Systems and methods for estimating muscle strength
Patent Information
- Application Number
- PCT/US2024/024026
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-08-08
- Filing Date
- 2024-04-11
- Publication Date
- 2024-11-21
AI Technical Summary
Current methods for estimating muscle strength, such as direct determination and submaximal approaches, are burdensome, risky, and inaccurate, particularly for individuals with disabilities or elderly users, as they require incremental loading, potential injury, and high variability in concentric velocity measurements.
A system and method using concentric velocity measurements with a trained one-repetition maximum weight estimation statistical model to estimate muscle strength, updating weights based on obtained concentric velocities, and employing sensors like inertial measurement units to track velocities during exercises, allowing for accurate and safe estimation without altering the user's exercise plan.
The method provides reduced variability and higher accuracy in estimating muscle strength, reducing the risk of injury and burden on users, while allowing for continuous tracking without additional effort or equipment, effectively updating weight estimates over time.
Smart Images

Figure US2024024026_21112024_PF_FP_ABST
Abstract
Description
SYSTEMS AND METHODS FOR ESTIMATING MUSCLE STRENGTH RELATED APPLICATIONS
[0001] This application claims the benefit under 35 U.S.C. § 119(e) to U.S. Provisional Application Serial No.63 / 518,233, entitled “SYSTEMS AND METHODS FOR ESTIMATING MUSCLE STRENGTH”, filed on August 8, 2023, and claims the benefit under 35 U.S.C. § 119(e) to U.S. Provisional Application Serial No.63 / 459,233, entitled “METHOD OF ESTIMATING MUSCLE STRENGTH WITH WEARABLE SENSORS”, filed on April 13, 2023, each of which is herein incorporated by reference in its entirety. FIELD
[0002] Disclosed embodiments are related to systems and methods for estimating muscle strength. STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT
[0003] This invention was made with government support under 2129012 awarded by National Science Foundation (NSF). The government has certain rights in this invention. BACKGROUND
[0004] In exercise, a one repetition maximum (1RM) is the heaviest load of an exercise that a user can successfully complete for only one repetition. A 1RM may be useful for estimating the muscle strength of a user, which may inform what loads to use for the corresponding exercise. This may help to avoid selecting loads which are excessively heavy and pose safety risks to the user, as well as helping track muscle strength over time. SUMMARY
[0005] In some embodiments, a method of estimating muscle strength comprises obtaining a plurality of concentric velocities associated with a plurality of sets of an exercise performed with a weight, wherein the plurality of concentric velocities include at least a first concentric velocity associated with a first set of the exercise and a second concentric velocity associated with a second set of the exercise. The method further comprises providing one or more parameters associated with the plurality of sets to a trained one-repetition maximum weight estimation statistical model, wherein the one or more parameters are based at least in part H0776.70159WO00 12339990.1on the first concentric velocity and the second concentric velocity, and outputting a one- repetition maximum weight from the trained one-repetition maximum weight estimation statistical model.
[0006] In further embodiments, a method of estimating muscle strength comprises obtaining an estimated one-repetition maximum weight of a user for a target exercise and commanding a user to perform the target exercise with a first weight and a second weight different from the first weight, wherein the first weight and the second weight are determined based at least in part on the estimated one-repetition maximum weight. The method further comprises obtaining a first concentric velocity associated with the user performing the target exercise with the first weight, obtaining a second concentric velocity associated with the user performing the target exercise with the second weight, and updating the estimated one-repetition maximum weight based at least in part on the first concentric velocity and the second concentric velocity.
[0007] In some embodiments, a method of training a weight estimation statistical model comprises obtaining training data, wherein the training data includes concentric velocities associated with performing sets of an exercise with weights, and associated one-repetition maximum weights of one or more subjects performing the exercise with the weights. The method further comprises generating a trained weight estimation statistical model using the training data and storing the trained weight estimation statistical model on non-transitory computer readable memory for subsequent use.
[0008] It should be appreciated that the foregoing concepts, and additional concepts discussed below, may be arranged in any suitable combination, as the present disclosure is not limited in this respect. Further, other advantages and novel features of the present disclosure will become apparent from the following detailed description of various non-limiting embodiments when considered in conjunction with the accompanying figures.
[0009] In cases where the present specification and a document incorporated by reference include conflicting and / or inconsistent disclosure, the present specification shall control. If two or more documents incorporated by reference include conflicting and / or inconsistent disclosure with respect to each other, then the document having the later effective date shall control. BRIEF DESCRIPTION OF DRAWINGS
[0010] The accompanying drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component that is illustrated in various figures may H0776.70159WO00 12339990.1be represented by a like numeral. For purposes of clarity, not every component may be labeled in every drawing. In the drawings:
[0011] Fig.1 shows a strength estimation system according to some embodiments;
[0012] Fig.2 shows a schematic flow diagram of a first method for estimating muscle strength according to some embodiments;
[0013] Fig.3 shows a graphical representation of a method for estimating muscle strength according to some embodiments;
[0014] Fig.4 shows a schematic flow diagram of a second method for estimating muscle strength according to some embodiments;
[0015] Fig.5 shows a schematic flow diagram of a method for training a weight estimation statistical model according to some embodiments;
[0016] Fig.6 shows a graph including data relating to velocity during an exercise according to some embodiments; and
[0017] Fig.7 shows a schematic flow diagram of an example method for estimating muscle strength according to some embodiments. DETAILED DESCRIPTION
[0018] Determining a 1RM may be useful for helping a user to select appropriate loads and / or resistances associated with an exercise to help avoid injury, determine an exercise plan to increase muscle strength, and / or to track muscle strength over time for a subject. A commonly used approach to determining a one repetition maximum (1RM) for an exercise is direct determination (DM), which involves a user attempting incrementally increasing loads until the user cannot successfully complete one repetition of the exercise. Because the user must fail in order to determine their 1RM using DM, there is a high risk of injury to the user. The user may also require assistance using DM, which is burdensome and potentially costly. The user may need to travel somewhere (e.g., a clinic having exercise equipment) to perform DM, adding additional cost and burden to the user. Lastly, DM may be time-intensive due to the plurality of sets of the exercise required.
[0019] As an alternative to the above, a user performing an exercise with a submaximal load (e.g., less than the 1RM load) for as many consecutive repetitions as possible until failure may be used to estimate the 1RM. This approach may be referred to as the submaximal approach. In this approach, the 1RM is determined as a function of the repetitions successfully completed and the load used. This approach also requires a user to perform the exercise until failure which poses a risk of injury. The exercise may require assistance and also may require a H0776.70159WO00 12339990.1user to travel somewhere to complete the exercise, both of which may be costly and / or burdensome. In some instances, the user may be elderly and / or may have a disease or disorder (e.g., sarcopenia, or neurological disease) which may put the user at a higher risk of injury, necessitating traveling to a clinic having exercise equipment and professional assistance in order to safely complete the exercise. Accordingly, the user having a disease or disorder may be especially burdened by these existing approaches to estimating a 1RM.
[0020] In some cases, 1RM may be estimated using concentric velocities sensed while the exercise is performed by the user. As the load of an exercise increases, so does the difficulty of completing one repetition of the exercise and as such the user typically will perform the exercise with a lower concentric velocity. Generally, there is an approximately linear relationship between the load and concentric velocity. As used herein, "mean concentric velocity” refers to an average velocity at which a load is lifted against the gravity or other direction of resistance applied to a user’s body during the concentric phase of an exercise, and “concentric” refers to the phase of muscle contraction. Any concentric velocity as used herein may be any appropriate concentric velocity used with any one of the various 1RM estimation methods and system disclosed herein. For example, a concentric velocity of a repetition of an exercise used with any of the embodiments disclosed herein may be a mean concentric velocity of a repetition. In still other implementations, a maximum (e.g., peak) concentric velocity may be used with any of the embodiments disclosed herein. In still another embodiment, a median concentric velocity may be used with any of the embodiments disclosed herein. Of course it should be understood that any other appropriate statistic associated with the concentric velocities measured during a set of exercise may be used as the disclosure is not so limited. Additionally, a maximum, mean, median, or other appropriate characterization of the concentric velocities measured during a set of exercise may be used as the concentric velocity for a particular set in any of the embodiments disclosed herein. For an exercise, there may be a critical velocity associated with a 1RM. That is, if a repetition of an exercise is performed by the user near or at the critical velocity, the load may be the 1RM. For example, a critical velocity for a bench press may be about 0.17 m / s, or other appropriate velocity for a given type of exercise. Thus, if a user performs a repetition of an exercise with this predetermined concentric velocity, the load used for the repetition may be considered the 1RM. One approach for estimating 1RM involves measuring a first concentric velocity at a first load and measuring a second concentric velocity at a second load, and by assuming a linear relationship the 1RM may be estimated. For example, the first concentric velocity may be measured at about 40% of the load of the 1RM, and the second concentric velocity may be measured at about 80% of the load of the 1RM, and by H0776.70159WO00 12339990.1assuming a linear relationship between load and concentric velocity, which may be referred to as the load-velocity relationship, the projected load associated with 0.17 m / s may be estimated as the 1RM.
[0021] While there are improvements associated with using velocity changes versus load to estimate the 1RM, there are still challenges associated with such an approach including high variability in measured concentric velocities for relatively lower loads. Thus, if the initial estimate of 1RM is inaccurate, this error may lead to inappropriate loads being selected which may lead to a higher variability in estimated 1RM and increased error. An inaccurately estimated 1RM may reduce the effectiveness of training that is based on the 1RM, may increase the risk of injury, and / or may not accurately reflect the strength of a user over time. Further, it may be difficult or impossible to select, for example, 40% and 80% of the 1RM if the 1RM is unknown. In some embodiments, if the percentage difference in weights between the two loads a user is lifting with is small (e.g., the difference is less than 30%), the estimated 1RM may be inaccurate. This difficulty in selecting appropriate loads may further reduce the accuracy of an estimated 1RM, the error of which may be compounded over time as the muscle strength of the user changes.
[0022] In addition to the forgoing, each of the above-described approaches to estimating 1RM require performing repetitions of an exercise in a specific manner, which may not be within an exercise plan (e.g., a plan designed to optimally increase muscle strength) of the user. As such, each of the approaches present a burden in estimating the 1RM by requiring additional effort and / or time from the user and / or deviation from the desired exercise routine of the user.
[0023] In view of the above, the inventors have recognized a need for improved systems and methods for estimating 1RM. The improvements may include reduced variability and higher accuracy of estimated 1RM. Additionally, in some embodiments, the implemented methods and system may be used without altering the exercise plan of the user, thereby reducing burden on the user. As described herein, the systems and methods for estimating muscle strength may be implemented for any appropriate exercise, including but not limited to a bench press, squat, deadlift, bicep curl, row, overhead press, triceps extension, leg press, leg extension, lunge, pull- down, or pull-up, any type of pushup, sit-to-stand, stair training, and / or any other appropriate exercise. Such exercise may be performed without additional weights or with any types of equipment, including but not limited to a barbell, dumbbell, cable, kettlebell, weight training machine, and / or any other appropriate equipment.
[0024] A first method of estimating muscle strength recognized by the inventors includes obtaining an estimated 1RM (e.g., one-repetition maximum weight) for a target exercise and H0776.70159WO00 12339990.1commanding a user to perform the target exercise with a first weight and a second weight different from the first weight, where the first weight and the second weight are determined based at least in part on the estimated 1RM. A first concentric velocity and a second concentric velocity may be obtained while the user performs the target exercise with the first weight and second weight, respectively. The estimated 1RM for the target exercise may then be updated based at least in part on the first concentric velocity and the second concentric velocity. In some embodiments, updating the estimated 1RM includes a linear extrapolation based at least in part on the first concentric velocity, the second concentric velocity, and a critical velocity (which may be a predetermined critical velocity) associated with the 1RM for an exercise. In embodiments where the critical velocity is a predetermined critical velocity, the predetermined critical velocity may be between or equal to about 0.15 m / s and 0.20 m / s (e.g., about 0.17 m / s) according to some embodiments. In some embodiments, the critical velocity may be determined by an interpolation based on the first and second mean concentric velocities and the weights associated with the first and second mean concentric velocities. In contrast to existing approaches to estimating 1RM which do not update the estimated weights over time (e.g., do not update the first weight and the second weight), this method may update the amounts of weight that a user is commanded to user over time which may help to provide a more accurate estimation of the 1RM. For example, in preceding discussions where 40% and 80% of the load of the 1RM are used as weights, the 40% and the 80% may accordingly be updated based on the updated 1RM (e.g., when the estimate of the 1RM increases the corresponding weights determined on this initial estimate may also be updated).
[0025] In some embodiments, the updated estimated 1RM may be output to a user. In some embodiments, the method for estimating muscle strength may include determining a concentric velocity (e.g., mean concentric velocity) during a repetition of the target exercise. For example, the user may perform the concentric movement during the repetition of the exercise at various velocities throughout the repetition, and the mean velocity during the concentric movement of the repetition may be obtained by averaging the concentric velocity during the concentric movement of that repetition. In instances where multiple repetitions are performed, multiple mean concentric velocities may be obtained (i.e., a mean concentric velocity for each repetition) and a maximum mean concentric velocity, which may be the maximum (e.g., greatest) velocity of the mean concentric velocities, may be obtained. Updating the estimated one-repetition maximum weight may be based at least in part on the mean concentric velocity and / or on the maximum mean concentric velocity according to some embodiments. In some embodiments, obtaining the first concentric velocity and the second concentric velocity may H0776.70159WO00 12339990.1include sensing a velocity associated with movement of the first weight and / or the second weight using one or more sensors attached to a body part of the user or the weight the user is lifting. For example, an inertial measurement unit (IMU) may be coupled to a wrist of the user and may be configured to measure velocity of the wrist during the exercise. However, any method and / or sensor capable of measuring movement of a weight during an exercise (e.g., a camera or a linear velocity transducer) may be used and the sensor may be mounted on any appropriate portion of a user, weight, and / or machine used during the exercise. As previously discussed, the concentric velocity described herein may include a mean concentric velocity of a repetition, a maximum (e.g., peak) concentric velocity, a median concentric velocity, and / or any other appropriate concentric velocity.
[0026] As noted above, in some instances it may be desirable to measure the strength of a user during their normal exercise routine. Thus, the inventors have also recognized the benefits associated with a second method for estimating muscle strength, which may include obtaining a first concentric velocity associated with a first set of an exercise performed with a weight and obtaining a second concentric velocity associated with a second set of the exercise performed with the weight. The second method for estimating muscle strength may use concentric velocity and / or any other appropriate information from at least 2 and preferably 3 to 5 sets of the exercise, although any appropriate number of sets may be used. While each set may include any appropriate number of repetitions, for example a set may range from 1 repetition to 30 repetitions, a set may preferably range from about 6 repetitions to about 12 repetitions. There may be one or more parameters associated with each set, including but not limited to consecutive repetitions successfully performed, amount of weight, differences in concentric velocities, differences in mean concentric velocities, differences in peak concentric velocities, differences in estimated 1RM, differences in average range of motion across repetitions and any other appropriate parameter. For example, a parameter may include a change in the maximum mean concentric velocity between sets. In some embodiments, the change in maximum mean concentric velocity between the first set performed by the user and the last set performed by the user may be used. The one or more parameters associated with the first set and the second set may be provided to a trained one-repetition maximum weight estimation statistical model, and the trained one-repetition maximum weight estimation statistical model may output an estimated 1RM. According to some embodiments, the trained one-repetition maximum weight estimation statistical model may be a hyperparameter statistical model, feed forward neural network, any appropriate neural network, any appropriate regression model (e.g., random forest regressor, H0776.70159WO00 12339990.1support vector regressor, K-nearest neighbors regressor, etc.), any optimization approach, and / or any other appropriate statistical model as the disclosure is not so limited.
[0027] In view of the above, the inventors have also recognized the benefits associated with a method of training a weight estimation statistical model. Training data may be obtained including concentric velocities associated with performing sets of an exercise with weights and associated one-repetition maximum weights of one or more subjects performing the exercise with the weights. The training data may be used to generate a trained weight estimation statistical model using the training data and the trained statistical model may be stored on non- transitory computer readable memory for subsequent use. It should be understood that while specific embodiments of methods of training statistical models are presented herein, the methods presented in this application may use any appropriate method of training statistical models as the disclosure is not so limited.
[0028] Any of the above methods, or other methods disclosed herein, may be implemented on any appropriate computing device including one or more processors and one or more associated non-transitory computer readable memory. For example, an application may be used to implement the methods disclosed herein. The application implemented on smartphone, smartwatch, or other mobile computing device, may direct the user to perform the exercise, for example by directing the user to perform the exercise using a specified amount of weight or range of weight, for a specified number or range of repetitions and / or sets, and / or when to perform the sets and / or repetitions. In some embodiments, the application may direct the user to rest for a specified time or range of time between performing sets. The user may be able to input information into the application, including but not limited to repetitions and / or sets successfully performed, difficulty of repetitions and / or sets, timing of repetitions and / or sets, perceived exertion, discomfort associated with the exercise, questions associated with the exercise, and any other appropriate input. The application may provide answers to the questions provided by the user and / or may direct the user to perform the exercise differently or the same based at least in part on input from the user. Additionally, it should be understood that the application or any other method disclosed herein may be implemented on any appropriate computing device as the disclosure is not so limited.
[0029] As referred to herein, “concentric” refers to a portion of an exercise where a user activates his or her muscles to move a load (e.g., the user is moving a portion of their body engaged with an interface of an exercise machine and / or a load such that movement of the portion of the user’s body is moving in a direction at least partially opposite the direction of the load applied to the user’s body during the exercise). In some exercises, the concentric movement H0776.70159WO00 12339990.1and associated concentric velocity may involve the user moving a weight against a force of gravity. For example, the concentric movement of a bench press includes a user pushing a weight upwards against a force of gravity acting in a downwards direction that is at least partially opposite to the direction of force applied to the user to the weight. As such, the concentric velocity for the bench press is directed upwards in this embodiment. As previously mentioned, the systems and methods herein are not limited to bench press and may be any appropriate exercise with any appropriate direction for the concentric movement associated with the exercise.
[0030] As referred to herein, “range of motion” may refer to a distance or path associated with an exercise performed by a user. For example, a range of motion of an exercise may be defined at a starting position of a weight moved by the user and an ending position of a weight. The range of motion may be a linear distance, angular range, a combination of the above, or other appropriate measure of movement between the starting position and the ending position according to some embodiments. It should be understood that the range of motion may include a plurality of distances or other measurements and may depend on the exercise being performed and the user performing the exercise.
[0031] As referred to herein, a plurality of consecutive sets may refer to two or more sets performed consecutively. For example, a user may perform a first set, second set, third set, and fourth set, in this order. Consecutive sets accordingly may include the first set and the second set, the second set, the third set and the fourth set, and any other consecutive sets. As referred to herein, non-consecutive sets may refer to two or more sets which are not performed consecutively. In the aforementioned example, the first set and the fourth set are non- consecutive sets. However, it should be understood that when a first set and second set are referred to herein that these sets may either be performed consecutively or there may be one or more intermediate sets performed between the sets as the disclosure is not limited in this fashion.
[0032] As referred to herein, sub-maximal may refer to repetitions and / or weight which are less than a maximum weight and / or number of repetitions that a user may be capable of performing for a set of an exercise.
[0033] Concentric velocities for any of the methods disclosed herein may be determined from the measured velocity information in any appropriate manner. For example, in some embodiments this may include the use of one or more of inertial sensors, (e.g., an Inertial Measurement Unit), vision sensors (e.g., RGB, RGB-D, and / or stereo camera), laser / lidar sensors, linear transducer, and / or any other appropriate sensors configured to sense a velocity. H0776.70159WO00 12339990.1This may also include the use of deterministic algorithms, directional velocity thresholds, trained statistical models to identify the exercise and / or velocities associated with the exercise being performed (e.g., see US provisional application number 63 / 438,460 filed on January 11, 2023 entitled Methods and Systems for Exercise Detection and US provisional application number 63 / 438,465 filed on January 11, 2023 entitled Methods and Systems for Quantification of Movement Kinematics the disclosures of each of which are incorporated by reference herein in their entirety), object tracking algorithms, human pose estimation algorithms and / or any other appropriate method for determining the concentric velocity of a load applied to a user during an exercise as the disclosure is not limited in this fashion.
[0034] Turning to the figures, specific non-limiting embodiments are described in further detail. It should be understood that the various systems, components, features, and methods described relative to these embodiments may be used either individually and / or in any desired combination as the disclosure is not limited to only the specific embodiments described herein.
[0035] Any appropriate strength estimation system may be used with any of the methods disclosed herein for any appropriate type of exercise. One embodiment of a strength estimation system 100 is shown in Fig.1. The strength estimation system 100 includes a user 102 performing an exercise such that a weight 104 is moved by the user. In the depicted embodiment, the weight 104 may be moved by the user 102 in the directions indicated by the arrows 106. A sensor 108 may be coupled to a body part of the user 102 and may be configured to sense a velocity associated with the exercise. In the depicted embodiment of Fig.1, the user 102 is performing a bench press and the sensor 108 is coupled to a wrist of the user, though the sensor could be attached directly to a component attached to the weight or to the weight itself. Additionally, instances in which one or more sensors are mounted on a portion of an exercise machine that corresponds to a movement under an applied load (e.g., a handle bar or other moving portion of a machine) may be used. In either case, the sensor 108 may be configured to measure a velocity associated with the exercise, and preferably a concentric velocity. In the depicted embodiment of Fig.1, the sensor may be configured to measure concentric velocity of the wrist of the user 102, that is, when the user 102 moves the weight 104 away from his / her body in a direction opposite to the load applied to the portion of the user’s body supporting the weight. In some embodiments, the concentric velocity is the velocity of the weight being moved in a direction that is opposite to the direction of gravity though this may again vary depending on the type of exercise being performed. In the depicted embodiment of Fig.1, gravity may be acting in a direction downward towards the body of the user. The sensor 108 may be any appropriate sensor as the disclosure is not so limited, including an IMU, a motion camera, a H0776.70159WO00 12339990.1standard camera, smart device, any other appropriate photosensitive detector, combinations of the foregoing, and / or any other appropriate sensor or combination of sensors configured to directly or indirectly sense the velocity associated with the exercise.
[0036] A first method for estimating muscle strength 200 according to some embodiments is shown in Fig.2. Initially, a 1RM weight (which may be an initial 1RM weight) of a user for a target exercise may be obtained, see 202. In some embodiments, the initial 1RM may be obtained by any 1RM measurement or estimation approaches or based on a user’s anticipation. A first weight and a second weight may then be determined based at least in part on the estimated 1RM max, see 204. The first weight and the second weight may be, for example, predetermined percentages of the 1RM and specifically may be about 40% and about 80% of the estimated 1RM weight in some embodiments. At 206, a user may be commanded to perform the target exercise with the first weight and the second weight. As previously discussed, concentric velocities may then be obtained while the user performs the exercise. A first concentric velocity associated with the user performing the target exercise with the first weight during a first set and a second concentric velocity associated with the use performing the target exercise with the second weight during a second set may be obtained, see 208 and 210. As described herein, obtaining the concentric velocity may include sensing a concentric velocity, optionally via one or more sensors configured to directly or indirectly sense a concentric velocity associated with the target exercise according to some embodiments. The estimated 1RM weight may be updated based at least in part on the first concentric velocity and the second concentric velocity, see 212. In some embodiments, this updated estimate may be stored in non-transitory computer readable memory for subsequent recall and use as the input for the estimated 1RM weight for subsequent iterations of the method by the user. This may lead to more accurate estimates of the 1RM overtime as the prior iteration of the process may be used to more accurately select the weights for use in estimating the 1RM during the next iteration of the method. In addition to the above, optionally, the updated estimated 1RM weight may also be output to the user using any appropriate display, see 214.
[0037] The first method for estimating muscle strength 200 may benefit the user by allowing the user to receive an estimated 1RM without directly determining their 1RM. The first method for estimating muscle strength 200 may provide an updated first weight and an updated second weight according to some embodiments. For example, upon estimating a 1RM, an updated 40% 1RM and an updated 80% 1RM may be provided. In some embodiments, the updated first weight and the updated second weight may be obtained using linear interpolation based at least in part on the first concentric velocity and the second concentric velocity. H0776.70159WO00 12339990.1
[0038] In some embodiments, a user may perform the commanded sets of exercise with the two different weights using any appropriate number of repetitions. For example, a set of three repetitions of an exercise may be commanded and a concentric velocity during each repetition may be sensed. In some instances, the mean concentric velocity may be determined for each repetition within a set. The maximum (e.g., greatest) mean concentric velocity may then be determined for each set. The maximum mean concentric velocity for each set may then be used to estimate the 1RM using the above noted interpolation as elaborated on further below. As mentioned, mean concentric velocity may be replaced with peak concentric velocity or any other velocity measures.
[0039] Fig.3 shows an embodiment of a graphical representation of the first method for estimating muscle strength 200 described above with respect to Fig.2. The graph 300 has a horizontal axis representing concentric velocity which is labeled as “Velocity.” The vertical axis of graph 300 represents weight, labeled as “Weight.” Following from the first method for estimating muscle strength 200, a first concentric velocity ^^and a second concentric velocity ^^associated with a first weightand a second weight ^^may be obtained. In some embodiments, the first concentric velocityand the second concentric velocity ^^may be mean concentric velocities or maximum mean concentric velocities as described herein. As shown on graph 300, a first point ^^may be associated with the first concentric velocityand the first weightand a second point ^^may be associated with the second concentric velocity ^^and the second weight ^^. As shown on graph 300, a line connects the first point ^^and the second point ^^. Assuming the relationship between concentric velocity and weight is approximately linear, additional points may be plotted on graph 300 on a line following the linear relationship (e.g., linear equation). For example, a third point ^^associated with an estimated 1RM weight ^^^^and a critical velocity ^^^^^, which may be defined as an expected threshold velocity associated with the 1RM for a user for a specific type of exercise, may be plotted following the linear relationship. As discussed herein, the critical velocity may be approximately 0.17 m / s for bench press or found based on the abovementioned load-velocity relationship for certain exercises according to some embodiments. However, the critical velocity may be any appropriate velocity associated with the 1RM for a given exercise as the disclosure is not so limited and may be different for different people or different exercises. In either case, using the critical velocity in combination with the measured velocities for the different weights, the 1RM for a user may be estimated. It should be understood that the present disclosure is in no way limited to a graphical representation, and rather the depicted embodiment of Fig.3 is H0776.70159WO00 12339990.1provided as a visual illustration of the disclosed systems and methods. Thus, the disclosed relationship and methods for determining the 1RM may be determined in any appropriate fashion by a computing device including one or more processors configured to execute the disclosed method.
[0040] The inventors have also recognized that parameters associated with changes in the observed concentric velocity between sets of an exercise with a single weight may be used to estimate the 1RM of a user. Thus, in some embodiments, another method for estimating muscle strength 400 may be implemented as shown in Fig.4. In the depicted embodiment, a first concentric velocity (feature 2 in Fig 7) and the number of repetitions (feature 1 in Fig 7) associated with a first set of an exercise performed with a weight, and a second concentric velocity associated with a second set of an exercise performed with a weight may be obtained, see 402 and 404. As previously discussed, one or more parameters may be associated with the first set and the second set of the exercise and may be based at least in part on the first concentric velocity and the second concentric velocity. Similar to the above embodiment, in some applications, a maximum mean concentric velocity may be determined for each set of exercise performed by the user with the weight. Appropriate parameters that may be input to the model for determining an estimated 1RM may include consecutive repetitions successfully performed, amount of weight, differences in concentric velocities, differences in mean concentric velocities, range of motion, differences in estimated 1RM based on any approach, and any other appropriate parameter, see 405. At 406, one or more parameters associated with the first set and the second set may be provided to a trained one-repetition maximum weight estimation statistical model. For example, in one embodiment, a difference in the maximum mean concentric velocity between the two or more sets of the exercise may be determined and input to the trained one-repetition maximum weight estimation statistical model (feature 4 in Fig 7). In another embodiment, a difference in the estimated 1RM between the first and second sets using any submaximal approaches based on the number of repetitions may be determined and input to the trained one-repetition maximum weight estimation statistical model (feature 3 in Fig 7). Another parameters may be difference in the range of motion between the two sets. Regardless of the specific parameters used, the trained one-repetition maximum weight statistical model may output an estimated one-repetition maximum weight, see 408.
[0041] As described herein, the quantity of sets for the second method may be two or more sets. The quantity of sets for the second method may be any appropriate number of sets as the disclosure is not limited in this fashion. If a user performs more than two sets of an exercise, H0776.70159WO00 12339990.1the above noted differences may be calculated using consecutive sets (e.g., the second and the third sets) and / or any non-consecutive sets (e.g., the first and the last sets).
[0042] The second method for estimating muscle strength 400 may provide the user the benefit of estimating a 1RM without interrupting the exercise plan of the user and not requiring additional effort outside of the exercise plan. For example, a user may perform a first set of an exercise and a second set of an exercise (in some embodiments, more than two sets) within a predetermined training interval for the exercise to be performed. The user may rest for any appropriate period of time in between sets according to some embodiments. This method may be reported daily, weekly, monthly, or at any other appropriate interval and may be used to track the user’s strength over time without requiring a separate measurement and testing session. As such, an estimated 1RM may be updated with any desired timing without interrupting or requiring additional effort beyond the exercise plan of the user. In some embodiments, the exercise plan of the user may be updated based at least in part on the updated estimated 1RM.
[0043] The statistical models described herein may be trained in any appropriate manner as the disclosure is not so limited. One embodiment of training a weight estimation statistical model is depicted in Fig.5. Concentric velocity training data 500 and sub-maximum weight training data 502 may be provided to a weight estimation training module at 504. In some embodiments, the one-repetition maximum weight training data 502 may be sub-maximal weight training data. The concentric velocity training data may include any one or more of the above noted parameters (for example, including but not limited to feature 2 and feature 4 in Fig. 7 and the parameters discussed in paragraph
[0034] ), and in some embodiments, may include a difference between a concentric velocity (e.g., mean concentric velocity) of the two or more sets of repetitions of the exercise. Sub-maximal weight training data may include estimated 1RMs based on, but not limited to any sub-maximal 1RM estimation approaches. A weight estimation statistical model 506, which may now be trained using this training data, may be output at 504. The training data herein may refer to data from a single person or a group (e.g., plurality) of people. Appropriate training methods may include, but are not limited to a hyperparameter statistical model, feed forward neural network, any appropriate neural network, any appropriate regression model, any optimization approach, and / or any other appropriate statistical model. The trained one-repetition maximum weight statistical model may be stored for subsequent use, see 508. The one-repetition maximum weight statistical model may be trained for a single person or a group of people. Any appropriate statistical model may be used herein as the disclosure is not so limited. Preferably, the statistical models as described herein may be a neural network, feedforward neural network, any other appropriate neural network, any appropriate regression H0776.70159WO00 12339990.1model (e.g., random forest regressor, support vector regressor, K-nearest neighbors regressor, etc.), any optimization approach.
[0044] The training data 500 and 502 in some embodiments may be obtained from one or preferably more users performing an exercise with any appropriate frequency and sets of repetitions over any appropriate period of time. For example, a plurality of users may perform the exercise on 17 different days over the course of 8 weeks. The users may perform the exercise 2 to 3 times per week, with about 2 days of rest between each day of exercise. Additionally any appropriate delay between sets for a given exercise may be used during the training data collection process where the delay between sets may correspond to a delay period between exercise sets to be performed by a user. Any appropriate number of concentric velocity data and sub-maximum weight data may be collected for a desired application and any appropriate number of the collected data may be used as verification data to test the accuracy of the model. Appropriate numbers of data points may be greater than or equal to 10, 50, 100, 200, 500, or any other appropriate number. The number of data points may also be less than or equal to 5000, 1000, 500, 200, 100, or any other appropriate number. Combinations of the forgoing are contemplated including a number of data points between or equal to 10 and 5000, though numbers of training data both less than and greater than the above are also contemplated.
[0045] As shown in Fig.6, mean concentric velocity (indicated by the dotted lines 601 in Fig 6) for any of the methods disclosed herein may be determined by calculating the average velocity during the concentric phase of one repetition. In some embodiments, a peak concentric velocity (indicated by the dashed-dotted lines 602 as shown in Fig 6), may be the maximum velocity of the concentric phase in a repetition and may be used in place of mean concentric velocity as described herein. In some embodiments, a velocity measurement other than a mean or peak concentric velocities may be used with the various embodiments disclosed herein, and it may be obtained using a hyperparameter statistical model, feed forward neural network, any appropriate neural network, any appropriate regression model (e.g., random forest regressor, support vector regressor, K-nearest neighbors regressor, etc.), any optimization approach, and / or any other appropriate statistical model as the disclosure is not so limited. The graphical data presented in the depicted embodiment of Fig.6 is merely an example of data that may be associated with the systems and methods described herein, as any appropriate velocities over any appropriate time may be associated with the systems and methods described herein. In some embodiments, gravity may act in a direction opposite to the concentric phase velocity, indicated by the dashed arrow in the depicted embodiment of Fig.6. H0776.70159WO00 12339990.1
[0046] An example of the disclosed methods is illustrated in the depicted embodiment of Fig.7. As a user performs an exercise, a sensor 702 coupled to a wrist of the user senses data, including concentric velocities associated with the exercise. As previously mentioned, the sensor may be mounted on any appropriate portion of a user, weight, and / or machine used during the exercise. The sensed data 704 is input to a processor 706 including a parameter sensing module 708, which may be configured to obtain features from the sensed data 704. The sensed data may include any concentric velocity or other appropriate data as discussed herein. For example, as discussed above, the concentric velocity may include a mean concentric velocity of a repetition, a maximum (e.g., peak) concentric velocity, a median concentric velocity, and / or any other appropriate concentric velocity. For example, in the depicted embodiment of Fig.7, the parameter sensing module 708 may 2, feature 3, and feature 4, and any other appropriate feature. Feature 1 includes weight and obtain feature 1, feature quantity of repetitions associated with the exercise, which may be sub-maximal. Feature 2 includes velocities associated with the exercise. Feature 3 includes changes in the repetitions and / or weight among two or more sets of repetitions. Feature 4 includes changes in the velocities among two or more sets of repetitions. One or more of the features from the parameter sensing module 708 may be input into the trained statistical model module 710, which may be used to output a 1RM 714. A memory module 712 may be included in the processor 706 and may store data and / or features for subsequent use and / or training.
[0047] Example: Cyclic Estimation of One Repetition Maximum Weight Methodology:
[0048] The first method for estimating muscle strength using different weights and measured velocities (referred to below as VEL) was implemented to evaluate the performance of the systems and methods. As previously discussed, using fixed weights (e.g., 40% and 80%) of a 1RM without adjusting the fixed weights may increase error associated with estimating 1RM over time. This was addressed by implementing a weight suggestion algorithm that updated the submaximal weights for the exercise based on the 1RM estimated from a previous exercise session. This method accounted for the changes in 1RM over time, and thereby reduced the error in estimated 1RM. The two implemented algorithms were a 1RM velocity detection algorithm that finds the velocity (e.g., critical velocity) that corresponds to the 1RM weight based on extrapolation of an individual user’s maximum mean concentric velocities and the weights (e.g., 40% and 80%) and a weight update algorithm that suggests the target weights to lift based on interpolation for the next session (VEL-A). If there was no prior VEL session, ^^was either the 1RM measured through the 1RM test, or a user-anticipated 1RM, which was updated using the previously estimated 1RM. The users performed sets of the exercise using weights W equal to H0776.70159WO00 12339990.140% and 80% of W0, ^^^%and ^^^%, respectively. Concentric velocity was obtained using an inertial measurement unit (IMU) placed on the body part of the user that moves the weight or the weight itself, and the IMU was configured to sense concentric velocities during the exercise.
[0049] The 1RM velocity detection algorithm provided critical velocity that is different from user-to-user and day-to-day, instead of a fixed number (e.g., approximately 0.17 m / s for bench press). Critical velocity, ^^^^^, was calculated using the equation:
[0050] MCVmaxwas measured as the maximum mean concentric velocity measured for the set of repetitions performed by the user with his / her maximum velocities for the exercise completed for the indicated weight (e.g., 40% and 80%). Further details and results of the implementation of the first method of estimating muscle strength are described in a preceding paragraph in conjunction with the second method of estimating muscle strength.
[0051] The weight update algorithm suggested target weights, ^&'() ^^%and , ^&'() ^^%, to lift for the next time the user performs sets of exercise.40% and 80% of the weights, ^^^%and ^^^%, were calculated using the equations:Where ^^^^^^,^^%, ^and ^^^^^^,^^%, ^are the maximum mean concentric velocities measured in the first VEL-A trial, ^^^^^^,^^%,and ^^^^^^,^^%,are the maximum mean concentric velocities measured and ^^^%and ^^^%are the lifted submaximal weights in the latest VEL-A trial.
[0052] Example: Statistical Model for Estimating One Repetition Maximum Weight Methodology:
[0053] As previously discussed, existing 1RM estimation methods are burdensome as their procedures require excessive time and effort. The second method for estimating muscle strength enables a user to estimate the 1RM for an exercise while avoiding excessive and effort. The second method for estimating muscle strength (labeled as DAILY in the proceeding paragraphs) was also implemented to evaluate performance. Users performed the exercise for 6 to 12 repetitions for 2 to 5 sets of volitional (e.g., user-selected) weights lifted at volitional speed without failure with a rest period in between sets. This reduced the risk of injury as well as the expertise in strength training required of the user. The parameters provided to the trained one- H0776.70159WO00 12339990.1repetition maximum weight estimation statistical model were as follows: 1) Submaximal 1RM prediction in the set 1 (1RMset1): The first feature was the submaximal 1RM prediction based on the data of the first set. The 1RM was assumed to be the average between two submaximal 1RM prediction regression-based equations, 1RMEand 1RMM, equations 1 and 2, respectively. Equations 3 and 4 used weight w and the total number of repetitions r as inputs to predict the estimated 1RM of the user.
[0054] 2) Velocity feature in the set 1 (^-.^^): The feature ^-.^^represented the maximum mean concentric velocity of the determined mean concentric velocities v_concrep of repetitions in the first set, shown in Equation 5.
[0055] 3) Submaximal fatigue feature / 0: Submaximal fatigue was defined as the difference between the submaximal 1RM prediction based on the first set and the 1RM prediction based on the last set (Equation 6 and 7).
[0056] 4) Velocity fatigue feature / ^.1: Velocity fatigue feature was defined as the decrease in maximum concentric velocity from the first set Vset1to the last set Vset2(Equation 8). H0776.70159WO00 12339990.1
[0057] The number of repetitions for each set were calculated using a repetition counting algorithm. A feed-forward neural network was used to examine the relationship between the aforementioned features and 1RM, but as previously discussed herein, any other appropriate statistical models, including but not limited to linear regression, may be used as the disclosure is not so limited.
[0058] Example: Performance Evaluation:
[0059] To evaluate the performance of the first method for estimating muscle strength and the second method for evaluating muscle strength, sixteen participants (eight male and eight female) participated in a study over an 8 week period of time. The target exercise was a free- weight bench press. A 1RM for each user was measured in the 1st, 5th, and 8th weeks, using direct determination, an existing approach to estimating 1RM, following protocol from the National Strength and Conditioning Association. After measuring the 1RM for each user in the 5th and 8th week, the users had at least a 48-hour break before performing a first bout and a second bout of the Two-point method to estimate 1RM: The first bout choosing the 40% and 80% of 1RM measured in the 1st week (VEL-F), and the second bout using the suggested weights based on the first method for estimating muscle strength (VEL-A). The order of the bouts was chosen randomly for each participant. Between each bout, each participant took a five-minute break. After performing VEL-F and VEL-A, the users had at least a 48-hour break before performing 5 sets of bench press at self-selected weights and volitional movement speeds. For each set, the users chose a weight they could successfully perform 6-12 repetitions with. Participants performed the repetitions until either 12 repetitions were successfully completed or until volitional fatigue. Volitional fatigue as referred to in the study was defined as stopping 1-2 repetitions before failure. The users rested between sets for 3 minutes. The measured concentric velocities for the self-selected weights were analyzed using the Daily algorithm described above. While performing the bench press throughout the study, the users wore IMUs on their wrist configured to sense concentric velocities of the bench press.
[0060] Table 1 shown below tabulates a statistical evaluation of the 1RM estimates at the 4th and 8th weeks of the study using the different 1RM estimation approaches (VEL-F, VEL-A, and DAILY). The results of the study suggest that both methods for evaluating muscle strength disclosed in the current application, VEL-A and DAILY, outperformed VEL-F where no updates or additional analysis were performed. The mean absolute errors were 6.9 lbs and 6.7 lbs for VEL-A and DAILY, respectively which are both less than the error of 10.0 lbs observed for Vel- F. The results of the study also suggest that VEL-A exhibits a higher level of linearity (r > 0.99) compared to VEL-F (r < 0.99). The statistical comparison was conducted using mean absolute H0776.70159WO00 12339990.1error (MAE), mean absolute percentage error (MAPE), and standard deviations (SD) of the error. The Shapiro-Wilk test was used to confirm the normal distribution of the data (p > 0.05). A paired sample t-test was performed to compare the ground truth with the new method. Pearson’s correlation coefficient (r) was also calculated to assess the correlation between the two methods and was classified as trivial (< 0.1), small (0.1-0.3), moderate (0.3-0.5), high (0.5-0.7), very high (0.7-0.9), or practically perfect (> 0.9). The standard error of the estimate (SEE) and the root mean squared error (RMSE) were also calculated. Table 1. Evaluation of 1RM Prediction Methods.
[0061] The various methods disclosed above may be implemented by one or more controllers including at least one processor operatively coupled to the various controllable portions of a strength estimation system as disclosed herein. Alternatively or additionally, in some embodiments, the disclosed methods may be performed at least in part, and in some instances completely, on a computing device that is separate and removed from the disclosed strength estimation systems. In either case, the disclosed methods may be embodied as computer readable instructions stored on non-transitory computer readable memory associated with the at least one processor such that when executed by the at least one processor the associated system, which may be a strength estimation system in some embodiments, may perform any of the actions related to the methods disclosed herein. Additionally, it should be understood that the disclosed order of the steps is exemplary and that the disclosed steps may be performed in a different order, simultaneously, and / or may include one or more additional intermediate steps not shown as the disclosure is not so limited.
[0062] The above-described embodiments of the technology described herein can be implemented in any of numerous ways. For example, the embodiments may be implemented using hardware, software or a combination thereof. When implemented in software, the software code can be executed on any suitable processor or collection of processors, whether H0776.70159WO00 12339990.1provided in a single computing device or distributed among multiple computing devices. Such processors may be implemented as integrated circuits, with one or more processors in an integrated circuit component, including commercially available integrated circuit components known in the art by names such as CPU chips, GPU chips, microprocessor, microcontroller, or co-processor. Alternatively, a processor may be implemented in custom circuitry, such as an ASIC, or semicustom circuitry resulting from configuring a programmable logic device. As yet a further alternative, a processor may be a portion of a larger circuit or semiconductor device, whether commercially available, semi-custom or custom. As a specific example, some commercially available microprocessors have multiple cores such that one or a subset of those cores may constitute a processor. Though, a processor may be implemented using circuitry in any suitable format.
[0063] Further, it should be appreciated that a computing device may be embodied in any of a number of forms, such as a rack-mounted computer, a desktop computer, a laptop computer, or a tablet computer. Additionally, a computing device may be embedded in a device not generally regarded as a computing device but with suitable processing capabilities, including a Personal Digital Assistant (PDA), a smart phone, tablet, or any other suitable portable or fixed electronic device.
[0064] Also, a computing device may have one or more input and output devices. These devices can be used, among other things, to present a user interface. Examples of output devices that can be used to provide a user interface include display screens for visual presentation of output and speakers or other sound generating devices for audible presentation of output. Examples of input devices that can be used for a user interface include keyboards, individual buttons, and pointing devices, such as mice, touch pads, and digitizing tablets. As another example, a computing device may receive input information through speech recognition or in other audible format.
[0065] The various methods or processes outlined herein may be implemented in any suitable hardware. Additionally, the various methods or processes outlined herein may be implemented in a combination of hardware and of software executable on one or more processors that employ any one of a variety of operating systems or platforms. Examples of such approaches are described above. However, any suitable combination of hardware and software may be employed to realize any of the embodiments discussed herein.
[0066] Also, the various methods or processes outlined herein may be coded as software that is executable on one or more processors that employ any one of a variety of operating systems or platforms. Additionally, such software may be written using any of a number of H0776.70159WO00 12339990.1suitable programming languages and / or programming or scripting tools, and also may be compiled as executable machine language code or intermediate code that is executed on a framework or virtual machine.
[0067] In this respect, various inventive concepts may be embodied as at least one non- transitory computer readable storage medium (e.g., a computer memory, one or more floppy discs, compact discs, optical discs, magnetic tapes, flash memories, circuit configurations in Field Programmable Gate Arrays or other semiconductor devices, etc.) encoded with one or more programs that, when executed on one or more computers or other processors, implement the various embodiments of the present disclosure. The non-transitory computer-readable medium or media may be transportable, such that the program or programs stored thereon may be loaded onto any computer resource to implement various aspects of the present disclosure as discussed above.
[0068] The terms “program” or “software” are used herein in a generic sense to refer to any type of computer code or set of computer-executable instructions that can be employed to program a computer or other processor to implement various aspects of embodiments as discussed above. Additionally, it should be appreciated that according to one aspect, one or more computer programs that when executed perform methods of the present disclosure need not reside on a single computer or processor, but may be distributed in a modular fashion among different computers or processors to implement various aspects of the present disclosure.
[0069] Computer-executable instructions may be in many forms, such as program modules, executed by one or more computers or other devices. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. Typically, the functionality of the program modules may be combined or distributed as desired in various embodiments.
[0070] The embodiments described herein may be embodied as a method, of which an example has been provided. The acts performed as part of the method may be ordered in any suitable way. Accordingly, embodiments may be constructed in which acts are performed in an order different than illustrated, which may include performing some acts simultaneously, even though shown as sequential acts in illustrative embodiments.
[0071] Further, some actions are described as taken by a “user.” It should be appreciated that a “user” need not be a single individual, and that in some embodiments, actions attributable to a “user” may be performed by a team of individuals and / or an individual in combination with computer-assisted tools or other mechanisms. H0776.70159WO00 12339990.1While the present teachings have been described in conjunction with various embodiments and examples, it is not intended that the present teachings be limited to such embodiments or examples. On the contrary, the present teachings encompass various alternatives, modifications, and equivalents, as will be appreciated by those of skill in the art. Accordingly, the foregoing description and drawings are by way of example only. H0776.70159WO00 12339990.1
Claims
CLAIMS What is claimed is:
1. A method of estimating muscle strength, the method comprising: obtaining a plurality of concentric velocities associated with a plurality of sets of an exercise performed with a weight, wherein the plurality of concentric velocities include at least a first concentric velocity associated with a first set of the exercise and a second concentric velocity associated with a second set of the exercise; providing one or more parameters associated with the plurality of sets to a trained one- repetition maximum weight estimation statistical model, wherein the one or more parameters are based at least in part on the first concentric velocity and the second concentric velocity; and outputting a one-repetition maximum weight from the trained one-repetition maximum weight estimation statistical model.
2. The method of claim 1, wherein the first set and the second set are consecutive or non- consecutive.
3. The method of claim 1, wherein the trained one-repetition maximum weight estimation statistical model is a neural network, regression model, or optimization approach 4. The method of claim 1, wherein the one or more parameters includes a number of consecutive repetitions associated with the first set and the second set.
5. The method of claim 1, wherein the one or more parameters include amounts of weight associated with the first set and the second set.
6. The method of claim 1, wherein the first concentric velocity is a maximum mean concentric velocity of repetitions of the first set, and wherein the second concentric velocity is a maximum mean concentric velocity of repetitions of the second set.
7. The method of claim 1, wherein the first concentric velocity is a maximum peak concentric velocity of repetitions of the first set, and wherein the second concentric velocity is a maximum peak concentric velocity of repetitions of the second set. H0776.70159WO00 12339990.
18. The method of claim 1, wherein the first concentric velocity and the second concentric velocity are velocity measures obtained from a statistical model, such as a neural network, regression model, or optimization approach.
9. The method of claim 1, wherein the one or more parameters include a difference between the first concentric velocity and the second concentric velocity.
10. The method of claim 1, wherein the one or more parameters include a first estimated one-repetition maximum weight based on the first set and a second estimated one-repetition maximum weight.
11. The method of claim 1, wherein the exercise is a bench press, squat, deadlift, bicep curl, row, overhead press, triceps extension, leg press, chest press, lunge, pull-down, pull-up, push- up, sit-to-stand, or stair training.
12. The method of claim 1, wherein obtaining the plurality of concentric velocities includes sensing the first concentric velocity and the second concentric velocity with one or more sensors attached to one or more pieces of exercise equipment.
13. The method of claim 1, wherein obtaining the plurality of concentric velocities includes sensing the first concentric velocity and the second concentric velocity with one or more sensors attached to one or more portions of a body of the user.
14. The method of any one of the preceding claims, wherein the first set is performed with a first weight and the second set is performed with a second weight different from the first weight.
15. The method of any one of claims 12 and 13, wherein the sensors are configured to sense at least one or more of acceleration, velocity, movement trajectory, angular velocity, and magnetometer data.
16. A method of estimating muscle strength, the method comprising: obtaining an estimated one-repetition maximum weight of a user for a target exercise; H0776.70159WO00 12339990.1commanding a user to perform the target exercise with a first weight and a second weight different from the first weight, wherein the first weight and the second weight are determined based at least in part on the estimated one-repetition maximum weight; obtaining a first concentric velocity associated with the user performing the target exercise with the first weight; obtaining a second concentric velocity associated with the user performing the target exercise with the second weight; and updating the estimated one-repetition maximum weight based at least in part on the first concentric velocity and the second concentric velocity.
17. The method of claim 16, wherein the updating the estimated one-repetition maximum weight includes a linear extrapolation based at least in part on the first concentric velocity, the second concentric velocity, and a predetermined critical velocity.
18. The method of claim 17, wherein the predetermined critical velocity is either predetermined or determined based on extrapolation of mean concentric velocities and the first weight and the second weight.
19. The method of claim 16, further comprising outputting the updated estimated one- repetition maximum weight to a user.
20. The method of claim 16, wherein obtaining the first concentric velocity and the second concentric velocity includes determining a mean concentric velocity during a repetition of the target exercise.
21. The method of claim 16, wherein the first concentric velocity is a maximum mean concentric velocity of repetitions of the target exercise with the first weight, and wherein the second concentric velocity is a maximum mean concentric velocity of repetitions of the target exercise with the second weight.
22. The method of claim 16, wherein the first weight is about 40% of the estimated one- repetition maximum of the user for the target exercise. H0776.70159WO00 12339990.
123. The method of claim 16, wherein the second weight is about 80% of the estimated one- repetition maximum of the user for the target exercise.
24. The method of claim 16, wherein obtaining the estimated one-repetition maximum weight includes obtaining a previously updated estimated one-repetition maximum weight.
25. The method of claim 16, wherein the target exercise is a bench press, squat, deadlift, bicep curl, row, overhead press, triceps extension, leg press, pull-down, or pull-up.
26. The method of claim 16, wherein obtaining the first concentric velocity and the second concentric velocity includes sensing a velocity associated with movement of the first weight and / or the second weight using one or more sensors attached to a body part of the user.
27. The method of claim 16, wherein obtaining the first concentric velocity and obtaining the second concentric velocity includes sensing the first concentric velocity and the second concentric velocity with one or more sensors attached to one or more pieces of exercise equipment.
28. The method of claim 16, wherein obtaining the first concentric velocity and obtaining the second concentric velocity includes sensing the first concentric velocity and the second concentric velocity with one or more sensors attached to one or more portions of a body of the user.
29. The method of any one of claims 27 and 28, wherein the sensors are configured to sense at least one or more of acceleration, velocity, movement trajectory, angular velocity, and magnetometer data.
30. The method of any one of claims 27 and 28, wherein the sensors are inertial measurement units.
31. A method of training a weight estimation statistical model, the method comprising: obtaining training data, wherein the training data includes concentric velocities associated with performing sets of an exercise with weights, and associated one-repetition maximum weights of one or more subjects performing the exercise with the weights; H0776.70159WO00 12339990.1generating a trained weight estimation statistical model using the training data; and storing the trained weight estimation statistical model on non-transitory computer readable memory for subsequent use.
32. The method of claim 31, wherein the trained one-repetition maximum weight estimation statistical model is a neural network, regression model, or optimization approach.
33. The method of claim 31, wherein the training data is collected from a single subject and the one-repetition maximum weight estimation statistical model is trained for the single subject.
34. The method of claim 31, wherein the training data is collected from a group of subjects and the one-repetition maximum weight estimation statistical model is trained for a plurality of people.
35. The method of claim 31, wherein the one or more parameters include consecutive repetitions associated with the sets of the exercise.
36. The method of claim 31, wherein the one or more parameters include amounts of weight associated with the sets of the exercise.
37. The method of claim 31, wherein the concentric velocities are maximum mean concentric velocities of repetitions of the sets of exercise.
38. The method of claim 31, wherein the one or more parameters include differences between two or more concentric velocities.
39. The method of claim 31, wherein the one or more parameters include a first estimated one-repetition maximum weight based on the first set and a second estimated one-repetition maximum weight.
40. The method of claim 31, wherein the one or more parameters include range of motion. H0776.70159WO00 12339990.
141. The method of claim 31, wherein the exercise is a bench press, squat, deadlift, bicep curl, row, overhead press, triceps extension, leg press, chest press, leg extension, lunge, pull-down, pull-up, push-up, sit-to-stand, or stair training.
42. A non-transitory computer readable medium including processor executable instructions that when executed by a processor perform the method of any one of claims 1-41.
43. A system for estimating muscle strength comprising: one or more processors configured to perform the method of any one of claims 1-41. H0776.70159WO00 12339990.1
Citation Information
Patent Citations
Systems, methods, and devices for assisting walking for developmentally-delayed toddlers
US20160346156A1
Compositions comprising polar lipids for maintaining or increasing mobility and vitality
US20230041432A1
Technological device to assist user in workouts and healthy living
WO2016157217A2
