System and method for analysis of the musculoskeletal system using MRI images

Abstract

A method for characterizing aspects of the musculoskeletal system of a subject includes accessing medical images of the subject, which include a plurality of voxels. The method includes segmenting the images to generate a 3D model that identifies at least two groups of the voxels that belong to at least two of a muscle, a tendon, a bone, scar tissue, edema, or fat. Generating a report that characterizes aspects of the musculoskeletal system, including at least two of a dimension of the muscle, a dimension of the tendon, a dimension of the bone, a dimension of the scar tissue, a dimension of the edema, a dimension of the fat, fat infiltration in the muscle, an extent of intra-muscular scar tissue, an extent of intra-muscular edema, an extent of free tendon, a surface area of a junction of the muscle and the tendon, or a density of the bone.

Description

SYSTEM AND METHOD FOR AN ALYSIS OF THE MUSCULOSKELETAL SYSTEM USING MRI IMAGESCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application is based on, claims priority to, and incorporates herein by reference in its entirety, US Provisional Application Serial No. 63 / 534,916, Filed August 28, 2023, and entitled “SYSTEM AND METHOD FOR ANALYSIS OF THE MUSCULOSKELETAL SYSTEM USING MRI IMAGES.”BACKGROUND

[0002] Image analysis of the musculoskeletal system is often performed manually, which is time consuming. Moreover, manual analysis leads to subjective results that vary based on who performs the analysis, the processing and visualization tools available, and imaging protocols used. For example, facioscapulohumeral muscular dystrophy (FSHD) is often evaluated by manually examining short tau inversion recover (STIR) images in which bright signal corresponds with inflammation. These images are ranked on a four-point scale where 0 indicates normal appearance, 1 indicates very mild diffuse increase, 2 indicates mild diffuse increase, 3 indicates moderate areas of increased signal intensity, and 4 indicates severe involvement of the entire muscle. This system only provides a qualitative measure of disease progression. Thus, a need exists for methods to assess various aspects of the musculoskeletal system in an automated and quantitative way. Such methods may advantageously improve standardization for inter- and intra-subject comparisons and increase efficiency of image analysis.SUMMARY OF THE DISCLOSURE

[0003] The foregoing and other aspects and advantages of the present disclosure will appear from the following description. In the description, reference is made to the accompanying drawings that form a part hereof, and in which there is shown by way of illustration a preferred embodiment. This embodiment does not necessarily represent the foil scope of the invention, however, and reference is therefore made to the claims and herein for interpreting the scope of the invention. Like reference numerals will be used to refer to like parts from Figure to Figure in the following description.

[0004] In some aspects, the present disclosure provides a method for characterizing fat infiltration in a muscle of a subject. The method includes accessing medical images of themuscle of the subject, which include a plurality of voxels, using a computer processor. The method further includes segmenting the medical images to generate a 3D model that identifies a group of the voxels belonging to a muscle and determining a fat content in each of the group of voxels. The method also includes generating a report that characterizes fat infiltration in the muscle based on the fat content in each of the voxels.

[0005] In other aspects, the present disclosure provides a method for characterizing aspects of the musculoskeletal system of a subject. The method includes accessing medical images of the subject, which include a plurality of voxels, using a computer processor. The method also includes segmenting the images to generate a 3D model that identifies at least two groups of the voxels that belong to at least two of a muscle, a tendon, a bone, scar tissue, edema, or fat. The method further includes generating a report that characterizes aspects of die musculoskeletal system, including at least two of a dimension of the muscle, a dimension of the tendon, a dimension of the bone, a dimension of the scar tissue, a dimension of the edema, a dimension of the fat, fat infiltration in the muscle, an extent of intra-muscular scar tissue, an extent of intra-muscular edema, an extent of free tendon, a surface area of a junction of the muscle and the tendon, or a density of the bone.

[0006] In other aspects, the present disclosure provides a method for characterizing a primary anatomical structure of the musculoskeletal system of a subject based on medical images. The method includes accessing medical images of the subject using a computer processor. The medical images include at least part of the primary anatomical structure and a secondary tissue. The method further includes segmenting the medical images to generate a 3D model of the primary anatomical structure, which identifies a first plurality of voxels belonging to the primary' anatomic al struc ture and a second plurality7of voxels belonging to the secondary tissue. The method further includes measuring dimensions of the primary7anatomical structure and of the secondary tissue based on the 3D model and generating a report including the dimensions of the primary anatomical structure and the secondary tissue.BRIEF DESCRIPTION OF THE DRAWINGS

[0007] FIG. 1A shows a flowchart setting forth the steps of an example process for automated image analysis.

[0008] FIG. IB shows a flowchart setting forth the steps of an example process for generating a model to analyze the musculoskeletal system.

[0009] FIG. 1C shows a flowchart setting forth the steps of an example process for generating a report for the musculoskeletal system.

[0010] FIG. 2 A shows an example segmentation, 3D segmentation model, and analysis, which may be generated in accordance with the present disclosure.

[0011] FIG. 2B shows an example report including muscle size asymmetries, which may be generated in accordance with the present disclosure.

[0012] FIG. 2C shows an example report including muscle-specific characterization of fat infiltration, which may be generated in accordance with the present disclosure.

[0013] FIG. 3A shows an example report including localized characterization of fat infiltration, which may be generated in accordance with the present disclosure.

[0014] FIG. 3B shows an example report including assessment of fat infiltration across the pectoralis major muscles, which may be generated in accordance with the present disclosure.

[0015] FIG. 3C shows an example report comparing fat infiltration across time points, which may be generated in accordance with the present disclosure.

[0016] FIG. 4A shows an example report including visualization of edema and muscles, which may be generated in accordance with the present disclosure.

[0017] FIG. 4B shows another example report including visualization of edema, which may be generated in accordance with the present disclosure.

[0018] FIG. 5A shows an example of a segmentation model that identifies edema, which may be generated in accordance with the present disclosure.

[0019] FIG. 5B show's an example report including visualization of edema, which may be generated in accordance with the present disclosure.

[0020] FIG. 5C show's an example results of automated edema analysis compared to manual edema ranking in accordance with some aspects of the present disclosure.

[0021] FIG. 6A shows an example segmentation model that identifies tendons, which may be generated in accordance with the present disclosure.

[0022] FIG. 6B show's an example report including 3D visualization of tendons and muscles, which may be generated in accordance with the present disclosure.

[0023] FIG. 6C shows an example report including 3D visualization of the interaction of tendons and muscles, w'hich may be generated in accordance with the present disclosure.

[0024] FIG. 6D show's another example report including 3D visualization of the interaction of tendons and muscles, w'hich may be generated in accordance with the present disclosure.

[0025] FIG. 6E show's example reports including 3D visualization of the Achillestendon, which may be generated in accordance with the present disclosure.

[0026] FIG. 6F shows an example report including assessment of the cross-sectional area of the Achilles tendon, which may be generated in accordance with the present disclosure.

[0027] FIG. 7 illustrates an example of processing which may be used in accordance with the present disclosure.

[0028] FIG. 8 shows an example report including visualization of scar tissue, which may be generated in accordance with the present disclosure.

[0029] FIG. 9 is a block diagram of an example magnetic resonance imaging (“MRI”) system that can implement the methods described in the present disclosure.

[0030] FIG. 10 is a block diagram of an example automated analysis system that can implement the methods of the present disclosure.

[0031] FIG. 11 is a block diagram of example components that can implement the system of FIG. 10.DETAILED DESCRIPTION

[0032] The present disclosure provides systems and methods for providing comprehensive analysis of the musculoskeletal system using magnetic resonance imaging (“MRI”) images. For example, the analysis may provide clinically useful information about muscles, bones, fat, tendons, ligaments, cartilage, connective tissue, intra- and extra-muscular edema, intra- and extra- muscular scarring, or other components of the musculoskeletal system. Advantageously, systems and methods are provided for segmentation of anatomical structures of interest, followed by quantitative analysis. For example, individual bones, muscles, tendons, ligaments, or other structures may be segmented. In this way, the analysis may be on a structure-specific level. For example, analysis can be provided for individual muscles, such as the right and left biceps, pectoralis major, erector spinae, multifidus, quadratus lumborum, psoas major, gluteus maximus, gluteus medius, and so on. Analysis may also be provided for individual tendons, such as the semimembranosus distal tendon, semimembranosus proximal tendon, biceps femoris semitendinosus joint tendon, biceps femoris distal tendon, semitendinosus distal tendon, Achilles tendon, patellar tendon, and so on, and individual bones, such as scapula, spine, and others. Additionally, the analysis may be generated for muscle groups as a whole, or as functional anatomical groups (e.g., the collective hamstring muscles and tendons).

[0033] The analyses generated provide quantitative metrics for individual structuresand tissue types. For example, these metrics may include volume, cross-sectional area, and surface area. The following non-limiting example analyses may be provided and will be described in further detail below, muscle size, total fat infiltration, localized fat infiltration, scar or edema volume and distribution of intra- and extra-muscular involvement, surface area of tendon adhering to muscle (e.g., myotendinous junction surface area), tendon size, cross- sectional area of a tendon, length of free tendon, center of boundary of muscle or tendon, muscle or contractile muscle volume and fat volume within each muscle, and extent of STIR content (a measure of inflammation). The analysis can also include comparative analysis between the left and right side of an individual (e.g., left-right asymmetry), various longitudinal timepoints for an individual (e.g., before and after injury, before and after rehabilitation or physical therapy, throughout aging), population norms (e.g., to assess risk of injury), or other individuals.

[0034] The present disclosure also provides a method for generating an analysis model. The analysis model can be produced by aggregating image analysis metrics and performance metrics from a group of subjects. The analysis model can be used to compare image analysis metrics of a given subject to population norms, predict functional or performance metrics, and quantify disease or injury severity based on MRI images.

[0035] FIGS. 1A-1C set forth three related example processes, which may be used independently or in conjunction to produce comprehensive analysis of the musculoskeletal system based on MRI image. Referring particularly to FIG. 1A, an example process 100 is provided for performing automated image analysis that can provide image-based metrics of the musculoskeletal system. The process 100 may be fully or partially automated and carried out by a computer system. The process may be used to characterize aspects of the musculoskeletal system of a subject. For example, the process can be used to characterize fat infiltration in a muscle.

[0036] Medical images are accessed in block 102. For example, the medical images may include MRI images, such as water-fat images (e.g., Dixon), Ti-weighted images, T2- weighted images, short tau inversion recovery (STIR) images, or images with other contrasts. Accessing the medical images can include acquiring images using an MRI system and may include reconstructing raw data (e.g., k-space data) or other image processing (e.g., artifact correction, motion correction, etc.).

[0037] The images are segmented in block 104. The segmentation can be automated and may be performed by a computer processor. For example, the segmentation may be performed using details as described in US 11,526,993 B2, US 10,314,511, and Appl. No.17 / 664,893, the entire contents of which are incorporated herein by reference.

[0038] Segmenting the MRI images can delineate anatomical structures or tissue types, such as individual or groups of muscles, bones, tendons, ligaments, scar tissue, edema, inflammation, fatty tissue, cartilage, or other structures. Segmentation may also delineate secondary tissues or tissue states, such as edema, inflammation, scar tissue, tendons associated with muscles, intra-muscular fat, or others. Such segmentation produces a 3D model of the anatomical structures and tissues. For example, the 3D model may include voxels that are labeled as part of one or more anatomical structures. The 3D model may also identify the boundary of one or more anatomical structures. In this way, the 3D model can identify one or more groups of voxels that belong to a type of structure or tissue, which may include at least a primary anatomical structure and a secondary tissue. An example segmentation is shown in FIG. 2A, in which the muscles and bones of a subject were segmented using water and fat MRI images over a 3D volume of the subject’s lower extremities.

[0039] Referring again to FIG. 1A, an analysis can be performed in block 106. The analysis can be performed using the medical images, the 3D segmentation model, or a combination thereof. The use of the 3D segmentation model can advantageously allow for the analysis to be performed with anatomical structure specificity. Analysis may include one of several analyses that will be described herein. As non-limiting examples, the analysis may include measuring a dimension or extent of an anatomical structure or tissue. Analysis may also include characterizing the tissue contents of an anatomical structure or characterizing bone density. Analysis may also include normalization. For example, the muscle size measurements may be normalized by the size of the subject. The analysis may additionally include comparisons of analyses over longitudinal time points for a given subject, between anatomical structures within a given subject (e.g., left hamstring vs. right hamstring), or between a given subject and a population norm.

[0040] The analysis of block 106 can be used to generate image analysis metrics in block 108. The metrics may include a dimension of an anatomical structure or tissue. For example, the dimension may include a volume, cross-sectional area, or surface area of an anatomical structure. The dimension may also include an extent of fat infiltration, edema, inflammation, or scar tissue within an anatomical structure. The metric may also include structure-wise or voxel-wise percentage of fat infiltration, bone density, ratio of muscle and fat content. The metric can also include a characterization of an inter- or intra-subject comparison, such as a size difference between two muscles or between a subject’s muscle and average muscle size in a particular population.

[0041] Referring to FIG. IB, a process 120 is illustrated, which can be used to generate an analysis model. The process 120 can also be used to establish several normative databases of musculoskeletal metrics from various subject populations. The process 120 may be frilly or partially automated and carried out by a computer system. The method may include aggregating image analysis metrics and performance metrics of one or more populations to allow for comparisons with population norms. The model can also be fit or trained to provide prediction of performance metrics or prediction of injury based on MRI images of a subject. The input data (e.g., image analysis metrics and performance metrics) may be collected over longitudinal time points, which may allow for temporal predictions for a given subject (e.g., risk of injury).

[0042] A subject of a given population may be identified in block 122. The population may include the general population or a specific population. For example, the population may be defined based on a clinical condition (e.g., FSHD patient) or received treatment or intervention (e.g., surgery vs. physical therapy, training vs. non-training, completing vs. foregoing rehabilitation, resting vs. continuing activity). The population may also be based on demographic data, such as sex, age (e.g., age 40-50). The population may also be based on clinical history (e.g., patients that have experienced injury to their anterior cruciate ligament, patients that have pulled a hamstring muscle). The population may also be based on a performance metric (e.g., able to lift 200 lbs.) or a lifestyle or profession (e.g., professional athlete). For example, the population may be based on a particular sport that the subject plays (e.g., basketball, football, swimming, and so on), what position they play (e.g., forward, defense, kicker), at what level they play (e.g., Olympic, professional, college, club, and so on). The population may also be based on a combination of subject characteristics. Identifying the subject may also include gathering demographic or clinical history data for the subject.

[0043] Metric data can be accessed for the subject in block 124. The metric data may include image analysis metrics. The image analysis metrics can be generated using process 100, for example. Image analysis metrics may also be accessed from memory by a computer. The image analysis metrics may include one or more measurements or characterizations generated based on MRI images, as described herein. The metric data may also include performance data. As non-limiting examples, the performance data may include maximum or average sprinting, pitching, or swimming speed; maximum throwing or jumping distance; maximum jumping height; maximum weight able to be lifted; scores on performance tests that measure agility, power, strength, acceleration, or speed; or other measures of functional or athletic performance. Metric data may also include demographic data, such as age, height, orweight. Metric data may also include clinical history data, such presence of a past or present injury, extent of injury, pain level, recovery times, or presence of medical condition.

[0044] Data collection can be repeated for a group of N subjects in block 128. Data can be aggregated by repeating block 122 and 124 for a group of subjects within the defined population. N may be chosen to be sufficient for a desired prediction or comparison. For example, N may be defined as N > 5, N > 10, N > 50, N > 100, N > 1,000, or N > 10,000. For comparison between two individuals, N = 2.

[0045] Once the data is aggregated, it can be used to generate an analysis model in block 130, which may include a predictive model or a comparative model. The model can relate image analysis metrics with performance metrics or injury metrics. For example, the model can establish a relationship between individual muscle, muscle group, or tendon volume with functional performance or injury metrics. In this way, the model can be used to predict injury based on volume of a muscle or group of muscles, for example. The model can also establish a relationship between a muscle-fat ratio with injury or performance metrics. The model may also relate longitudinal changes of muscle and fat metrics with performance or injury metrics. For example, the longitudinal growth of a muscle or muscle group may be associated with improved performance over time. As another example, the model may be used to establish a reduction in inflammation required for an injured athlete to safely return to play.

[0046] Generating the model may include performing cluster analysis. For example, hierarchical cluster analysis may be used to investigate patterns in muscle development and characteristics. Further details of such example are provided in Appendix B. As another nonlimiting example, cluster analysis can be used to relate muscle profiles with disease severity, treatment groups, age, athlete type, and other subject characteristics. Cluster analysis may also be used to relate image-based metrics with function or performance metrics, such as sprint speed or distance covered in a six-minute walk test.

[0047] Generating the model may also include fitting a regression model or training a machine learning model on the aggregated data. In this way, a model can be generated to identify positively or negatively correlated parameters, which can provide predictive parameters. For example, the model may identify a negative correlation with tendon cross- sectional area with presence of tendon injury, indicating that small tendon cross-sectional area is predictive of injury' risk. The model can also be used to predict future changes of the musculoskeletal system. For example, cunent measures or patterns of fat infiltration or bright STIR signal may be used to predict progression of fat infiltration or edema, respectively. As another example, the model may characterize correlations between muscle sizes within amuscle group. In this way, the model can be used to establish healthy or normal relationships between muscles within a given subject. As another example, the model may predict athlete strength, speed, jump ability, or other performance metrics based on muscle size or muscle composition (e.g., ratio of muscle to fat content).

[0048] The analysis model may also characterize population norms or distributions of various metrics. For example, the model can provide a distribution of a muscle size, such as a quadricep, for professional athletes or the general population. Such a comparative model can be used to compare image analysis metrics of a subject with that of a desired population. An example of such analysis is provided in Appendix C.

[0049] The process 140 can be used to generate an analysis report of the musculoskeletal system. The report may be specific to a given subject or may provide a characterization of the musculoskeletal system for a group of subjects in a population. The report may be based on MRI images by using automated image analysis metrics and may optionally be further based on an analysis model. The image analysis metrics can be accessed in block 142. Accessing the image analysis metrics may include performing process 100, as previously described, or accessing image analysis metrics from a computer memory. These metrics may optionally be used with an analysis model, which can be accessed and applied in block 144. For example, block 144 may include performing process 120 or accessing a trained analysis model from a computer memory. Block 144 may also include applying the analysis model to the image analysis metrics to produce a prediction or comparison.

[0050] A report may be generated in block 146, which may report image analysis metrics, provide 3D visualization of the 3D segmentation model, and provide predictive or comparative results from the analysis model. In this way, the report can characterize several aspects of the musculoskeletal system. For example, the report can include the structurespecific analysis, a visualization of the 3D segmentation model, 2D and 3D visualizations of the analysis, medical images, injury or performance predictions, and data comparisons. Several example reports are provided in FIGS. 2A-8. As non-limiting examples, the report may include evaluation of muscle size, fat infiltration in a muscle, an extent of the scar tissue within or around a muscle, an extent of the edema within or around a muscle, an extent of free tendon, a surface area of a junction of a muscle and a tendon (e.g., shared surface area of the muscle and the tendon), a density of a bone, a prediction of injury risk, a prediction of athletic performance, a comparison of muscle size with population nonns, tracking of image analysis metrics over time, or a combination thereof.

[0051] Referring now to FIG. 2 A, a report is provided for the lower extremities of asubject. Water and fat MRI images were acquired, which characterize the water and fat signal in each voxel, respectively. The images were automatically segmented as described above. The segmentation delineated individual muscles and bones within the lower extremities of the subject. The segmentations can be provided as boundaries drawn onto the 2D images or as a 3D segmentation model. Such 2D and 3D segmentation models may be interactive. For example, a user can scroll through the image slices, rotate the 3D model, toggle visualization of a subset of segmented structures, click on the segmentation boundary or 3D visualization to query the structure name and analyses, etc.

[0052] FIG. 2A also demonstrates two example analyses. The total muscle volume was measured for each individual muscle based on the 3D segmentation model. These muscle volumes can be visualized by color-coded muscles, where the color indicates the size of the muscle. For example, the size can be characterized relative to a population norm, relative to the standard deviation (SD) of muscle size distribution in a reference population (e.g., healthy subjects), relative to another timepoint, or as a continuous or histogram measure of total muscle volume. Similar comparisons can be reported for fat infiltration or other image analysis metrics. The muscle size may also be characterized based on the length of the muscle. For example, the length can be determined by identifying the centroid of the segmented muscle for each slice and connecting the centroids along the length of the muscle.

[0053] The fat content or fat infiltration may also be measured for each muscle. As a non-limiting example, the fat infiltration measure for a given muscle may be defined as the number of voxels characterized as fat compared to the total number of voxels within the boundary of the muscle. Thus, the fat infiltration can be presented as a percentage of fat content within each muscle. The fat infiltration may also be characterized on a voxel- wise basis, which will be described further below.

[0054] FIG. 2B shows an example of an asymmetry analysis of the lower extremities. The various analyses, such as muscle volume, fat infiltration, and others, may be compared between the corresponding structures on the right and left sides. As a non-limiting example, the muscle sizes are compared in the report shown in FIG. 2B. hi this case, the 3D visualization provides a color-coded characterization of each of the muscles’ sizes compared to their counterparts. In addition, the muscle sizes can be compared with population norms. To compare muscle sizes between subjects, they may be normalized by the subject’s overall size, which may be characterized based on their height, weight, or the size (e.g., length, volume) of a bone (e.g., the scapula as measured by the 3D segmentation model). Similarly, the fat infiltration can be compared or analyzed across the whole body of a subject. For example, thefat infiltration can be compared between muscle groups, as shown in FIG. 2C, for example. Such a report may identify specific muscle groups that a subject should focus on in strength training or physical therapy, for example.

[0055] FIG. 3A shows another example analysis and report, which provides a localized characterization of the fat infiltration in muscle. As a non-limiting example, the fat infiltration can be measured on a voxel- wise basis according to:where F(x,y, z) represents the fat infiltration of a given voxel at position (x, y, z~) .Mfat(x, y, z) represents the signal of the fat or fat-only MR. I image that characterizes a fat signal in a voxel at a position (x, y, z) and Mwater(x, y, z) represents the signal of the water or water-only MRI image that similarly characterizes a water signal. Such analysis can be performed over the whole 3D volume or for a volume of a muscle or other structure of interest.

[0056] The localized fat infiltration may be displayed using a 3D visualization, as shown for the right and left pectoralis major in FIG. 3A, for example. This visualization provides a pixel -by-pixel 3D view of the localized fat content and progression of fat infiltration. The fat infiltration or other image analysis metric can be visualized as a color-binned map (as shown), where each color represents a range of fat infiltration percentages, or as a continuous color-coded heatmap or colormap. For example, each pixel within each labeled muscle can be binned based on the fat infiltration. Such a 3D visualization may be used to help locate fat infiltration, monitor progression, and assess fat infiltration cluster shape.

[0057] FIG. 3B provides another example analysis in which the fat infiltration is analyzed over a cross-sectional area of the muscle. One or more cross-sectional areas can be defined along a length of the muscle. For example, cross-sectional areas may be defined as the boundary of the muscle, according to the 3D segmentation model, in a series of planes defined by the MRI acquisition slices. Alternatively, the planes may be defined based on the natural or anatomical length of the muscle (e.g., from the hip to the knee in the quadricep). For example, a length of the muscle can be defined as a line through a contiguous volume of the muscle. The cross-sectional areas can be defined as areas along the line that are perpendicular to the line. The cross-sectional areas may have any desired thickness and be defined by the boundary of the muscle or other structure according to the 3D segmentation model.

[0058] The fat infiltration can be analyzed as the average fat infiltration across the cross-sectional area of the muscle (top panel). It may further be analyzed as the sum of the fatinfiltration across the whole cross-sectional area of the muscle (bottom panel), separated by muscle and fat. This quantification provides a localized characterization of the extent of both muscle and fat along a muscle. Such quantification may be used to further interpret fat infiltration metrics. For example, the user can understand if an increase in fat infiltration at a given location is caused by an increase in fat content or a decrease in muscle content (e.g., muscle atrophy). This analysis may be particularly useful for pediatric patients who are growing or patients undergoing exercise regimens.

[0059] In addition, the muscle composition can be similarly analyzed. As a nonlimiting example, the muscle composition can be defined as:FIG. 3B shows an example plot of the fat infiltration and muscle composition summed over the cross-sectional area of the muscle from an inferior slice to a superior slice. In this way, the fat and muscle composition can be compared across the muscle and compared between the right and left sides. Fat infiltration can also be expressed as a ratio of muscle to intra-muscular or extra-muscular fat.

[0060] Fat content or infiltration can also be characterized based on a segmentation of fat (e.g., adipose or fatty tissue). For example, the segmentation can identify fatty tissue based on Ti-weighted images. The segmentation model can also include delineation of muscle. In this way, the extent of fatty tissue within the muscle boundary can be measured. For example, the total fat volume can be compared to total muscle volume or compared across cross-sectional areas, as previously described.

[0061] These analyses also can be compared across timepoints, as shown, for example, in FIG. 3C. For example, the fat infiltration can be compared before and after an injury, throughout a physical therapy regimen for injury recovery, or over the course of a workout plan for strength building. Fat infiltration can also be compared to population norms. To obtain population norms, the fat infiltration can be measured for a given muscle for a group of subjects. As non-limiting examples, the population may be defined as healthy subjects, athletes, patients with a predefined condition (e.g., Facioscapulohumeral muscular dystrophy), or another population of interest.

[0062] FIG. 4A shows another example report generated based on an analysis of injury extent. As a non-limiting example, the 3D segmentation can identify edema based on medical images, such as T2-weighted MRI images. The segmented edema can be visualized with the 3D muscle segmentation, as shown for a patient with a hamstring injury in FIG. 4A. The extentof injury can be quantified based on the volume of edema. The injury can be readily characterized based on muscle involvement based on the 3D segmentation model.

[0063] FIG. 4B shows another example report of a patient with an injury. The edema spans 38 cm of the hamstring. The site of pain was identified by the patient and marked the MRI images with a vitamin E tablet. The perceived location of pain was located 7.5 cm distal from the most superior occurrence of edema and superior to the start of the proximal myotendinous interface, measured using a visualization of the segmentation model. Further examples of assessing injury, including edema and scar tissue, are provided in Appendix A.

[0064] Referring now to FIGS. 5A-5C, an example analysis is shown for STIR images, which are often acquired for FSHD patients to characterize inflammation or edema. In standard practice, the STIR images are subjectively analyzed, as the bright signal corresponds to edema. Each muscle region is subjectively rated on a scale from zero to four based on visual inspection of the STIR images. As an alternative, as disclosed herein, the STIR images can be segmented using an artificial intelligence (Al) network. Such segmentation can be fully automated or semi-automated, accepting partial input for a user or clinician, which may be referred to as “vetted Al”. For example, the images may be initially segmented using an automated Al model. A radiologist or other user can view the segmentations to verify the automated result or modify the automated segmentation to create a “vetted Al” segmentation. The segmentation can identify bright signals in STIR images that correspond to inflammation or edema, allowing for a quantification of the inflammation. An example segmentation of STIR images of a subject’s calves is shown in FIG. 5A. The standard subjective ratings on a scale of 0-4 are shown for comparison.

[0065] By segmenting both the individual muscles and the edema, several quantitative analyses can be performed. For example, the overall volume of edema can be measured over a region (e.g., calf) of a subject. The edema extent can also be measured on a muscle- or another structure-specific basis. For example, the volume of edema within each muscle can be measured and compared to the overall muscle volume. FIG. 5B provides an example of a visualization of edema within a muscle. The extent of the edema was measured as a percentage of the muscle based on the volume of edema within the muscle boundary. This extent was compared in 40 lower extremity muscles for 31 subjects, as plotted in FIG. 5C. The extent of the edema as measured by the automated segmentation is highly correlated with the subjective rank as evaluated by a clinician.

[0066] Additionally, the extent of edema can be assessed based on its relation to another structure. For example, the edema can be characterized by an intra-structure extent andan extra-structure extent. This may be determined based on the boundaries or overlapping volumes of the structures or tissues. In this way, the edema can be characterized as intramuscular and extra-muscular. The edema can be reported as a raw volume or as a percentage of its associated muscle, both of which provide a quantitative metric to evaluate muscle injury severity^.

[0067] The segmentation may also identify individual tendons, as shown, for example, in FIGS. 6A-6F. The segmentation of muscles and tendons may be performed using a single set of medical images. Alternatively, the segmentation of muscles can be performed using one set of medical images while the segmentation of tendons can be performed using another set of medical images, and the 3D tendon and muscle segmentation models can be co-registered to create a combined 3D segmentation model.

[0068] The 3D segmentation model can be used to measure the size of individual tendons, such as the cross-sectional area, total volume, or surface area. The model can also be used to assess the interaction of tendons and muscles. For example, the shared surface area of a muscle and tendon can be measured to evaluate the junction between a tendon and muscle.

[0069] As a non-limiting example, the hamstring muscles (e.g., semimembranosus, semitendinosus, and biceps femoris) and the tendons associated with the hamstring muscles can be segmented, as shown in FIG. 6A. Segmentation can also be performed to identify the Achilles tendon patellar tendon, and others.

[0070] The segmentation can produce a 3D segmentation model, as shown, for example, in FIG. 6B. The 3D segmentation model can be used to perform a variety of analyses. Analyses can include measuring tendon and muscle volumes, the cross-sectional areas of the tendons and muscles, comparison of the cross-sectional areas of the tendons with the cross- sectional areas of the muscles, measuring the surface areas of the tendons and muscles, measuring the surface areas of the muscle-tendon interfaces, and the like. The cross-sectional area and surface area can also be measured in a localized way, such as along the length of the tendon or muscle. The shared surface area between the adjacent structures or tissues can be measured. For example, the surface area of muscle-tendon interfaces can be measured to evaluate the junction between a muscle and tendon. Such analyses and measures can be compared between the right and left side of a subject or with population norms. For example, the cross-sectional area of a tendon across its length may be predictive of injury' risk.

[0071] Segmenting the muscles and tendons allows for separate characterization of the free tendon and internal tendon. For example, FIG. 6C illustrates a visualization of the proximal free tendon, the distal free tendon, the proximal internal tendon (or also called “aponeurosis”),and the distal internal tendon, all associated with the biceps femoris long head muscle. These 3D segmentation models can be used to measure tendon length and the area of attachment between the muscle and the internal tendons, which may be associated with muscle injury risk. The length of the muscle-tendon unit can also be evaluated based on the location of the centroid path of the muscle and tendon, as shown in FIG . 6D. For example, a centroid of the muscle cross-sectional area can be identified for each slice through the length of the muscle based on the 3D model of the muscle boundary. This centroid path can be used to aid in the creation of musculoskeletal models. The centroid path may also be combined with measurements of running mechanics to estimate stretch of the muscle-tendon unit during movements like running.

[0072] Another non-limiting example of tendon segmentation is shown in FIGS. 6E- 6F. In this example, the Achilles tendon was segmented in 10 patients, as shown in FIG. 6E. The cross-sectional area (CSA) was compared across all 10 patients along the length of the tendon. For example, the CSA can be represented by a heatmap and plotted along the length of the tendon.

[0073] The segmentation may include a padding step, as illustrated in FIG. 7. This may be especially advantageous for small structures whose dimensions are near that of the imaging resolution. For example, the finite image resolution may create holes in the 3D visualization of the tendon, as shown. Such holes can be reduced by padding diagonally adjacent pixels. For example, the padding may include automatically identifying diagonally adjacent pixels in each plane and increasing the boundary of the segmented structure by one pixel in each direction diagonally. As another example, the padding may include filling in gaps between the two structures based on a desired threshold. For example, any gap between a tendon and a muscle of two pixels or fewer may be filled in as tendon.

[0074] Referring now to FIG. 8, another example report is provided, which provides visualization of scar tissue. The scar tissue can be segmented based on the medical images. In this way, the 3D segmentation model can provide a model of the muscles as well as the scar tissue. The model can allow for assessment of the scar tissue and muscle involvement. For example, the scar tissue can be visualized with an interactive 3D visualization, as shown in FIG. 8. The extent (e.g., volume and shape) of intra-muscular and extra-muscular scar tissue can also be measured and quantified. In some implementations, the 3D model of the scar tissue may be registered to the 3D model of the individual muscles. In this way, the scar tissue can be quantified based on the effected muscles. Further examples are provided is Appendix A.

[0075] Additional example reports and analyses are provided in Appendices A-C.Appendix A provides examples of injury analysis, including segmentation and quantification of scar tissue and edema. Such analysis may include co-registering separate 3D models (e.g., as muscle segmentation, tendon segmentation, edema segmentation, scar tissue segmentation, and others) to quantify muscle or tendon involvement. Appendix B provides an example of cluster analysis where the population was defined as 10 Division 1 NCAA basketball athletes and performance metrics included jump height and sprint velocity. Appendix C provides an example of more detailed population analyses. These examples provide characterization of individual athletes with respect to a normative and elite population database and evaluates positional differences between players.

[0076] Referring particularly now to Fig. 9, an example of an MRI sy stem 900 that can implement the methods described herein is illustrated. The MRI system 900 includes an operator workstation 902 that may include a display 904, one or more input devices 906 (e.g., a keyboard, a mouse), and a processor 908. The processor 908 may include a commercially available programmable machine running a commercially available operating system. The operator workstation 902 provides an operator interface that facilitates entering scan parameters into the MRI system 900. The operator workstation 902 may be coupled to different servers, including, for example, a pulse sequence server 910, a data acquisition server 912, a data processing server 914, and a data store server 916. The operator workstation 902 and the servers 910, 912, 914, and 916 may be connected via a communication system 940, which may include wired or wireless network connections.

[0077] The MRI system 900 also includes a magnet assembly 924 that includes a polarizing magnet 926, which may be a low-field magnet. The MRI system 900 may optionally include a whole-body RF coil 928 and a gradient system 918 that controls a gradient coil assembly 922.

[0078] The pulse sequence server 910 functions in response to instructions provided by the operator workstation 902 to operate a gradient system 918 and a radiofrequency (“RF”) system 920. Gradient waveforms for performing a prescribed scan are produced and applied to the gradient system 918, which then excited gradient coils in an assembly 922 to produce the magnetic field gradients (e.g., Gx. Gy, and Gz) that can be used for spatially encoding magnetic resonance signals. The gradient coil assembly 922 forms part of a magnet assembly 924 that includes a polarizing magnet 926 and a whole-body RF coil 928.

[0079] RF waveforms are applied by the RF system 920 to the RF coil 928, or a separate local coil to perform the prescribed magnetic resonance pulse sequence. Responsive magneticresonance signals detected by the RF coil 928, or a separate local coil, are received by the RF system 920. The responsive magnetic resonance signals may be amplified, demodulated, filtered, and digitized under direction of commands produced by the pulse sequence server 910. The RF system 920 includes an RF transmitter for producing a wide variety' of RF pulses used in MRI pulse sequences. The RF transmitter is responsive to the prescribed scan and direction from the pulse sequence server 910 to produce RF pulses of the desired frequency, phase, and pulse amplitude waveform. The generated RF pulses may be applied to the whole-body RF coil 928 or to one or more local coils or coil arrays.

[0080] The RF system 920 also includes one or more RF receiver channels. An RF receiver channel includes an RF preamplifier that amplifies the magnetic resonance signal received by the coil 928 to which it is connected, and a detector that detects and digitizes the / and Q quadrature components of the received magnetic resonance signal. The magnitude of the received magnetic resonance signal may, therefore, be determined at a sampled point by the square root of the sum of the squares of the / and Q components:

[0081] and the phase of the received magnetic resonance signal may also be determined according to the following relationship:

[0082] The pulse sequence server 910 may receive patient data from a physiological acquisition controller 930. By way of example, the physiological acquisition controller 930 may receive signals from a number of different sensors connected to the patient, including electrocardiograph (“ECG”) signals from electrodes, or respiratory signals from a respiratory bellows or other respiratory monitoring devices. These signals may be used by the pulse sequence server 910 to synchronize, or “gate,” the performance of the scan with the subject’s heartbeat or respiration.

[0083] The pulse sequence server 910 may also connect to a scan room interface circuit932 that receives signals from various sensors associated with the condition of the patient and the magnet system. Through the scan room interface circuit 932, a patient positioning system 934 can receive commands to move the patient to desired positions during the scan.

[0084] The digitized magnetic resonance signal samples produced by the RF system 920 are received by the data acquisition server 912. The data acquisition server 912 operates in response to instructions downloaded from the operator workstation 902 to receive the realtime magnetic resonance data and provide buffer storage, so that data are not lost by dataoverrun. In some scans, the data acquisition server 912 passes the acquired magnetic resonance data to the data processor server 914. In scans that require information derived from acquired magnetic resonance data to control the further performance of the scan, the data acquisition server 912 may be programmed to produce such information and convey it to the pulse sequence server 910. For example, during pre-scans, magnetic resonance data may be acquired and used to calibrate the pulse sequence performed by tire pulse sequence server 910. As another example, navigator signals may be acquired and used to adjust the operating parameters of the RF system 920 or the gradient system 918, or to control the view order in which k-space is sampled. In still another example, the data acquisition server 912 may also process magnetic resonance signals used to detect the arrival of a contrast agent in a magnetic resonance angiography (“MRA”) scan. For example, the data acquisition server 912 may acquire magnetic resonance data and process it in real-time to produce information that is used to control the scan.

[0085] The data processing server 914 receives magnetic resonance data from the data acquisition server 912 and processes the magnetic resonance data in accordance with instructions provided by the operator workstation 902. Such processing may include, for example, reconstructing two-dimensional or three-dimensional images by performing a Fourier transformation of raw k-space data, performing other image reconstruction algorithms (e.g., iterative or backpr ejection reconstruction algorithms), applying filters to raw k-space data or to reconstructed images, generating functional magnetic resonance images, or calculating motion or flow images.

[0086] Images reconstructed by the data processing server 914 are conveyed back to the operator workstation 902 for storage. Real-time images may be stored in a data base memory cache, from which they may be output to operator display 902 or a display 936. Batch mode images or selected real time images may be stored in a host database on disc storage 938. When such images have been reconstructed and transferred to storage, the data processing serv er 914 may notify the data store server 916 on the operator workstation 902. The operator workstation 902 may be used by an operator to archive the images, produce films, or send the images via a network to other facilities.

[0087] The MRI system 900 may also include one or more networked workstations 942. For example, a networked workstation 942 may include a display 944, one or more input devices 946 (e.g., a keyboard, a mouse), and a processor 948. The networked workstation 942 may be located within the same facility as the operator workstation 902, or in a different facility, such as a different healthcare institution or clinic.

[0088] The networked workstation 942 may gain remote access to the data processing server 914 or data store server 916 via the communication system 940. Accordingly, multiple networked workstations 942 may have access to the data processing server 914 and the data store server 916. In this manner, magnetic resonance data, reconstructed images, or other data may be exchanged between the data processing server 914 or the data store server 916 and the networked workstations 942, such that the data or images may be remotely processed by a networked workstation 942.

[0089] Referring now to FIG. 10, an example of an MRI system 1000 is shown, which may be used in accordance with some aspects of the systems and methods described in the present disclosure. As shown in FIG. 10, a computing device 1050 can receive one or more types of data (e.g., signal evolution data, k-space data, receiver coil sensitivity data) from data source 1002. In some configurations, computing device 1050 can execute at least a portion of an automated analysis system 1004 to reconstruct images from magnetic resonance data (e.g., k-space data). In some configurations, the automated analysis system 1004 can implement an automated pipeline to provide MRI images, water images, fat images, 3D visualization, reports, etc.

[0090] Additionally or alternatively, in some configurations, the computing device 1050 can communicate information about data received from the data source 1002 to a server 1052 over a communication network 1054, which can execute at least a portion of the automated analysis system 1004. In such configurations, the server 1052 can return information to the computing device 1050 (and / or any other suitable computing device) indicative of an output of the automated analysis system 1004.

[0091] In some configurations, computing device 1050 and / or server 1052 can be any suitable computing device or combination of devices, such as a desktop computer, a laptop computer, a smartphone, a tablet computer, a wearable computer, a server computer, a virtual machine being executed by a physical computing device, and so on. The computing device 1050 and / or server 1052 can also reconstruct images from the data.

[0092] In some configurations, data source 1002 can be any suitable source of data (e.g., measurement data, images reconstructed from measurement data, processed image data), such as an MRI system, another computing device (e.g., a server storing measurement data, images reconstructed from measurement data, processed image data), and so on. hi some configurations, data source 1002 can be local to computing device 1050. For example, data source 1002 can be incorporated with computing device 1050 (e.g., computing device 1050 can be configured as part of a device for measuring, recording, estimating, acquiring, orotherwise collecting or storing data). As another example, data source 1002 can be connected to computing device 1050 by a cable, a direct wireless link, and so on. Additionally or alternatively, in some configurations, data source 1002 can be located locally and / or remotely from computing device 1050, and can communicate data to computing device 1050 (and / or server 1052) via a communication network (e.g., communication network 1054).

[0093] In some configurations, communication network 1054 can be any suitable communication network or combination of communication networks. For example, communication network 1054 can include a Wi-Fi network (which can include one or more wireless routers, one or more switches, etc.), a peer-to-peer network (e.g., a Bluetooth network), a cellular network (e.g., a 3G network, a 4G network, etc., complying with any suitable standard, such as CDMA, GSM, LTE, LTE Advanced, WiMAX, etc.), other types of wireless network, a wired network, and so on. In some configurations, communication network 1054 can be a local area network, a wide area network, a public network (e.g., the Internet), a private or semi-private network (e.g., a corporate or university intranet), any other suitable type of network, or any suitable combination of networks. Communications links shown in FIG. 10 can each be any suitable communications link or combination of communications links, such as wired links, fiber optic links, Wi-Fi links, Bluetooth links, cellular links, and so on.

[0094] Referring now to FIG. 11, an example of hardware 1100 that can be used to implement data source 1002, computing device 1050, and server 1052 in accordance with some configurations of the systems and methods described in the present disclosure is shown.

[0095] As shown in FIG. 11, in some configurations, computing device 1050 can include a processor 1102, a display 1104, one or more inputs 1106, one or more communication systems 1108, and / or memory 1110. In some configurations, processor 1102 can be any suitable hardware processor or combination of processors, such as a central processing unit (“CPU”), a graphics processing unit (“GPU”), and so on. In some configurations, display 1104 can include any suitable display devices, such as a liquid crystal display (“LCD”) screen, a light-emitting diode (“LED”) display, an organic LED (“OLED”) display, an electrophoretic display (e.g., an “e-ink” display), a computer monitor, a touchscreen, a television, and so on. In some configurations, inputs 1106 can include any suitable input devices and / or sensors that can be used to receive user input, such as a keyboard, a mouse, a touchscreen, a microphone, and so on.

[0096] In some configurations, communications systems 1108 can include any suitable hardware, firmware, and / or software for communicating information over communication network 1054 and / or any other suitable communication networks. For example,communications systems 1108 can include one or more transceivers, one or more communication chips and / or chip sets, and so on. In a more particular example, communications systems 1108 can include hardware, firmware, and / or software that can be used to establish a Wi-Fi connection, a Bluetooth connection, a cellular connection, an Ethernet connection, and so on.

[0097] In some configurations, memory 1110 can include any suitable storage device or devices that can be used to store instructions, values, data, or the like, that can be used, for example, by processor 1102 to present content using display 1104, to communicate with server 1052 via communications system(s) 1108, and so on. Memory 1110 can include any suitable volatile memory, non-volatile memory, storage, or any suitable combination thereof. For example, memory 1110 can include random-access memory (“RAM”), read-only memory (“ROM”), electrically programmable ROM (“EPROM”), electrically erasable ROM (“EEPROM”), other forms of volatile memory, other forms of non-volatile memory, one or more forms of semi-volatile memory, one or more flash drives, one or more hard disks, one or more solid state drives, one or more optical drives, and so on. In some configurations, memory 1110 can have encoded thereon, or otherwise stored therein, a computer program for controlling operation of computing device 1050. In such configurations, processor 1102 can execute at least a portion of the computer program to present content (e.g., images, user interfaces, graphics, tables), receive content from server 1052, transmit information to server 1052, and so on. For example, the processor 1102 and the memory 1110 can be configured to perform the methods described herein.

[0098] In some configurations, server 1052 can include a processor 1112, a display 1114, one or more inputs 1116, one or more communications systems 1118, and / or memory 1120. In some configurations, processor 1112 can be any suitable hardware processor or combination of processors, such as a CPU, a GPU, and so on. In some configurations, display 1114 can include any suitable display devices, such as an LCD screen, LED display, OLED display, electrophoretic display, a computer monitor, a touchscreen, a television, and so on. In some configurations, inputs 1116 can include any suitable input devices and / or sensors that can be used to receive user input, such as a keyboard, a mouse, a touchscreen, a microphone, and so on.

[0099] In some configurations, communications systems 1118 can include any suitable hardware, firmware, and / or software for communicating information over communication network 1054 and / or any other suitable communication networks. For example, communications systems 1118 can include one or more transceivers, one or morecommunication chips and / or chip sets, and so on. In a more particular example, communications systems 1118 can include hardware, firmware, and / or software that can be used to establish a Wi-Fi connection, a Bluetooth connection, a cellular connection, an Ethernet connection, and so on.

[0100] In some configurations, memory 1120 can include any suitable storage device or devices that can be used to store instructions, values, data, or the like, that can be used, for example, by processor 1112 to present content using display 1114, to communicate with one or more computing devices 1050, and so on. Memory 1120 can include any suitable volatile memory, non-volatile memory, storage, or any suitable combination thereof. For example, memory 1120 can include RAM, ROM, EPROM, EEPROM, other types of volatile memory, other types of non-volatile memory, one or more types of semi-volatile memory, one or more flash drives, one or more hard disks, one or more solid state drives, one or more optical drives, and so on. In some configurations, memory 1120 can have encoded thereon a server program for controlling operation of server 1052. In such configurations, processor 1112 can execute at least a portion of the server program to transmit information and / or content (e.g., data, images, a user interface) to one or more computing devices 1050, receive information and / or content from one or more computing devices 1050, receive instructions from one or more devices (e.g., a personal computer, a laptop computer, a tablet computer, a smartphone), and so on.

[0101] In some configurations, the server 1052 is configured to perform the methods described in the present disclosure. For example, the processor 1112 and memory 1120 can be configured to perform the methods described herein.

[0102] In some configurations, data source 1002 can include a processor 1122, one or more data acquisition systems 1124, one or more communications systems 1126, and / or memory 1128. In some configurations, processor 1122 can be any suitable hardware processor or combination of processors, such as a CPU, a GPU, and so on. In some configurations, the one or more data acquisition systems 1124 are generally configured to acquire data, images, or both, and can include an MRI system. Additionally or alternatively, in some configurations, the one or more data acquisition systems 1124 can include any suitable hardware, firmware, and / or software for coupling to and / or controlling operations of an MRI system. In some configurations, one or more portions of the data acquisition system(s) 1124 can be removable and / or replaceable.

[0103] Note that, although not shown, data source 1002 can include any suitable inputs and / or outputs. For example, data source 1002 can include input devices and / or sensors that can be used to receive user input, such as a keyboard, a mouse, a touchscreen, a microphone, atrackpad, a trackball, and so on. As another example, data source 1002 can include any suitable display devices, such as an LCD screen, an LED display, an OLED display, an electrophoretic display, a computer monitor, a touchscreen, a television, etc., one or more speakers, and so on.

[0104] In some configurations, communications systems 1126 can include any suitable hardware, firmware, and / or software for communicating information to computing device 1050 (and, in some configurations, over communication network 1054 and / or any other suitable communication networks). For example, communications systems 1126 can include one or more transceivers, one or more communication chips and / or chip sets, and so on. In a more particular example, communications systems 1126 can include hardware, firmware, and / or software that can be used to establish a wired connection using any suitable port and / or communication standard (e.g., VGA, DVI video, USB, RS-232, etc.), Wi-Fi connection, a Bluetooth connection, a cellular connection, an Ethernet connection, and so on.

[0105] In some configurations, memory 1128 can include any suitable storage device or devices that can be used to store instructions, values, data, or the like, that can be used, for example, by processor 1122 to control the one or more data acquisition systems 1124, and / or receive data from the one or more data acquisition systems 1124; to generate images from data; present content (e.g., data, images, a user interface) using a display; communicate with one or more computing devices 1050; and so on. Memory 1128 can include any suitable volatile memory, non-volatile memory, storage, or any suitable combination thereof. For example, memory 1128 can include RAM, ROM, EPROM, EEPROM, other types of volatile memory, other types of non-volatile memory, one or more types of semi-volatile memory, one or more flash drives, one or more hard disks, one or more solid state drives, one or more optical drives, and so on. In some configurations, memory 1128 can have encoded thereon, or otherwise stored therein, a program for controlling operation of medical image data source 1002. In such configurations, processor 1122 can execute at least a portion of the program to generate images, transmit information and / or content (e.g., data, images, a user interface) to one or more computing devices 1050, receive information and / or content from one or more computing devices 1050, receive instructions from one or more devices (e.g., a personal computer, a laptop computer, a tablet computer, a smartphone, etc.), and so on.

[0106] In some configurations, any suitable computer-readable media can be used for storing instructions for performing the functions and / or processes described herein. For example, in some configurations, computer-readable media can be transitory or non-transitory. For example, non-transitory computer-readable media can include media such as magnetic media (e.g., hard disks, floppy disks), optical media (e.g., compact discs, digital video discs,Blu-ray discs), semiconductor media (e.g., RAM, flash memory, EPROM, EEPROM), any suitable media that is not fleeting or devoid of any semblance of permanence during transmission, and / or any suitable tangible media. As another example, transitory computer- readable media can include signals on networks, in wires, conductors, optical fibers, circuits, or any suitable media that is fleeting and devoid of any semblance of permanence during transmission, and / or any suitable intangible media.

[0107] As used herein in the context of computer implementation, unless otherwise specified or limited, the terms "component," "system," "module,” "controller," "framework," and the like are intended to encompass part or all of computer-related systems that include hardware, software, a combination of hardware and software, or software in execution. For example, a component may be, but is not limited to being, a processor device, a process being executed (or executable) by a processor device, an object, an executable, a thread of execution, a computer program, or a computer. By way of illustration, both an application running on a computer and the computer can be a component. One or more components (or system, module, and so on) may reside within a process or thread of execution, may be localized on one computer, may be distributed between two or more computers or other processor devices, or may be included within another component (or system, module, and so on).

[0108] In some implementations, devices or systems disclosed herein can be utilized or installed using methods embodying aspects of the disclosure. Correspondingly, description herein of particular features, capabilities, or intended purposes of a device or system is generally intended to inherently include disclosure of a method of using such features for the intended purposes, a method of implementing such capabilities, and a method of installing disclosed (or otherwise known) components to support these purposes or capabilities. Similarly, unless otherwise indicated or limited, discussion herein of any method of manufacturing or using a particular device or system, including installing the device or system, is intended to inherently include disclosure, as embodiments of the disclosure, of the utilized features and implemented capabilities of such device or system.

[0109] As used herein, the phrase "at least one of A, B, and C" means at least one of A, at least one of B, and / or at least one of C, or any one of A, B, or C or combination of A, B, or C. A, B, and C are elements of a list, and A, B, and C may be anything contained in the Specification. The present disclosure has described one or more preferred embodiments, and it should be appreciated that many equivalents, alternatives, variations, and modifications, aside from those expressly stated, are possible and within the scope of the invention.

Claims

CLAIMS1. A method for characterizing fat infiltration in a muscle of a subject, the method comprising: accessing medical images of the muscle of the subject using a computer processor, the medical images comprising a plurality of voxels; segmenting the medical images using the computer processor to generate a 3D model that identifies a group of the plurality of voxels belonging to a muscle; determining a fat content in each of the group of voxels; and generating a report characterizing fat infiltration in the muscle based on the fat content in each of the group of voxels.

2. The method of claim 1, wherein the 3D model further identifies a group of voxels belonging to a secondary structure, wherein the secondary structure comprises at least one of scar tissue, edema, inflamed tissue, or tendon, and wherein the report further characterizes a dimension of the secondary structure.

3. The method of claim 1 , wherein the fat content in each of the group of voxels comprises a percentage of fat content in each of the group of voxels.

4. The method of claim 3, wherein the medical images comprise MRI images, the MRI images comprising water images that characterize a water signal in the plurality of voxels and fat images that characterize a fat signal in the plurality of voxels and determining the fat content in each of tire group of voxels comprises dividing the fat signal by a sum of the fat signal and the water signal in the group of voxels.

5. The method of claim 1, wherein the report comprises a measure of fat content in a cross-sectional area of the muscle.

6. The method of claim 5, wherein the report further comprises a measure of muscle content in the cross-sectional area of the muscle.

7. The method of claim I , wherein the report comprises a measure of fat content in a plurality of cross-sectional areas across a length of the muscle.

8. The method of claim 7, wherein the measure of fat content in the report is linked to relative location along the muscle length to quantify fat content distribution.

9. The method of claim 8, further comprising quantifying the fat content distribution into discrete categories based on location and pattern of highest fat content along the muscle.

10. The method of claim 8, wherein the measure of fat content and relative location along the muscle length is further compared across at least more than one timepoint of the subject and muscle, and further comprising quantifying fat infiltration progression using changes in fat content distribution across timepoints.

11. The method of claim 8, wherein the length of the muscle and cross-sectional areas of the muscle are provided in the report.

12. The method of claim 7, wherein the report further comprises a measure of muscle content in the plurality' of cross-sectional areas across the length of the muscle.

13. The method of claim 12, further comprising defining the length of the muscle as a line through a contiguous volume of the muscle, and wherein each of the plurality of cross-sectional areas is perpendicular to the line.

14. The method of claim 1 , wherein the report includes at least one of variation, mean, skewness, or kurtosis of the fat content across each of the group of voxels.

15. The method of claim 1, wherein a relative location of each voxel belonging to the group of voxels is utilized in conjunction with the fat content to calculate Moran’s index to generate a value indicating a level of clustering of fat.

16. The method of claim 1, wherein the report comprises a 3D visualization of fat infiltration, wherein the visualization comprises the group of pixels color-coded based on the percentage of fat content.

17. The method of claim 1 , wherein medical images comprise MRI images, the MRI images comprising Ti -weighted images, and determining the fat content in each of the group of voxels comprises segmenting the Ti -weighted images to label at least muscle and fat in the Ti-weighted images.

18. The method of claim 1 , wherein the medical images comprise short tau inversion recovery (STIR) images, and wherein the 3D model further identifies edema based on a bright signal in the STIR images.

19. A method for characterizing aspects of a musculoskeletal system of a subject, the method comprising: accessing medical images of the subject using a computer processor, the medical images comprising a plurality of voxels: segmenting the images using the computer processor to generate a 3D model that identifies at least two groups of the plurality of voxels belonging to at least two of a muscle, a tendon, a bone, scar tissue, edema, or fat: and generating a report characterizing aspects of the musculoskeletal system, wherein the aspects comprise at least two of a dimension of the muscle, a dimension of the tendon, a dimension of the bone, a dimension of the scar tissue, a dimension of the edema, a dimension of the fat, fat infiltration in the muscle, an extent of intra-muscular scar tissue, an extent of intra-muscular edema, an extent of free tendon, a surface area of a junction of the muscle and the tendon, or a density of the bone.

20. The method of claim 19, wherein the medical images comprise MRI images, the MRI images comprising water images that characterize a water signal in the plurality of voxels and fat images that characterize a fat signal in the plurality of voxels; and the method further comprises determining a fat content in the group of voxels by dividing the fat signal by a sum of the fat signal and the water signal in the group of voxels.

21. The method of claim 19, wherein the report further comprises a measure of fat content in a cross-sectional area of the muscle and a measure of muscle content in the cross- sectional area of the muscle.

22. The method of claim 19, wherein the report further comprises a measure of fat content in a plurality of cross-sectional areas across a length of the muscle and a measure of muscle content in the plurality of cross-sectional areas across the length of the muscle.

23. A method for characterizing a primary anatomical structure of a musculoskeletal system of a subject based on medical images of the subject, the method comprising: accessing, using a computer processor, medical images of the subject, wherein the medical images include at least part of the primary anatomical structure and a secondary' tissue; segmenting the medical images using the computer processor to generate a 3D model of the primary anatomical structure, wherein the 3D model identifies a first plurality of voxels belonging to the primary anatomical structure and a second plurality of voxels belonging to the secondary tissue; measuring a dimension of the primary anatomical structure and a dimension of the secondary tissue based on the 3D model; and generating a report comprising the dimension of the primary anatomical structure and the dimension of the secondary tissue.

24. The method of claim 23, wherein the primary anatomical structure is at least one of a muscle, a tendon, or a bone.

25. The method of claim 24, wherein the secondary tissue is at least one of fat, scar tissue, edema, or tendon.

26. The method of claim 25, wherein the report further comprises aspects of the musculoskeletal system, wherein the aspects comprise at least two of fat infiltration in the muscle, an extent of intra-muscular scar tissue, an extent of intra-muscular edema, an extent of free tendon, a surface area of a junction of the muscle and the tendon, or a density of the bone.

27. The method of claim 23, wherein the dimension of the primary anatomical structure is at least one of a 3D volume, a cross-sectional area, or a surface area of theprimary anatomical structure, and the dimension of the secondary tissue is at least one of a 3D volume, a cross-sectional area, or a surface area of the secondary' tissue.

28. The method of claim 23, wherein the method farther comprises measuring a shared surface area of the primary anatomical structure and the secondary tissue; and the report further comprises a measure of shared surface area of the primary anatomical structure and the secondary tissue.

29. The method of claim 23, wherein segmenting the medical images further comprises identifying a boundary of the primary' anatomical structure, and wherein the method farther comprises measuring an intra-structure dimension of the secondary tissue based on the 3D model of the primary anatomical structure, wherein the intra-structure dimension is measured based on the second plurality of voxels belonging to the secondary tissue within the boundary of the primary anatomical structure.

Images