Quantitative MRI MSK Techniques: An Update

Abstract

For many years now, MRI of the musculoskeletal system has relied mostly on conventional sequences with qualitative analysis. More recently, utilizing quantitative MRI applications to complement qualitative imaging has gained increasing interest in the MRI community, providing more detailed physiological or anatomical information. In this article, we review the current state of quantitative MRI, technical and software advances, along with quantitative MRI’s most relevant clinical and research musculoskeletal applications.

Introduction

MRI has been shown to be a very helpful tool in the diagnosis of many musculoskeletal (MSK) disorders and has established itself as a very reliable modality for noninvasive evaluation of the MSK system, thereby becoming an indispensable clinical diagnostic tool [1, 2]. Most clinical MRI requests are for evaluation of connective tissue pathology such as meniscal tears, rotator cuff tears, ligaments, and tendons pathology. MSK MRI protocols have been tailored to accommodate this scenario.

For many years, MRI of the MSK system has relied mostly on conventional sequences with qualitative analysis. The rate of MRI studies and MSK applications keeps expanding [3], and lately, quantitative MRI (QMRI) applications are gaining increasing interest in the MRI community. QMRI can complement qualitative imaging with more detailed physiological or anatomical information, providing measures that could aid in earlier disease detection, comparative studies, and monitor treatment [4]. However, despite recent advances, technical challenges and specific recommendations for QMRI must be taken into consideration.

In this article, we review what is going on regarding QMRI, technical and software advances, along with QMRI’s most relevant clinical and research MSK applications. In particular, we focus on quantitative techniques that are currently available or are among the most promising for clinical use.

T2 Mapping

Different MRI parametric mapping techniques are available, such as T2, T2* (T2-star), and T1 mapping, but the most widely available and frequently obtained is T2-mapping. This technique typically requires acquisition of multiple images with different echo times that will yield signal intensities which follow a T2 relaxation curve, such as with the Single Echo Spin Echo (SESE), Multi Echo Spin Echo (MESE), and Double Echo Steady State (DESS) sequences [5, 6]. When implementing T2 mapping in clinical practice, considerations should be given to acquisition, post-processing, and interpretation.

Acquisition of T2 mapping techniques can be lengthy, and patient motion during the sequence can result in inaccurate mapping. In addition, a technical challenge that can be seen with some parametric mapping techniques, such as MESE T2 mapping, is sensitivity to B₁ inhomogeneity [7]. B₁ inhomogeneity is particularly prominent at high field strengths, such as 3T, and can lead to inaccuracies in quantification. Different T2-mapping methods have been proposed to overcome this limitation, including the use of a T2-preparation pulse (T2prep) to assure B1 insensitivity, followed by either a gradient echo read-out or 3D turbo spin echo (3D TSE) read-out [8]. The T2prep consists of a spin-echo–like preparation pulse to transmit the T2-weighted contrast and a readout design for fast image data gathering. T2prep techniques have faster acquisition times as compared to the MESE technique, but acquire less echoes and, therefore, provide less data points along the T2 decay curve, with possible compromise in T2 values accuracy [9].

Postprocessing of the multiple images generated by the T2 mapping sequences can be performed online on the scanner or offline using algorithms written in separate programs, such as MATLAB (The Mathworks Inc, Natick, MA). Automated processing on the scanner typically generates a pixel-by-pixel map of T2 relaxation times, with a scalebar based on the range of values in the field of view. With offline processing, additional steps can be performed, including image registration, segmentation of structures, and selection of regions of interest to generate fitting curves, which can be used to assess the quality of the data (Figure 1). Interpretation of T2 values and visual maps requires many considerations. First, care must be taken when comparing absolute T2 relaxation times, since values may differ depending on anatomic location [10], with the highest values typically seen at the magic angle (Figure 1) [11]. In addition, values can vary depending on the scanner and sequence used for acquisition. However, abrupt changes or irregularities, which may be highlighted on T2 maps, can be considered abnormal. Despite the challenges, T2 mapping remains useful in clinical practice and has been validated for non-invasive quantitative analysis of tissue composition and structure. T2-mapping can be easily implemented on most clinical MRI systems and has the advantage of not requiring contrast agent administration [12].

Fig. 1–

(A) 43-year-old asymptomatic man and (B) 67-year-old woman with knee osteoarthritis. Image from the asymptomatic volunteer (A) shows the highest T2 values in the posterior weight bearing aspect of the medial femoral condyle, which is at the magic angle. In the patient with osteoarthritis of the knee (B), determined by clinical examination and radiographs, there are irregular areas of high T2 values at the femoral trochlea, posterior weight-bearing aspect of the medial femoral condyle, and the medial tibial plateau, indicating areas of abnormal cartilage. Sequence: 2D multi echo spin echo T2 measurement (TE=6.1, 12.2, 18.3, 24.4, 30.4, 36.5, 42.6, 48.7, 54.8, 60.9, 67.0, 73.1 ms).

T2 maps can be used for quantitative evaluation of nearly any musculoskeletal tissue, but the most frequent use is for the assessment of articular hyaline cartilage. For cartilage, density and organization of the extracellular matrix appear as variations of T2 values that can be represented by either a gray-scale or color map [13]. T2-mapping has been shown to be effective in detecting and quantifying early changes related to water content and collagen concentration, even before detectable structural changes [14, 15]. It can be applied to identify early-stage degeneration and cartilage with irreversible damage [16, 17], to assess functional potential, to study reparative tissue, or to monitor the effects of chondroprotective therapy [18, 19].

T2 maps can also be used for evaluation of muscle composition, especially for quantification of edema and inflammatory changes [20]. A challenge with muscle T2-mapping is the impact of fatty degeneration on T2 values, since both edema/inflammatory changes and fatty degeneration lead to increased T2-values [21, 22]. These effects should be considered when evaluating patients with inflammatory myopathies and neuromuscular disorders, since patients can show concomitant findings of muscle edema and fatty infiltration [23]. Interpretation of T2-mapping in these scenarios requires careful analysis due to the multiple factors impacting T2 values [24, 25]. To reduce the effect of fat content on T2 maps, the implementation of fat suppression techniques is an alternative [26], despite recognized limitations for the complete removal of the fat content effect on T2 values [27]. The use of both fat-saturated and non-fat-saturated acquisitions can help analyze the specific findings in muscle T2 maps [26]. Fat infiltration in obese patients and muscle edema related to exercise are also confounding factors that can contribute to elevated T2 and need to be taken into consideration when evaluating muscle T2 maps [25, 28]. Overall, T2 mapping can be very useful in clinical practice, but the radiologist should be aware of the potential challenges since standardized T2 measurement protocols are absent, which make meta-analyses and multi-site comparison difficult.

T1ρ Mapping

T1ρ, also referred to as T1rho or “spin lock” relaxation, is another technique used to evaluate biochemical changes in tissues. T1ρ is the time constant of spin-lattice relaxation in the rotating frame, characterized by magnetic relaxation of spins under the influence of a radiofrequency (RF) pulse. It is sensitive for low-frequency interactions between macromolecules and bulk water [29, 30]. T1ρ is similar to T2 relaxation, except that there is an additional radiofrequency pulse (the “spin-locking” pulse) applied immediately after the magnetization is tipped into the transverse plane. Conventionally, the spin-locking pulse is a continuous wave RF pulse with long duration and low energy. Since the magnetization and RF field are along the same direction, this effectively locks the magnetization vector into the transverse plane without phase decay (as with T2 decay). The signal decay is exponential with a time constant, T1ρ, and is typically calculated from multiple images by changing the duration of the spin-locking pulse. Changing the amplitude of the spin locking pulse can also select for different properties within the tissue [5, 29, 31]. Since the conventional continuous wave spin-lock pulse is susceptible to field inhomogeneities, recent studies have evaluated techniques using adiabatic T1ρ spin-lock pulses, where amplitude and frequency are varied in time, showing promising results in osteoarthritis, with reduction of the effects of magnetic field inhomogeneities and also reduced sensitivity to the magic angle effect [32, 33].

Several studies have shown that T1ρ imaging is sensitive to detecting changes in PG content of articular cartilage, as found in the early stages of osteoarthritis [29, 34, 35]. T1ρ has also been shown to be reliable in mapping cartilage damage in patients with rheumatoid arthritis and osteoarthritis [36]. T1ρ depicts changes between protons and the macromolecular environment of cartilage. Considering that the motion of water molecules in articular cartilage is restricted by the macromolecules in the extracellular matrix, alterations such as PG loss can therefore be reflected in T1ρ values [5, 34, 37]. In comparison with other techniques, T1ρ mapping has the advantage of providing noninvasive analysis of PG content in cartilage without the need for contrast agent administration or any extra hardware (Figure 2) [35]. However, it requires a pulse sequence that is not widely available. Other limitations of T1ρ mapping are related to variability of results between different pulse sequences [38], angular and layer dependence of T1ρ values [39]. T1ρ measurements can also be confounded by the presence of multiple tissue components [40].

Fig. 2–

70-year-old asymptomatic woman. (A) T1ρ measurement obtained with a conventional technique with longer echo time (TE=10 ms) can be used to evaluate cartilage. (B and C) Using an ultrashort echo time (TE=0.032 ms), additional information is obtained from the short T2 components. High signal intensity at the posterior portion of the MFC suggests loss of proteoglycan (arrow). In addition, using the UTE technique, T1ρ measurements to be obtained from tissues with short mean T2 values, such as the menisci (arrowheads). Sequence: 2D Spiral T1ρ measurement (TSL = 0, 20,40,80 ms); 3D UTE-Cones T1ρ measurement (TSL = 0, 5,10, 20 ms).

dGEMRIC

Delayed gadolinium-enhanced MRI of cartilage (dGEMRIC) provides indirect measurement of cartilage structural composition and is based on the inverse relation between cartilage glycosaminoglycan (GAG) content and cartilage distribution of the negatively charged gadolinium contrast agent [41]. GAG molecules play an important role for cartilage integrity by helping keep water molecules within cartilage and, therefore, maintaining swelling pressure and strength. Decreased GAG molecules in the cartilage is one of the early steps of osteoarthritis development.

GAG molecules and gadolinium are both negatively charged, and gadolinium will accumulate in articular cartilage in an inversely proportional manner to the GAG concentration: in areas with decreased GAG concentration (i.e., lesser negative charge), more negatively charged gadolinium will penetrate the cartilage. Images are generally obtained after 90–120 min following intravenous injection of contrast agent to allow its diffusion within the cartilage. dGEMRIC is based on a T1 relaxation-time measurement technique that uses the negative ionic charge of gadolinium to generate a color-coded map of the charge density of cartilage GAGs [42]. Accumulation of the contrast agent in areas of low GAG content in the cartilage will result in shorter T1 relaxation time.

dGEMRIC maps can reliably detect GAG loss even before conventional MRI can detect cartilage changes. Different studies have shown that dGEMRIC is a valuable non-invasive tool in the evaluation of cartilage changes, like is seen in early stage osteoarthritis and focal cartilage defects [41, 43, 44], femoroacetabular joint disorders [45], follow up of surgically treated cartilage [46], and injured joints after dislocations and ligament tears [47, 48]. In osteoarthritis research studies, dGEMRIC has being used as a standard tool for the assessment of articular cartilage GAG [41, 49].

Practical limitations of this technique are the prolonged acquisition protocol due to the long delay between the contrast administration and MR acquisition, lack of standard physical activity protocol required before image acquisition, and the increased costs and risks associated with intravenous administration of contrast agent [42, 49].

Proton MR Spectroscopy

In vivo proton MR spectroscopy (MRS) offers non-invasive molecular characterization of different tissues, sampling the relative levels of metabolites in specific regions of interest [50]. MRS has been successfully used to study the intracellular contents of skeletal muscle tissue, monitoring metabolites like choline, lipids, total creatine, and trimethyl ammonium (Figure 3) [51]. However, compared to other body regions, like the brain, MRS has had rather restricted applications in the MSK system.

Fig. 3–

28-year-old asymptomatic man. (A) T1-weighted MRI showing voxel location used for MRS in one subject, and (B) the corresponding water suppressed spectrum from that region. Lipids dominate the signal in this region with distinct peaks from CH₂ groups on extra- and intra-myocellular lipids (red and blue arrows, respectively). A distinct peak from trimethylamine is visible (orange arrow), as are the peaks from the CH₃ and CH₂ groups of creatine (green and yellow arrows, respectively).

Multiple studies have used MRS for characterization of MSK tumors. By quantitatively assessing choline content, differentiation between benign and malignant lesions is possible [52, 53]. Another application of MRS is to quantify skeletal muscle lipid storage, evaluating both intramyocellular and extramyocellular lipid compartments [54]. Muscular lipid metabolism measurements offer an ample spectrum of clinical and research applications for MRS: correlations to exercise physiology, muscular function, insulin resistance, obesity, and other metabolic disorders [55–59]. Muscle metabolites other than lipids, such as creatine, have also been evaluated, providing a potential avenue to improve understanding of muscle metabolism and to monitor response to therapies [60].

Specific MRS techniques, like phosphorus-31 MRS, require specialized MRI hardware which limits its clinical adequacy. Proton MRS, on the other hand, does not involve special hardware and can therefore be performed as part of routine MRI. However, MRS measurements are affected by many imaging-related factors. One described limitation of muscle MRS is related to the reliance on ratios between water and fat or metabolite content and to the assumption of consistent water content within muscle, which can lead to inaccurate measurements in patients that present both muscle edema/inflammation and fatty degeneration, such as patients with dystrophinopathies [61].

Chemical-Shift

MRI allows quantification of fatty infiltration using several approaches, including chemical shift water-fat MRI, which relies on the difference in chemical shift between water and fat, enabling assessment and quantification of the fatty elements in muscle tissue using fat fraction maps (Figure 4) [62, 63]. Chemical shift imaging (CSI), proposed by Dixon in 1984, uses the phenomenon of signal intensity alterations detected in MRI that result from the inherent differences in the resonant frequencies of lipid and water [64]. These differences can be encoded into images, producing sets of images based on water and fat. Fat and water signals can then be used to calculate the fat fraction, expressed as the fraction of fat signal in the total signal in each voxel [65].

Fig. 4–

28-year-old asymptomatic man. Chemical shift imaging with fat (A) and water (B) fraction maps, which allows quantification of skeletal muscle. Low fat and high water fractions in the musculature of the leg in this volunteer are normal.

CSI methods have been applied mostly to quantify fat replacement of skeletal muscle. Fat fractions obtained by CSI have been shown to correlate well with histology and MRS [66]. Two- and three-point Dixon imaging have demonstrated good correlations with fat levels based on muscle biopsy and clinical severity in dystrophinopathies [67, 68]. It has also been applied to provide outcome measures and to detect disease progression [69].

In comparison with signal intensity approaches, CSI methods are less influenced by B₀ and B₁ magnetic field inhomogeneities and are less biased by partial volume effects [66]. However, it has been suggested that low signal-to-noise images affect the accuracy of the results, especially when one of the components is predominant, namely, low fat fraction or low water fraction [70]. Another challenge to CSI is related to the definition of regions of interest and delineation of muscle contours. Manual segmentation is very time consuming and recent advances on semi-automatic or automatic segmentation are still difficult to apply in the clinical setting. Faster and simpler automated or semi-automated muscle segmentation would be highly beneficial for the analysis of quantitative muscle MRI [71].

Diffusion Weighted Imaging

Diffusion-weighted imaging (DWI) is an MRI technique capable of measuring differences in the magnitude of diffusion of water molecules within a tissue [72]. DWI is an established technique in neuroradiology, but has not been widely used for MSK imaging, although promising applications have been studied [73]. DWI generates MRI images based on the contrast derived from the diffusion property of water molecules, allowing mapping of the diffusion process that will reflect the difference in rate of diffusion in tissues. Diffusion, also known as Brownian motion, denotes the random thermal movement of molecules. Diffusion of water molecules follows a pattern according to each tissue composition and structure, and some pathological conditions can alter this diffusion, allowing abnormalities to be detected by DWI [74, 75]. DWI can be generated by applying two extra diffusion gradients, equal in magnitude, on conventional MR sequences—one dephasing and one, exactly opposite, rephasing gradient. The first gradient introduces phase shift to the molecules, while the second gradient will cancel these changes. With diffusion of protons, the second gradient is not able to completely reverse the changes induced by the first gradient on moving spins, and signal attenuation can be detected. The detected signal loss is related to the resultant spin dephasing, varying according to time between pulses, strength, and duration of gradients applied. Apparent diffusion coefficient (ADC) maps can then be obtained, derived from at least two diffusion-weighted images, displaying the spatial distribution of the different diffusion rates. ADC maps also reduce T1 and T2 contrast in the images and allow diffusion quantification [74, 75].

For muscle MRI, while conventional sequences are sensitive to detect larger abnormalities, edema and hemorrhage, DWI can help detect minor lesions and fatigue-induced muscle disorders that would otherwise remain undetected, improving diagnosis [76]. DWI can also be combined with the diffusion tensor imaging (DTI) technique. Considering that the diffusion of molecules in structured tissue is anisotropic, DTI parameters can be used to measure anisotropy, to allow noninvasive evaluation of tissue microstructure and mechanical properties, and to map the orientation of skeletal muscle fibers [73, 77]. DTI is able to measure the magnitude and direction of mobility of molecules in a particular voxel, and can become a useful tool for investigation of muscle disorders. DTI measurements allow calculation of parameters for overall diffusivity and for the degree of anisotropy (fractional and relative anisotropy) [78]. Certain conditions, like mechanical injury and exercise-related trauma, can lead to disorganization of muscle fibers and altered diffusion, with consequent decreased anisotropy, that can sometimes be detected even before it becomes apparent on conventional images, helping early diagnosis and definition of the lesion extension [76, 79]. DTI can be applied to study skeletal muscle physiology, anatomy, and pathology, and could become diagnostically relevant for prognosis and treatment of sports-related muscle injury [76, 80]. However, DWI and DTI are dependent on many acquisition parameters and are prone to artifacts from misregistration of data, to motion artifacts, and to susceptibility variants. Muscle DTI parameters are also highly sensitive to age, gender, body mass index, exercise status, and temperature [80]. Further research is necessary, and improvements in the technique and in postprocessing analysis are needed to increase the application in both research and clinical medicine [73].

UTE/ZTE

Tendons, ligaments, menisci and bones contain a high fraction of components with “short” and “ultrashort” transverse relaxation times and, therefore, have short mean transverse relaxation times [81]. These tissues yield little or no signal with conventional MRI pulse sequences, and thus are not able to be properly characterized using these sequences with longer echo times. Tendons, ligaments, and menisci have T2s in the range of 2–8 ms and cortical bone and the deep layers of articular cartilage have T2s of about 0.2–2 ms [81]. To detect and explore signals from these very short T2 tissues, which are especially relevant in the MSK system, different ultrashort echo-time (UTE) and zero echo-time (ZTE) sequences have been designed. They are being increasingly improved and studied, thus providing the opportunity to visualize and detect abnormalities of these tissues in a manner not previously possible (Figure 2) [81–85].

UTE - T2*

UTE-T2* mapping, similar to quantitative conventional T2*, is based on a series of multiple images at different echo times (TEs), including TE of 0.5 ms or shorter [86]. T2* is the most popular relaxation constant used to detect potential collagen matrix alteration in tendons [87]. The number of components and T2* values vary according to spatial resolution and tendon orientation [88]. Biexponential T2* analysis has been successfully performed in vivo using both UTE and variable TE (vTE) sequences, and fractions and T2* values vary depending on tendon location, consistent with different tendon compositions [81]. Recent studies have shown that biexponential T2* offers robust measurements in both healthy individuals and patients with Achilles and patellar tendinopathy [83, 89, 90]. UTE-T2* can become a reliable marker to guide clinical outcome, detecting tendinopathy. Bi-component analysis can also be useful in quantifying the injured or postoperative tendon (Fig. 4). T2* limitations are related to magnetic field inhomogeneities and magic angle effects.

UTE/ZTE techniques can obtain signal from the short T2* components of articular cartilage, allowing direct visualization of the deep layers and discrimination of the calcified layer, which may be related to the pathogenesis of cartilage degeneration (Figure 5) [91, 92]. Conventional MRI sequences cannot differentiate the deep radial and calcified layers from subchondral bone [93]. Williams et al. [86], using a mono-exponential decay model, found that UTE-T2* values were more sensitive to matrix degeneration compared with conventional T2 values based on histological standards, showing lower UTE-T2* values with severely degraded cartilage. In a recent study with patients 2 years after anterior cruciate ligament reconstruction, UTE-T2* assessment could identify deep articular pathology in subclinical disease that was not evident on conventional MRI [94]. Chu et al. showed that UTE-T2* values were significantly elevated in ACL-reconstructed patients with arthroscopically normal articular cartilage and menisci [95]. Shao et al. recently demonstrated that UTE bi-component analysis can characterize the short and long T2* values and fractions across the cartilage depth, including the deep radial and calcified cartilage [96].

Fig. 5–

54-year-old male cadaveric specimen with osteoarthritis. UTE imaging highlighting the deep layers of articular cartilage of the medial femorotibial compartment (A) and at the femoral trochlea (B). Abnormal areas, represented by the missing bright lines, can be identified on the weight-bearing medial femoral condyle (thick arrow), medial tibial plateau (thin arrow), and the patella (arrowhead). Subtle areas (red lines) can also be quantified (C).

UTE T2* can also be used to assess bone. For cortical bone, recent studies have demonstrated that bi-exponential T2* fitting and adiabatic inversion recovery-UTE techniques can reliably measure bound water and pore water components in vitro and in vivo [84, 97, 98]. Future studies can confirm the potential of these techniques to assess bone quality and strength and to determine its implications for clinical evaluation of bone diseases. Challenges related to UTE-T2* are the lack of availability on most existing scanners, additional cost and extra clinical exam times, since it will need to be performed in addition to clinical sequences. For clinical imaging, limitations related to imperfect registration and patient motion may affect measurements [99]. Another source of potential limitations can be related to issues in the processing of exponential fitting of multi-echo images [40].

UTE - Magnetization Transfer

For certain anisotropic tissues such as tendons and cortical bone, low mean transverse relaxation time is not the only challenge for imaging. An additional concern is the “magic angle effect,” which is related to unaveraged dipolar interactions of proton nuclear spins [100]. Magnetization transfer (MT) refers to the interactions of protons residing in different macromolecular environments and transfer of longitudinal magnetization from the bound proton pool to the free proton pool. When combined with the UTE sequence, MT can be performed on short T2 tissues [101]. UTE-MT may provide unique information that cannot be directly obtained by other methods, such as regular UTE techniques, and multiple parameters can be obtained, including water and macromolecular proton fractions, as well as relaxation and exchange rates [102]. UTE‐MT with two‐pool modeling measurements has demonstrated much less orientational dependence and has shown potential as a clinically compatible quantitative technique that is resistant to the magic angle effect [103]. However, UTE-MT protocols with relatively small saturation pulses may lead to inaccuracy in the measurement of T2 of the water pool, related to underestimation of the exchange rate between macromolecular and water pools [103]. In addition, the two-pool model does not account for the presence of fat, which would confound the measurements. Fat-saturation methods or a three-pool model could be used in future studies.

Recent studies using the UTE-MT technique on rotator cuff tendons have shown promising results that are much less sensitive to magic angle effects compared with transverse relaxation times (Figure 6) [103, 104]. For cortical bone, studies by Chang et al. [105] and Ma et al. [106] demonstrated encouraging results investigating 2D UTE-MT, that can provide useful quantification information for cortical bone (Figure 7).

Fig. 6–

(A and B) 25-year-old man with intermittent shoulder pain and (C and D) 33-year-old asymptomatic woman. Coronal oblique UTE magnetization transfer imaging of the supraspinatus tendon with macromolecular maps show lower macromolecular fraction in the symptomatic patient (A and B) than in the asymptomatic volunteer (C and D), suggesting tendinopathy. Fitting curves from two-pool MT modeling shows excellent fit with mean macromolecular fraction (f) of 12.7% for the symptomatic patient (C) and 13.8% for the asymptomatic one (D). This technique enables acquisition of macromolecular fractions that are much less sensitive to magic angle effects.

Fig. 7–

35-year-old asymptomatic man. UTE magnetization transfer imaging of the left leg, providing color mapping with quantification information of the tibial cortical bone: macromolecular fraction (A), proton exchange rate from the macromolecular to water pools (B), and spin‐lattice relaxation rate of the water pool (R1w) (C). Color bars indicate the gradation of MT measures and regional variations can be seen in the different portions of the tibial cortex, reflecting compositional and structural differences. UTE-Cones MT modeling (flip angle=500, 1000, 1500°, frequency offset=2, 5, 10, 20, 50 kHz). UTE-Cones T1 measurement for MT modeling (actual flip angle imaging: flip angle=45° and TR=20/100 ms; variable TR: FA=45° and TR=20, 50, 150 ms).

Conclusion

MRI is continuously being refined and, nowadays, an ample range of quantitative sequences and techniques are available, enabling the visualization of previously “invisible” structures and characterization of multiple MSK tissues. However, many of the newer QMRI techniques are not widely available in clinical packages or cannot be performed in clinically feasible scan times, and specific recommendations are not adopted due to the lack of standardization and validation, especially across different equipment and systems. Further refinements are desired in order to facilitate and speed up its adoption, making it easier to compare its results over time, between subjects and with different equipment.