 37: 98–104.
14. Galbán CJ, Maderwald S, Uffmann K, et al. Diffusive sensitivity to muscle architecture: a magnetic resonance diffusion tensor imaging study of the human calf. Eur J Appl Physiol. 2004; 93: 253–262.
15. Basser PJ, Mattiello J, LeBihan D. Estimation of the effective self-diffusion tensor from the NMR spin echo. J Magn Reson B. 1994; 103: 247–254.
16. Noseworthy MD, Davis AD, Elzibak AH. Advanced MR imaging techniques for skeletal muscle evaluation. Semin Musculoskelet Radiol. 2010; 14: 257–268.
17. Zaraiskaya T, Kumbhare D, Noseworthy MD. Diffusion tensor imaging in evaluation of human skeletal muscle injury. J Magn Reson Imaging. 2006; 24: 402–408.
18. Qi J, Olsen NJ, Price RR, et al. Diffusion-weighted imaging of inflammatory myopathies: polymyositis and dermatomyositis. J Magn Reson Imaging. 2008; 27: 212–217.
19. Saotome T, Sekino M, Eto F, et al. Evaluation of diffusional anisotropy and microscopic structure in skeletal muscles using magnetic resonance. Magn Reson Imaging. 2006; 24: 19–25.
20. Cermak NM, Noseworthy MD, Bourgeois JM, et al. Diffusion tensor MRI to access skeletal muscle disruption following eccentric exercise. Muscle Nerve. 2012; 46: 42–50.
21. Galban CJ, Maderwald S, Stock F, et al. Age dependence of water diffusivity in human skeletal muscle. Proc Intl Soc Mag Reson Med. 2006; 14: 257
22. Galban CJ, Maderwald S, Stock F, et al. Age-related changes in skeletal muscle as detected by diffusion tensor magnetic resonance imaging. J Gerontol A Biol Sci Med Sci. 2007; 62: 453–458.
23. Galbán CJ, Maderwald S, Uffmann K, et al. A diffusion tensor imaging analysis of gender differences in water diffusivity within human skeletal muscle. NMR Biomed. 2005; 18: 489–498.
24. Okamoto Y, Kunimatsu A, Kono T, et al. Gender differences in MR muscle tractography. Magn Reson Med Sci. 2010; 9: 111–118.
25. Sinha U, Sinha S, Hodgson JA, et al. Human soleus muscle architecture at different ankle joint angles from magnetic resonance diffusion tensor imaging. J Appl Physiol (1985). 2011; 110: 807–819.
26. Sinha S, Sinha U. Reproducibility analysis of diffusion tensor indices and fiber architecture of human calf muscles in vivo at 1.5 tesla in neutral and plantarflexed ankle positions at rest. J Magn Reson Imaging. 2011; 34: 107–119.
27. Hatakenaka M, Yabuuchi H, Matsuo Y, et al. Effect of passive muscle length change on apparent diffusion coefficient: detection with clinical MR imaging. Magn Reson Med Sci. 2008; 7: 59–63.
28. Okamoto Y, Kunimatsu A, Miki S, et al. Fractional anisotropy values of calf muscles in normative state after exercise: preliminary results. Magn Reson Med Sci. 2008; 7: 157–167.
29. Deux JF, Malzy P, Paragios N, et al. Assessment of calf muscle contraction by diffusion tensor imaging. Eur Radiol. 2008; 18: 2303–2310.
30. Okamoto Y, Kunimatsu A, Kono T, et al. Changes in MR diffusion properties during active muscle contraction in the calf. Magn Reson Med Sci. 2010; 9: 1–8.
31. Heemskerk AM, Drost MR, van Bochove GS, et al. DTI-based assessment of ischemia-reperfusion in mouse skeletal muscle. Magn Reson Med. 2006; 56: 272–281.
32. Hata J, Yagi K, Hikishima K, et al. Diffusion fractional anisotropy-based transformation in skeletal muscle caused by pressure. Magn Reson Med Sci. 2012; 11: 179–184.
33. Damon BM, Ding Z, Anderson AW, et al. Validation of diffusion tensor MRI based muscle fiber tracking. Magn Reson Med. 2002; 48: 97–104.
34. Saupe N, White LM, Sussman MS, et al. Diffusion tensor magnetic resonance imaging of the human calf comparison between 1.5T and 3.0T- preliminary results. Invest Radiol. 2008; 43: 612–618.
35. Greenman RL, Panasyuk S, Wang X, et al. Early changes in the skin microcirculation and muscle metabolism of the diabetic foot. Lancet. 2005; 366: 1711–1717.
36. Hulsman NM, Geertzen JHB, Dijkstra PU, et al. Myopathy in CRPS-I: disuse or neurogenic. Eur J Pain. 2009; 13: 731–736.
37. Kos S, Klarhöfer M, Aschwanden M, et al. Simultaneous dynamic blood oxygen level-dependent magnetic resonance imaging of foot and calf muscles. Aging effects at ischemia and postocclusive hyperemia in healthy volunteers. Invest Radiol. 2009; 44: 741–747.
38. Dietrich O, Raya JG, Reeded SB, et al. Measurement of signal-to-noise ratios in MR images: influence of multichannel coils, parallel imaging, and reconstruction filters. J Magn Reson Imaging. 2007; 26: 375–385.
39. Jenkinson M, Bannister P, Brady JM, et al. Improved optimisation for the robust and accurate linear registration and motion correction of brain images. Neuroimage. 2002; 17: 825–841.
40. Jenkinson M, Beckmann CF, Behrens TE, et al. FSL. Neuroimage. 2012; 62: 782–790.
41. Cox RW. AFNI: software for analysis and visualization of functional magnetic resonance neuroimages. Comput Biomed Res. 1996; 29: 162–173.
42. Gupta RK, Hasan KM, Mishra AM, et al. High fractional anisotropy in brain abscesses versus other cystic intracranial lesions. AJNR Am J Neuroradiol. 2005; 26: 1107–1114.
43. Trivedi R, Malik GK, Gupta RK, et al. Increased anisotropy in neonatal meningitis: an indicator of meningeal inflammation. Neuroradiology. 2007; 49: 767–775.
44. Agarwal V, Kumar M, Singh JK, et al. Diffusion tensor anisotropy magnetic resonance imaging: a new tool to assess synovial inflammation. Rheumatology. 2009; 48: 378–382.
45. Vas LC, Pai R, Radhakrishnan M. Ultrasound appearance of forearm muscles in 18 patients with complex regional pain syndrome 1 of the upper extremity. Pain Pract. 2013; 13: 76–88.
46. Heerschap A, den Hollander JA, Reynen H, et al. Metabolic changes in reflex sympathetic dystrophy: a 31P NMR spectroscopy study. Muscle Nerve. 1993; 16: 367–373.
47. Huygen FJ, De Bruijn AG, De Bruin MT, et al. Evidence for local inflammation in complex regional pain syndrome type 1. Mediators Inflamm. 2002; 11: 47–51.
48. Krämer HH, Eberle T, Üçeyler N, et al. TNF-alpha in CRPS and “normal” trauma—significant differences between tissue and serum. Pain. 2011; 152: 285–290.
49. Gondin J, Callot V, Cozzone PJ, et al. Specific changes in water diffusivity due to passive shortening and lengthening of thigh muscles—a diffusion tensor imaging study. Proc Intl Soc Mag Reson Med. 2010; 18: 881
50. Damon BM. Effects of image noise in muscle diffusion tensor (DT)-MRI assessed using numerical simulations. Magn Reson Med. 2008; 60: 934–944.
magnetic resonance imaging; diffusion tensor imaging; skeletal muscle; normal foot; 3 T
Copyright © 2014 by Lippincott Williams & Wilkins
Highlight selected keywords in the article text.