Systems and method for actionable real-time signal electromyography analysis

Abstract

A system for monitoring subject movement is provided. The system includes one or more sensors attached to a subject. The one or more sensors monitor and provide muscle data and inertial measurement unit (IMU) data. The system also includes one or more computing devices in communication with the one or more sensors and including at least one processor programmed to: a) receive a plurality of sensor data including muscle data and IMU data; b) preprocess the plurality of sensor data to remove noise data; c) execute at least one artificial intelligence (AI) model trained with the muscle data and IMU data, where the at least one AI model outputs a classification and an identification of a subject movement; and d) provide feedback to the subject based on the classification and identification of the subject movement.

Description

47735-3SYSTEMS AND METHOD FOR ACTIONABLE REAL-TIME SIGNAL ELECTROMYOGRAPHY ANALYSISCROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims the benefit of U.S. Patent Application No. 63 / 738,123, filed December 23, 2024, which is incorporated herein by reference in its entirety.BACKGROUND

[0002] The field of the invention relates generally to Actionable Realtime Signal Electromyography (EMG) Analysis, and more specifically to systems and methods to near-real time wireless data collections and analysis of EMG data to improve rehabilitation, physical therapy, and training in real-time.

[0003] There are a lot of devices and services for external evaluation of the body (movements, speed, strength, etc.). However, there are no effective devices and / or sen-ices that can do the same for what is occurring within the body. Electromyography (EMG) has been around for a long time and just started to branch out into more practical use due to wireless technology. Before EMG was only used in a research setting. Even now, data analysis being compared to the Gold Standard can take up to a week for five minutes of data collection when using EMG. Within a rehab setting, muscle has already changed by the time results are available and the chance for accurate intervention may have passed. In current industry, research still uses wired EMG sensors to collect data.

[0004] The current problem is having safe, reliable data that can provide actionable feedback just like the various external metric devices / services. Currently, a medical professional is required for sensor placement and to oversee data collection. When collecting data with analysis compared to the Gold Standard, most sessions would have wired (sometimes wireless) sensors in a controlled environment47735-3within a motion capture lab. In many cases, video is required to be accurately partition data for analysis.

[0005] This Background section is intended to introduce the reader to various aspects of art that may be related to various aspects of the present disclosure, which are described and / or claimed below. This discussion is believed to be helpful in providing the reader with background information to facilitate a better understanding of the various aspects of the present disclosure. Accordingly, it should be understood that these statements are to be read in this light, and not as admissions of prior art. BRIEF DESCRIPTION

[0006] In one aspect, a system for monitoring subject movement is provided. The system including one or more sensors attached to a subject, wherein the one or more sensors monitor and provide Electromyography (EMG) and inertial measurement unit (IMU) data. The system also includes one or more computing devices each comprising at least one processor in communication with at least one memory device. The one or more computing devices are in communication with the one or more sensors. The at least one processor is programmed to: a) receive a plurality of sensor data including muscle data and inertial measurement unit (IMU) data from one or more sensors monitoring a subject; b) preprocess the plurality' of sensor data to remove noise data from the plurality of sensor data; c) execute at least one artificial intelligence (Al) model trained with the muscle data and IMU data, wherein the muscle data and IMU data of the plurality of sensor data are inputs for the at least one Al model, and wherein the at least one Al model outputs a classification and an identification of a subject movement; and d) provide feedback to the subject based on the classification and identification of the subject movement. The system may have additional, less, or alternate functionalities, including those discussed elsewhere herein.

[0007] In another aspect, a computer device for monitoring subject movement is provided. The computer device includes at least one processor (or "‘the processor’’) in communication with at least one memory device. The processor is programmed to a) receive a plurality of sensor data including muscle data and inertial47735-3measurement unit (IMU) data from one or more sensors monitoring a subject; b) preprocess the plurality of sensor data to remove noise data from the plurality of sensor data; c) execute at least one artificial intelligence (Al) model trained with muscle data and IMU data, wherein the EMG and IMU data of the plurality of sensor data are inputs for the at least one Al model, and wherein the at least one Al model outputs a classification and an identification of a subject movement; and d) provide feedback to the subject based on the classification and identification of the subject movement. The computer device may have additional, less, or alternate functionalities, including those discussed elsewhere herein.

[0008] In a further aspect, a computer implemented method for monitoring subject movement is provided. The method implemented by at least one processor (or “the processor”) in communication with at least one memory' device. The method includes a) receiving a plurality of sensor data including muscle data and inertial measurement unit (IMU) data from one or more sensors attached to a subject; b) preprocessing the plurality of sensor data to remove noise data from the plurality of sensor data; c) executing at least one artificial intelligence (Al) model trained with the muscle data and IMU data, wherein the muscle data and IMU data of the plurality of sensor data are inputs for the at least one Al model, and wherein the at least one Al model outputs a classification and an identification of a subject movement; and d) providing feedback to the subject based on the classification and identification of the subject movement. The method may have additional, less, or alternate functionalities, including those discussed elsewhere herein.

[0009] Various refinements exist of the features noted in relation to the above-mentioned aspects. Further features may also be incorporated in the above-mentioned aspects as well. These refinements and additional features may exist individually or in any combination. For instance, various features discussed below in relation to any of the illustrated embodiments may be incorporated into any of the above-described aspects, alone or in any combination.47735-3BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The Figures described below depict various aspects of the systems and methods disclosed. It should be understood that each Figure depicts an embodiment of a particular aspect of the disclosed systems and methods, and that each of the Figures is intended to accord with a possible embodiment thereof. Further, wherever possible, the following description refers to the reference numerals included in the following Figures, in which features depicted in multiple Figures are designated with consistent reference numerals. There are shown in the drawings arrangements presently discussed, it being understood, however, that the present embodiments are not limited to the precise arrangements.

[0011] Figure 1A illustrates an example sensor being place on a muscle , in accordance with at least one embodiment.

[0012] Figure IB illustrates an example diagram of different muscles that may be monitored on a subject, in accordance with at least one embodiment.

[0013] Figure 2 illustrates an example muscle monitoring and analysis system for monitoring and analyzing one or more muscles in a subject, in accordance with at least one embodiment.

[0014] Figure 3 illustrates an exemplary computer system for realtime EMG analysis, in accordance with at least one embodiment.

[0015] Figure 4 illustrates a flow diagram of a process for real-time EMG analysis, in accordance with at least one embodiment of the disclosure.

[0016] Figure 5 depicts an artificial intelligence (AI) / deep learning (DL) module, according to an embodiment.

[0017] Figure 6 depicts an exemplary configuration of user computer device, in accordance with one embodiment of the present disclosure.

[0018] Figure 7 depicts an exemplary configuration of a server computer device, in accordance with one embodiment of the present disclosure.47735-3

[0019] Unless otherwise indicated, the drawings provided herein are meant to illustrate features of embodiments of this disclosure. These features are believed to be applicable in a wide variety of systems including one or more embodiments of this disclosure. As such, the drawings are not meant to include all conventional features known by those of ordinary skill in the art to be required for the practice of the embodiments disclosed herein.DETAILED DESCRIPTION

[0020] The field of the invention relates generally to Actionable Realtime Signal Electromyography (EMG) Analysis, and more specifically to systems and methods to near-real time wireless data collections and analysis of EMG data to improve rehabilitation, physical therapy, and training in real-time.

[0021] The systems and methods described herein provide for practical use of EMG sensors for wireless solutions that offer real-time data collection. In some embodiments, the EMG sensors are combined with other technologies like accelerometers and heart rate monitors for holistic assessments. This is generally used for applications like physical therapy and sports performance.

[0022] The presently described systems and methods overcome the barriers to effectively using this powerful sensor data, These barriers include, but are not limited to, (1) it general requires multiple specialists to process (z.e., - signals analysis, artifact detection, partitioning, labeling, etc.) and interpret EMG / IMU (inertial measurement unit) data and (2) the sensor set up and workflow itself has historically had a lot of friction because it has only been used primarily as a research tool or highly specialized in-clinic diagnostic tool.

[0023] The presently described systems and methods integrate a wearable sensor, an application on a user device, such as a smartphone, a cloud platform, and a web portal to monitor and analyze muscle activity and movement during training or rehabilitation. The system provides real-time biofeedback and advanced analytics using clinical-grade EMG and IMU data streamed to the user devices. The cloud processes this data in real-time providing both real-time during training feedback47735-3and detailed reports to offer insights for progress tracking and planning. The web portal archives trends and workout histories, ensuring compliance with GDPR (General Data Protection Regulation) and HIPAA (Health Insurance Portability and Accountability Act) standards. The present system and method provides quick setup, artifact-filtered data, and user-friendly analysis, looking at both real-time feedback and trend analysis and reports.

[0024] The present systems and method streamlines and mitigates the need for highly specialized engineering and data science skills that are normally required of EMG and IMU analysis through multiple software layers. The system analyzes the nature and spectrum of phenotypes of data that can be received and processed via a layered software approach to perform the necessary signals analysis, filtering, artifact detection / aliasing, partitioning, labeling, and report quality' check to get research grade biomechanical reports that breakdown a comprehensive workout movement by movement and set by set.

[0025] To our knowledge, no other company can come close to doing this. All other companies we have found only stream the raw data coming over from the sensor. Their interfaces may be sleek and clean, but not only is this signal incredibly noisy and difficult to interpret live, it poses a risk to the user because the raw data is heavily vulnerable to artifacts that can cause significant error. The raw traces and smoothing algorithms other companies use to present a sleek UI to their users hides these artifacts and, again, poses significant user risk.

[0026] The present system mitigates the workflow friction issues by implementing software protocols and user onboarding processes that simplify connecting the sensors and being ready for data collection. The system also heavily monitors data quality' in uploads and reports generated through additional software to address user errors quickly.

[0027] Furthermore, medical professionals are not required during data collection. The system assists with sensor placement and does not require additional video information for accurate analysis. This makes Remote Patient47735-3Monitoring feasible with automated biosignal analysis with downloadable reports for EHR / EMR (electronic health record / electronic medical record) systems.

[0028] In some embodiments, one detailed session with the systems and methods provided herein provides data analysis that could be considered the equivalent to a PhD student’s thesis. Based on all aspects of collection, analysis, parti tioning / segmenting, software technique (not for non-provisional), the present system provides the opportunity for an exponential grow th of assessable, accurate data. The data provided by this system may be used to generate movement libraries based on all phenotypes related to body structure and identify trends not feasible in a lab / research environment or with other systems.

[0029] In the example embodiment, the systems described herein support layers of analysis that can process and normalize the complex signal of EMG and IMU concurrently while detecting and filtering out artifacts that contaminate this type of data. The system also makes use of low level machines (LLMs) to partition and label sets of exercises for easy user interpretation for immediate in-session biofeedback and adjustments. This real-time feedback and analysis is a major advantage of the approached described herein.

[0030] Once data collection is complete, the data is uploaded to cloud systems, such as the signal analysis computer device. The signal analysis computer device generates one or more detailed reports. This includes an in-depth analysis is performed to provide a repetition by repetition breakdown of the workout with sets of repetitions grouped together. Normally, this could take days or even weeks to perform by an EMG research laboratory'. However, the systems and methods described herein allows the system to process, partition, and create these comprehensive biomechanics reports. These reports can then be used by the mass market of users require to make actionable decisions. Normally, only a miniscule portion of users are willing to take on the processing and interpretation.

[0031] The compiled data and reports are available for review by the users and other via one or more web portals. The signal analysis computer device47735-3maintains an archive of data collections (workouts) for each individual user. This comprises overview trends of training periods over time and detailed repetition by repetition breakdowns of workouts. Users can then use this information to review training regimens and program the next week or training block’s worth of training.

[0032] All processes described herein can occur dynamically, realtime on edge devices or in standalone analysis software post data collection. One of the goals of the systems and methods described herein is to assess the effectiveness of a user's workout, archive the assessment, and be able to draw trends to predict future effectiveness / performance.

[0033] Figure 1A illustrates an example sensor 105 being place on a muscle 110, in accordance with at least one embodiment. The sensors 105 are affixed to a user’s skin above the desired muscle 110 to be measured. In the example embodiment, the sensors 105 are placed so that they have two or more points of contact that follow muscle fibers. In the exemplar - embodiment, the sensors 105 generate EMG data. In some further embodiments, the sensors 105 also generate IMU data, such as, but not limited to, accelerometers and heart rate monitors. The sensors 105 monitor muscle patterning (clinical grade electromyography - EMG) and movement (accelerometer + gyroscope IMU system). Muscle patterning (EMG) and movement (IMU) data is collected by the sensors 105 attached to the user’s skin.

[0034] In the example embodiment, the sensors 105 are placed on the epidermis of the subject being monitored. The sensors have two or more points of contact that follow some embodiments, the sensors 105 are attached near different points along the muscle 110 to be monitored. In some embodiments, the subject is palpitated to ensure that the proper muscle 110 is found. In some embodiments, the sensor 105 is attached via a suction or tacky surface. In other embodiments, the sensor 105 is attached with one or more fasteners to ensure that it doesn’t fall off during exercise or other activity performed by the subject. In the example embodiment, two sensors 105 are attached to the body of the subject measuring the same muscle 110 on different sides of the subject (aka left and right). In other embodiments, any number of sensors 105 may be attached to different muscle 110 on the subject’s body.47735-3

[0035] The sensors ae configured to collect high resolution, clinical gauge EMG and IMU data and stream it to one or more computers, such as the user computer device 220 and / or the signal analysis computer device 225 (both shown in Figure 2). In many embodiments, monitoring occurs when a user 215 (shown in Figure 2) is performing one or more sets of muscle movement, such as, but not limited to, during therapy, training, exercise, and / or other activities. A user 215 wears the sensors 105 affixed to their body as they train and the sensors 105 stream data to the user computer device 220 for real-time assessment of training intensity. This may occur as physiotherapy or physical training sessions with a healthcare provider, coach, or athletic trainer.

[0036] One having ordinary skill in the art would understand that other types of sensors 105 may be used to collect other types of data to work with the systems and methods presented herein. More specifically, in addition to EMG data, other forms of muscle data could be collected, such as, but not limited to, Mechanomyogram (MMG) and Vibromyography (VMG). Furthermore, the IMU data may also include movement data that was collected from video data, such as by using motion capture. Furthermore, the sensors 105 may also be remote sensors that are configured to collect the types of muscle data and IMU data described herein without touching the subject. In still further embodiments, one or more of the sensors 105 may be implanted in the subject and / or connect to one or more parts of the subject to collect the muscle data and IMU data.

[0037] Figure IB illustrates an example diagram of different muscles 110 (shown in Figure 1A) that may be monitored on a subject, in accordance with at least one embodiment. This list is exemplary7only and is not intended to be an exhaustive list of muscles 110 and / or muscle groups that may be used with the systems and methods described herein. In the head and neck age, muscles 110 include the Frontalis 1, the Masseter 2, and the Sternocleidomastoid 3. In the trunk of the body, the muscles 110 include the Pectoralis Major 4, the Serratus Anterior 5, the Rectus Abdominis 6, The Obliquis Extemus Abdominis 8, the Trapezius P. Descendenz (Upper) 9, the Trapezius P. Temsversus (Middle) 10, the Trapezius P. Ascendenz (Lower) 1 1, the Infraspinatus 12, the Latissius Dorsi 13, the Erector Spinae47735-3(Longissimus - Thorscic) 14, and the Erector Spinae (Ilicostalis - Lumbar) 15. The Arms include the Deltoideus P. Clavicularis (Anterior) 16. the Deltoideus P. Acromialis (medius) 17, the Deltoideus P. Scapularis (Posterior) 18, the Biceps Brachii (Long Head) 19, the Biceps Brachii (Short head) 20, the Triceps Brachii Lateral Head 21, the Triceps Brachii Medial Head 22, the Triceps Brachii Medial Head 23, the Brachioradialis 24. the Pronator Teres 25, the Flexor Carpi Radial 26. the Flexor Carpi Ulnaris 27, the Extensor Carpi Ulnaris 28, the Extensor Carpi Radialis Longus 29, the Extensor Carpi Radialis Brevis 30, the Extensor Digitorum 31, the Abductor Pollicis Longus 32, and the Extensor Pollicis Brevis 33. The legs include the abductors 34, the Rectus Femoris 35, the Vastus Medialis 36, the Vastus Lateralis 37. the Tibialis Anterior 38, the Tensor Fascia Latae 39, the Gluteus Medius 40, the Gluteus Maximus 41, the Biceps Femoris (Long Head) 42, the Semitendinosus / membranosus 43, the Gastrocnemius Medialis 44, the Gastrocnemius Lateralis 45, the Soleus 46, the Peroneus Brevis 47, and the Peroneus Longus 48. At each listed muscle 110 there are two dots to demonstrate the attachment points for the appropriate sensor 105. In some embodiments, the same sensor 105 is used for attaching to all of the listed muscles 110. In other embodiments, different sensors 105 may be used for different muscles 110 and / or different locations on the subject. One having ordinary skill in the art would understand that while the image includes two dots for the two points of contact on a sensor 105, other sensors 105 may have additional or fewer points of contact to the monitored muscle 110.

[0038] One having ordinary skill in the art would understand that while the systems and methods described herein relate to muscles and muscle groups, such as on humans, the systems and methods described herein would also work with other muscles than those listed and on animals.

[0039] Figure 2 illustrates an example muscle monitoring and analysis system 200 for monitoring and analyzing one or more muscles 110 (shown in Figure 1A) in a subject, in accordance with at least one embodiment. In the example embodiment, the system 200 includes a first sensor 205 and a second sensor 210 each configured to monitor a muscle 110 on a user 215 or subject. In the example embodiment, the first sensor 205 and the second sensor 210 are similar to the sensor47735-3105 (shown in Figure 1A). In the example embodiment, the first sensor 205 and the second sensor 205 are in wireless communication with a user computer device 220. The wireless communication may include, but is not limited to, one or more of Bluetooth, Near Field Communication (NFC), Zigbee, infrared transmission, Wi-Fi, and Ultra-Wideband (UWB). In the example embodiment, the user computer device 220 is in communication with a signal analysis computer device 225. In some embodiments, the signal analysis computer device 225 is a cloud-based server that is in communication with the user computer device 220. In other embodiments, the signal analysis computer device 225 is a part of the user computer device 220. In still further embodiments, the signal analysis computer device 225 is a computing device on the premises with the user computer device 220 and may be in communication with a plurality of user computing devices 220 on the premises.

[0040] In the example embodiment, the first sensor 205 and the second sensor 210 are attached to muscles 110 on the left and right sides of the user 215. In other embodiments, there may be a plurality of sensors 105 attached to the user 215 to monitor different muscles 110 of the user 215. In many embodiments, monitoring occurs when a user 215 is performing one or more sets of muscle movement, such as, but not limited to, during therapy, training, exercise, and / or other activities. A user 215 wears the sensors 105 affixed to their body as they train and the sensors 105 stream data to the user computer device 220 for real-time assessment of training intensity. This may occur as physiotherapy or physical training sessions with a healthcare provider, coach, or athletic trainer.

[0041] In the example embodiment, the user computer device 220 receives signals from the first sensor 205 and the second sensor 210. These signals may include EMG data and / or IMU data measured by the corresponding sensors 205 and 210.

[0042] In the example embodiment, the user computer device 220 and the signal analysis computer device 225 support layers of analysis that can process and normalize the complex signal of EMG and IMU concurrently while detecting and filtering out artifacts that contaminate this type of data. The system 200 also makes use47735-3of low level machines (LLMs) to partition and label sets of exercises for easy user interpretation for immediate in-session biofeedback and adjustments. This real-time feedback and analysis is a maj or advantage of the approached described herein.

[0043] Once data collection is complete, the data is uploaded to cloud systems, such as the signal analysis computer device 225. The signal analysis computer device 225 generates one or more detailed reports. This includes an in-depth analysis is performed to provide a repetition by repetition breakdown of the workout with sets of repetitions grouped together. Normally, this could take days or even weeks to perform by an EMG research laboratory. However, the systems and methods described herein allows the system 200 to process, partition, and create these comprehensive biomechanics reports. These reports can then be used by the mass market of users 215 require to make actionable decisions. Normally, only a miniscule portion of users are willing to take on the processing and interpretation.

[0044] The compiled data and reports are available for review by the users and other via one or more web portals. The signal analysis computer device 225 maintains an archive of data collections (workouts) for each individual user. This comprises overview trends of training periods over time and detailed repetition by repetition breakdowns of workouts. Users 215 can then use this information to review training regimens and program the next week or training block’s worth of training.

[0045] The user computer device 220 and / or the signal analysis computer device 225 execute a plurality of artificial intelligence / machine learning trained models. These models may include, but are not limited a preprocess Al 230, set detection Al 235, partition cycles Al 240, and / or classify cycles Al 245. These Al 230, 235.240, and 245 models may be executed on the user computer device 220 and / or the signal analysis computer device 225.

[0046] In the example embodiment, the preprocess Al 230 is trained to preprocess data from the sensors 105 to detect artifacts and remove noise. In some embodiments, this is similar to step 430 of process 400 (shown in Figure 4). In some embodiments, the preprocess Al 230 models described herein have layers that are47735-3configured to detect artifacts. For the purposes of this discussion, an artifact is a portion of an EMG signal that does not originate from the muscle 110. These artifacts can include movement artifacts, bias introduced by skin impedance, electrode placement, electromagnetic interference, etc.

[0047] As input, the preprocess Al 230 receives the EMG signal in time and frequency domain as received from the sensors 105. In some embodiments, the preprocess Al 230 performs normal filtering, which can include synchronization protocols and / or bandpass and high pass filters. In other embodiments, the preprocess Al 230 can also perform adaptive filtering based on what signal the user computer device 220 and / or the signal analysis computer device 225 sees (magnitude, frequency bandwidths, etc.). This includes, but is not limited to, dynamic bandpass filtering, dynamic thresholding, frequency domain analysis, signal decomposition, local min max detection, peak to peak detection, linear and polynomial regression, random forest, decision tree, and / or k means and clusters. The preprocess Al 230 outputs the start and end time stamps of artifact regions of data. When artifacts do not contaminate the signal to where a portion of the signal is in the operational range and distribution in bandwidth of EMG, the signal is decomposed into component signals and artifact bandwidths are taken out. The signal is then normalized to a control movement called the maximum voluntary contraction that is performed in every data collection. In at least one embodiment, the signal is normalized based on MVC (maximum voluntary contraction) data collection show n in step 420 (show n in Figure 4).

[0048] The set detection Al 235 is trained with EMG and IMU data. In some embodiments, this is similar to step 435 of process 400 (shown in Figure 4). In some embodiments, the set detection Al 235 is configured to detect sets (sense data between periods of rest within a period of data collection).

[0049] As input, the set detection Al 235 receives normalized EMG data cleared of artifacts from the preprocess Al 230. The set detection Al 235 performs one or more of dynamic thresholding, frequency domain analysis, local min max detection, peak to peak detection, linear and polynomial regression, hidden Markov model, support vector machines (SVM), and / or autocorrelators. The set detection Al47735-3235 outputs start and end time stamps of regions in which the test subject exerted muscle excitation (z.e., sets of a workout or physical activity). For example, if a workout calls for five sets of squat for ten repetitions each set, this layer of analysis detects the intervals of all five sets of squat containing ten squats each.

[0050] In some embodiments, the partition cycles Al 240 is configured to partition cycles of motion. As input, the partition cycles Al 240 receives the EMG and IMU data from the set detection Al 235. The partition cycles Al 240 performs one or more of local min max detection, peak to peak detection, linear and polynomial regression, exponential decay trimming, centroid mapping, random forest, decision tree, k means and clusters, extreme gradient boosting (XGBOOST), support vector machines, autocorrelators, hidden Markov model, deep neural network, convolution neural network, and / or reinforcement learning. The partiton cycles Al 240 outputs the start and end time stamps of every cycle of motion with extracted EMG metrics (z.e., average RMS, peak RMS, duration, median frequency, mean frequency, EMG rise / fall, etc.), and IMU metrics (z.e., periodicity, inflections in motion, average / peak acceleration and rotational velocity, acceleration / rotational velocity' rise / fall, etc.).

[0051] The classify cycles Al 245 classifies subject movement. In some embodiments, this is similar to step 440 of process 400 (shown in Figure 4). In some embodiments, the classify' cycles Al 245 is configured to classify' and identity' cycles of motion.

[0052] As input, the classify cycles Al 245 receives EMG and IMU data and features from cycles of motion from the partition cycles Al 240, as well as specific muscle 110 measured. The classify' cycles Al 245 performs one or more of local min max detection, peak to peak detection, centroid mapping, random forest, decision tree, k means and clusters, extreme gradient boosting (XGBOOST), hidden Markov model, support vector machines, autocorrelators, deep neural network, convolution neural network, and / or reinforcement learning. The classify' cycles Al 245 outputs classification and / or identification of movement into isometric, dynamic, endurance, or random movement. The classify- cycles Al 245 further narrows down the classification and / or identification to a specific movement. The classification and47735-3identification accuracy of the classify cycles Al 245 is increased if a pre-prescribed list of movements are provided by the user or previous data collections is known.

[0053] In some embodiments, these Al 230, 235, 240, and 245 are separate entities. In other embodiments, one or more of the aforementioned Al 230, 235, 240. and 245 are combined together. In some embodiments, these Al 230, 235.240, and 245 are executed on one of the user computer device 220 and / or the signal analysis computer device 225. In some embodiments, some of the Al are executed on the user computer device 220 (z.e., the preprocess Al 230), while other Al are executed by the signal analysis computer device 225.

[0054] One having ordinary skill in the art would understand that other t pes of sensors 105 may be used to collect other types of data to work with the systems and methods presented herein. More specifically, in addition to EMG data, other forms of muscle data could be collected, such as, but not limited to, Mechanomyogram (MMG) and Vibromyography (VMG). Furthermore, the IMU data may also include movement data that was collected from video data, such as by using motion capture. Furthermore, the sensors 105 may also be remote sensors that are configured to collect the types of muscle data and IMU data described herein without touching the subject 215. In still further embodiments, one or more of the sensors 105 may be implanted in the subject 215 and / or connect to one or more parts of the subject 215 to collect the muscle data and IMU data.

[0055] While Figure 2 only shows two sensors 105, first sensor 205 and second sensor 210, multiple sensors 105 may be attached to the user 215 and monitored as described herein.

[0056] Figure 3 illustrates an exemplary computer system 300 for realtime EMG analysis, in accordance with at least one embodiment. In the exemplary embodiment, the system 300 provides to near-real time wireless EMG data collection and analysis.

[0057] As described below in more detail, the signal analysis computer device 225 may be programmed for real-time EMG analysis. In some47735-3embodiments, the tower inspection computing device may be programmed to: a) receive 425 (shown in Figure 2) a plurality of sensor data including Electromyography (EMG) and inertial measurement unit (IMU) data from one or more sensors 105 (shown in Figure 1A) attached to a subject 215 (shown in Figure 2); b) preprocess 430 (shown in Figure 2) the plurality of sensor data to remove noise data from the plurality of sensor data; c) execute 435 (shown in Figure 2) at least one artificial intelligence (Al) model 510 (shown in Figure 5) trained with EMG and IMU data, wherein the EMG and IMU data of the plurality of sensor data are inputs for the at least one Al model 510, and wherein the at least one Al model 510 outputs a classification and an identification of a subject movement; and d) provide 455 (shown in Figure 4) feedback to the subject 215 based on the classification and identification of the subject movement.

[0058] In the example embodiment, user computer devices 220 are computers that include a web browser or a software application, which enables user computer devices 220 to communicate with signal analysis computer device 225 using the Internet, a local area network (LAN), or a wide area network (WAN). In some embodiments, the user computer devices 220 are communicatively coupled to the Internet through many interfaces including, but not limited to, at least one of a network, such as the Internet, a LAN, a WAN, or an integrated services digital network (ISDN), a dial-up-connection, a digital subscriber line (DSL), a cellular phone connection, a satellite connection, and a cable modem. User computer devices 220 can be any device capable of accessing a network, such as the Internet, including, but not limited to, a desktop computer, a laptop computer, a personal digital assistant (PDA), a cellular phone, a smartphone, a tablet, a phablet, wearable electronics, smart watch, virtual headsets or glasses (e.g., AR (augmented reality), VR (virtual reality), MR (mixed reality), or XR (extended reality) headsets or glasses), chat hots, voice hots, ChatGPT hots or ChatGPT-based hots, or other web-based connectable equipment or mobile devices.

[0059] In the example embodiment, the signal analysis computer device 225 is a computer that include a web brow ser or a software application, which enables the signal analysis computer device 225 to communicate with user computer devices 220 and other devices through various wared or wireless interfaces including47735-3without limitation a network, such as a local area network (LAN) or a wide area network (WAN), dial-in-connections, cable modems, Internet connection, wireless, and special high-speed Integrated Services Digital Network (ISDN) lines. Furthermore, signal analysis computer device 225 may include one or more artificial intelligence (Al) (e.g., preprocess Al 230, set detection Al 235, partition cycles Al 240, and classify cycles Al 245)(all shown in Figure 2) and / or an Al / deep learning module 510 (as shown in Figure 5) for training and / or updating the signal analysis computer device 225. In some embodiments, signal analysis computer device 225 may be implemented as a server computing device with artificial intelligence and deep learning functionality. In some of these embodiments, the signal analysis computer device 225 executes the signal analysis Al. In some embodiments, the signal analysis computer device 225 is communicatively coupled to the Internet through many interfaces including, but not limited to, at least one of a network, such as the Internet, a LAN, a WAN, or an integrated services digital network (ISDN), a dial-up-connection, a digital subscriber line (DSL), a cellular phone connection, a satellite connection, and a cable modem. The signal analysis computer device 225 can be any device capable of accessing a network, such as the Internet, including, but not limited to, a desktop computer, a laptop computer, a personal digital assistant (PDA), a cellular phone, a smartphone, a tablet, a phablet, wearable electronics, smart watch, virtual headsets or glasses (e.g., AR (augmented reality), VR (virtual reality), MR (mixed reality), or XR (extended reality) headsets or glasses), chat bots, voice bots, ChatGPT bots or ChatGPT-based bots, or other web-based connectable equipment or mobile devices.

[0060] A database server 305 is communicatively coupled to a database 310 that stores data. In one embodiment, the database 310 is a database that includes sensor information, user information, previous workout data, muscle data, and / or settings. In some embodiments, the database 310 is stored remotely from the signal analysis computer device 225. In some embodiments, the database 310 is decentralized. In the example embodiment, a person can access the database 310 via the user computer devices 220 by logging onto signal analysis computer device 225.

[0061] Sensors 205 and 210 may be any sensor and / or computer device to provide sensor information to signal analysis computer device 225 such as47735-3EMG and / or IMU data. In the example embodiment, the signal analysis computer device 225 is in communication with the sensors 205 and 210 attached to the user 215 (shown in Figure 2) . In some embodiments, the signal analysis computer device 225 is in direct communication with the sensors 205 and 210. In other embodiments, the signal analysis computer device 225 is in communication with the sensors 205 and 210 via one or more additional computer devices, such as, but not limited to, user computer device 220. The sensors 205 and 210 may be in communication with the user computer device 220 and / or the signal analysis computer device 225 via wireless communication such as through cable connection, Bluetooth, NFC, Zigbee, infrared transmission, WiFi, and Ultra-Wideband (UWB). In the example embodiment, sensors 205 and 210 are able to communicate with the signal analysis computer device 225 using the Internet, a local area network (LAN), or a wide area network (WAN). In some embodiments, the sensors 205 and 210 are communicatively coupled to the Internet through many interfaces including, but not limited to, at least one of a network, such as the Internet, a LAN, a WAN, or an integrated services digital network (ISDN), a dial-up-connection, a digital subscriber line (DSL), a cellular phone connection, a satellite connection, and a cable modem.

[0062] Figure 4 illustrates a flow diagram of a process 400 for realtime EMG analysis, in accordance with at least one embodiment of the disclosure. In the example embodiment, process 400 is performed by the user computer device 220 and / or signal analysis computer device 225 (both shown in Figure 2). In some of these embodiments, the signal analysis computer device 225 executes one or more the signal analysis AIs. While process 400 is written as the user computer device 220 is performing the steps, in some further embodiments, the signal analysis computer device 225 performs some or many of the steps of process 400. In many embodiments, process 400 occurs when a user 215 (shown in Figure 2) is performing one or more sets of muscle movement, such as, but not limited to, during therapy, training, exercise, and / or other activities. This may occur as physiotherapy or physical training sessions with a healthcare provider, coach, or athletic trainer.

[0063] In the example embodiment, a plurality of sensors 105 (shown in Figure 1 A) are connected 405 to the user 215 to monitor muscles 1 10 (shown in47735-3Figure 1A) of the user 215. The sensors 105 are affixed to a user's skin above the desired muscle 110 to be measured. While Figure 2 only shows two sensors 110, first sensor 205 and second sensor 210, multiple sensors 110 may be attached to the user 215 and monitored as described herein.

[0064] In the example embodiment, the sensors 105 are connected 410 to the user computer device 220. In the example embodiment, the sensors 105 are connected via a wireless connection, which may include, but is not limited to, one or more of Bluetooth, Near Field Communication (NFC), Zigbee, infrared transmission, Wi-Fi, and Ultra-Wideband (UWB). In the example embodiment, the sensors 105 are in communication with one or more applications or web services on the user computer device 220, which allows process 400 to proceed as described herein.

[0065] In the example embodiment, the user computer device 220 receives 415 identification markers including, but not limited to, muscles 110 that the sensor 105 is attached to, which side of the body the sensors 105 are attached to, name and / or other identification information for the user 215, and other information as needed. For example, the identification information may include height, weight, age, and other physical information about the user 215. In other examples, the user 215 has a profile on the user computer device 220 and the user computer device 220 retrieves the needed information based on the user’s name or profile name. These identification markers are used for tracking progress and statistics for each user 215. In the example embodiment, the tracking information is stored in the database 310 (shown in Figure 3) associated with the signal analysis computer device 225. In at least one embodiment, the database 310 is cloud-based. In a further embodiment, the database 310 is accessible by the user computer device 220 to provide information about the performance of the user 215 to the user 215.

[0066] In the example embodiment, the user computer device 220 performs 420 MVC (maximum voluntary contraction) data collection. In at least one embodiment, MVC of a muscle 110 is monitored for 45 seconds. This includes multiple repetitions of three to five second high effort isometric contractions with a single isolated movement that is resisted. This movement is easily repeatable movement for47735-3the particular muscle group being monitored. In at least one embodiment, the user computer device 220 performs 420 MVC for each muscle 110 or muscle group being tested by sensors 105. This includes left and right sides. In at least one embodiment, the MVC data collection is used for normalization of the sensor data from the corresponding muscles 110. Collecting MVC at the start allows each consecutive set to be normalized. This allows for accurate decision making in real time by the user 215.

[0067] In the example embodiment, the user computer device 220 receives 425 a plurality of sensor data. The sensors 105 monitor muscle patterning (clinical grade electromyography - EMG) and movement (accelerometer + gyroscope IMU system). Muscle patterning (EMG) and movement (IMU) data is collected by the sensors 105 attached to the user’s skin. This data is streamed in real-time to the user computer device 220. The EMG and IMU data may be partitioned together or separately. The movement data is derived from devices such as, but not limited to, accelerometers, gyroscopes, and magnetometers. The movement data may include, but is not limited to, acceleration, rotational velocity / speed, and bearing / course. The muscle excitation / activation data includes surface or intramuscular electromyography, which may be collectively referred to as EMG herein. As described herein, this EMG and IMU data is used to measure performance, rehabilitation, or assessment of human movement in various users, such as, but not limited to, athlete, patient, game player, etc.

[0068] In the example embodiment, the user computer device 220 preprocesses 430 the plurality of sensor data to remove noise. To initially clean the data, a multitude of adaptive fdters are used to filter out movement artifacts based on data found in abnormal bandwidths of biological data. The data is then normalized to the statistical p-end of highest RMS values. In some embodiments, the normalization is based on the MVC data. This preprocessing 430 is done dynamically in real-time.

[0069] In the example embodiment, the user computer device 220 executes 435 at least one trained artificial intelligence (Al) models to output a classification and an identification of the user’s movement related to the plurality of47735-3sensor data. The at least one model has been trained with EMG and IMU data. The at least one model is trained to receive the EMG and IMU data of the plurality of sensor data as inputs and the at least one model outputs a classification and an identification of the user’s movement.

[0070] To partition movement data, a combination of machine learning and regression models (i.e., autoencoders, hidden Markov models, autocorrelators, k-means, nearest neighbor, random forest, decision trees, neural networks, linear / logarithmic models, cluster models, and support vectors, etc.) are used to determine the presence of periodicity or major perturbations in movement patterns. Periods of repetitive periodicity or major perturbations in movement patterns are then partitioned into shorter segments of "sense" data and all "nonsense" data (periods of rest) is omitted from the rest of analysis.

[0071] These short sense movement data segments are then isolated and a combination of machine learning and regression models (i.e., autoencoders, hidden Markov models, autocorrelators, k-means, nearest neighbor, random forest, decision trees, neural networks, linear / logarithmic models, cluster models, and support vectors, etc.) are used to determine the periodicity of movement to estimate the number of repetitions of a movement that have been performed based on detection of repeated subunits of movement. Based on feature extractions such as duration of each period of motion, magnitude of acceleration, rotational velocity, jerk, etc., these movements can be classified based on descriptors and identified (i.e., squat, lunge, jump, complex, explosive movements, endurance, walking, cycling, sprinting, running, etc.).

[0072] To partition muscle data, a combination of machine learning and regression models (i.e., autoencoders, hidden Markov models, autocorrelators, k-means, nearest neighbor, random forest, decision trees, neural networks, linear / logarithmic models, cluster models, and support vectors, etc.) are used to detect periods in time that the user is exerting significant effort using their muscles. Periods of significant muscle patterning or exertion are then partitioned into shorter segments of "sense" data and all "nonsense" data is omitted from the rest of analysis.47735-3

[0073] These short sense muscle data segments are then isolated and a combination of machine learning and regression models (i.e., autoencoders, hidden Markov models, autocorrelators, k-means, nearest neighbor, random forest, decision trees, neural networks, linear / logarithmic models, cluster models, and support vectors, etc.) are used to detect all muscle contractions performed during the period of time. Based on feature extractions such as duration, magnitude of the signal, frequency bandwidth, and power differential in spectral diagrams, these muscle contractions can be grouped together and be classified based on descriptors (i.e., speed, power, strength, endurance, isometric, concentric, eccentric, slow twitch, fast twitch, etc.).

[0074] In the example embodiment, the user computer device 220 generates 445 real-time user feedback. The user computer device 220 also displays 450 the real-time feedback via UI to the user 215. The user computer device 220 maps the short, isolated segments of muscle data and movement data on top of each other to determine muscle exertion metrics (i.e., intensity, time under tension, contraction duration, work duration, performance, fatigue, etc.) for the entirety of the workout or for these short segments.

[0075] To isolate and determine muscle exertion metrics (i.e.. intensity, time under tension, contraction duration, work duration, performance, fatigue, etc.) for each individual repetition of movement per given short segment, a combination of machine learning and regression models (i.e., autoencoders, hidden Markov models, autocorrelators, k-means, nearest neighbor, random forest, decision trees, neural networks, linear / logarithmic models, cluster models, and support vectors, etc.) are used to gather individual muscle contractions into a muscle pattern correlated with a unit of movement (or repetition).

[0076] In the event that there is no periodicity but significant muscle excitation detected (this indicates an isometric, holding contraction), additional algorithms (i.e., greedy algorithms, gathering algorithms, k-means, random forest distributions, clustering, decision trees, etc.) are used to gather individual muscle contractions together into gross muscle contractions for analysis.47735-3

[0077] This repetition by repetition data is aggregated together into comprehensive, bottom up reports and observed longitudinally for many training cycles (z.e., months, even years). Trend analyses using regression models are then used to determine if progress is being made. Generative Al techniques are also used to recommend movements / training schemes based on previous trends that elicited desired responses.

[0078] In some embodiments, the Real-time Feedback includes data collection for the current session. This includes displaying 450 real-time muscle excitation, where the UI may use symbols, such as. but not limited to, bars and / or circles, to provide representation of max and average values during the session. This also allows for set by set muscle to muscle comparison and set by set data display, which may be characterized by intensity aka EMG RMS (Root Mean Square) values. Other feedback data can include time under tension, aka the duration of muscle excitation during movement (secs).

[0079] As a part of processing, session data is uploaded to a web portal, where the user 215 has access to all of the data.

[0080] For the comparison feedback and trend analysis, the signal analysis computer device 225 can provide information, including, but not limited to, session summaries, report downloads, and detailed session summaries. These session summaries and / or reports include trend data of muscles 110, same set by set data from the user computer device, and rep by rep data.

[0081] In some embodiments, the signal analysis computer device 225 can be used to create sport-specific protocol libraries that include a library of EMG-guided training and rehab protocols for specific sports or activities. These protocols can include: a) muscle activation targets for key movements (e.g., sprinting, throw ing): normative ranges for every listed exercise and muscle 110; b) recovery' strategies informed by fatigue patterns; c) color coded flags when there is a deviation from the norm (+ or -); d) training drills that maximize efficiency based on muscle group engagement; and e) the most effective exercises for each muscle.47735-3

[0082] In some further embodiments, the signal analysis computer device 225 can be used to create retum-to-play decision-making algorithms. These algorithms can be used to assess an athlete’s readiness to return to play. Features could include: a) comparative data against baseline EMG readings; % of normative ranges OR athlete baseline; b) automated scoring for dynamic movements or sport-specific drills (thresholds); c) setting thresholds (minimum and maximum) for running and throwing per muscle; d) alerts for discrepancies between historical or normative intensity; and / or e) color coded flags same as “recovery strategies” to immediately identify fatigue.

[0083] In some further embodiments, the signal analysis computer device 225 can be used to create dynamic biofeedback for real-time correction. This includes visual biofeedback to correct movement patterns during exercise or therapy. For example, visual biofeedback to correct movement patterns during exercise or therapy. This includes a) real-time feedback to ensure balanced muscle activation during rehab; b) implementation of minimum and maximum thresholds to ensure prescribed %MVC intensities; and / or c) setting thresholds same as “automated scoring.” This may also include gamified training solutions, which include developing gamification elements for practitioners and athletes to: a) increase engagement during training by setting EMG-based challenges (e.g., muscle endurance or activation targets); b) provide a green flash if the subject beats their highest threshold RMS and / or setting thresholds for HIMA (Holding Isometric Muscle Action) exercises (minimum and maximum); c) create competitive scenarios for individuals or teams based on EMG metrics; and / or d) % of the normative population (i.e., ranking).

[0084] Other provided information includes: a) Al generated suggestions based on data (e.g., - point out potential form adjustments, suggest lesser weight, etc.); b) local storage until service (Cellular or WiFi) is available for data submission / upload; c) trend review' - historical movements (e.g., - MVC); d) movement name suggestions based on data collected (e.g., - suggest squat); d) integration with other APIs (other equipment / devices); e) facilitate improved research for industry -faster, repetitive, accurate data; and / or f) movement detection.47735-3

[0085] In the exemplary embodiment, the user computer device 220 and / or the signal analysis computer device 225 receives 425 a plurality of sensor data including Electromyography (EMG) and inertial measurement unit (IMU) data from one or more sensors 105 attached to a subject 215. In some embodiments, the plurality7of sensor data is generated by pairs of sensors 105 attached to the subject 215. The pairs of sensors 105 are placed with one sensor 105 is placed on a muscle 110 on a left side and another sensor 105 is placed on a matching muscle 110 a right side of the subject 215. In at least one embodiment, the one or more sensors 105 are attached to the epidermis above a muscle 110 to be monitored.

[0086] In the exemplary embodiment, the user computer device 220 and / or the signal analysis computer device 225 preprocesses 430 the plurality7of sensor data to remove noise data from the plurality7of sensor data. In at least one embodiment, the user computer device 220 and / or the signal analysis computer device 225 executes a trained Al model 235 to preprocess 430 the plurality of sensor data to remove noise data from the plurality of sensor data. The trained Al model 230 receives the plurality of sensor data as input and output locations of artifact regions of data in the plurality7of sensor data. The user computer device 220 and / or the signal analysis computer device 225 removes the artifact regions of data from the plurality of sensor data.

[0087] In the exemplary7embodiment, the user computer device 220 and / or the signal analysis computer device 225 executes 435 at least one artificial intelligence (Al) model 230, 235, 240. and 245 trained with EMG and IMU data. The EMG and IMU data of the plurality of sensor data are inputs for the at least one Al model 230, 235, 240, and 245. The at least one Al model 230, 235, 240, and 245 outputs a classification and / or an identification of a subject movement. The user computer device 220 and / or the signal analysis computer device 225 identifies sets of activity7by the subject from the plurality of sensor data. The user computer device 220 and / or the signal analysis computer device 225 identifies cycles of motion from the plurality of sensor data.

[0088] In the exemplary embodiment, the user computer device 220 and / or the signal analysis computer device 225 provide 450 feedback to the subject47735-3215 based on the classification and / or identification of the subject movement. The user computer device 220 and / or the signal analysis computer device 225 extracts EMG metrics from the plurality of sensor data. The user computer device 220 and / or the signal analysis computer device 225 extract IMU metrics from the plurality of sensor data.

[0089] In some embodiments, the user computer device 220 and / or the signal analysis computer device 225 normalizes the plurality of sensor data with MVC (maximum voluntary contraction) data collected at the beginning of a session of subject movement.

[0090] In some embodiments, the user computer device 220 and / or the signal analysis computer device 225 identifies a plurality' of activities performed by the subject based on the plurality of sensor data. The identification of the plurality’ of activities is further based on an identified muscle.

[0091] In some embodiments, the user computer device 220 and / or the signal analysis computer device 225 receives identification markers for each sensor 105 of the one or more sensors 105. In some embodiments, the feedback to the subject 215 includes real-time intensity information as subject movement is occurring.

[0092] In some embodiments, the feedback to the subject includes a report of subject movements that were monitored by the one or more sensors 105. In some embodiments, the feedback to the subject includes trend analysis over a plurality of sessions of subject movement.

[0093] One having ordinary skill in the art would understand that other types of sensors 105 may' be used to collect other types of data to yvork with the systems and methods presented herein. More specifically, in addition to EMG data, other forms of muscle data could be collected, such as, but not limited to, Mechanomyogram (MMG) and Vibromyography (VMG). Furthermore, the IMU data may also include movement data that was collected from video data, such as by using motion capture. Furthermore, the sensors 105 may also be remote sensors that are configured to collect the types of muscle data and IMU data described herein without touching the subject47735-3215. In still further embodiments, one or more of the sensors 105 may be implanted in the subject 215 and / or connect to one or more parts of the subject 215 to collect the muscle data and IMU data.

[0094] Figure 5 depicts an artificial intelligence (Al)Zdeep learning (DL) module 510, according to an embodiment. In some embodiments. AI / DL module 510 includes a training set builder module 502 programmed to submit one or more queries to database 320 (show n in Figure 3) to retrieve data and / or subsets of data, and to use those subsets to build training data sets for generating predictive models 508. In some embodiments, AI / DL module 510 may be similar to one or more of preprocess Al 230, set detection Al 235, partition cycles Al 240, and classify cycles Al 245 (all shown in Figure 2).

[0095] In example embodiments, training set builder module 502 is programmed to retrieve training data sets from the retrieved subsets. Each training data set corresponds to detected issues and completed solutions. “Historical” in this context may be defined as previously determined risk, as opposed to risk assessment completed in real-time with respect to the time of retrieval by training set builder module 502. Each training data set includes “model input” data fields along w ith at least one “result” data field including issue data. The model input data fields represent factors that may be expected to, or unexpectedly be found during model training to have some correlation with other detected issues and completed solutions.

[0096] After training set builder module 502 generates training data sets, training set builder module 502 passes the training data sets to model trainer module 504. In some embodiments, model trainer module 504 is programmed to apply the model input data fields of each training data set as inputs to one or more machine learning models. Each of the one or more machine learning models is programmed to produce, for each training data set, at least one output intended to correspond to, or “predict,” a value of the at least one result data field of the training data set. “Machine learning” refers broadly to various algorithms that may be used to train the model to identify and recognize patterns in existing data in order to facilitate making predictions for subsequent new input data.47735-3

[0097] Model trainer module 504 is programmed to compare, for each training data set, the at least one output of the model to the at least one result data field of the training data set and apply a machine learning algorithm to adjust parameters of the model in order to reduce the difference or “error” between the at least one output and the corresponding at least one result data field. In this way, model trainer module 504 trains the machine learning model to accurately identify issues. In other words, model trainer module 504 cycles the one or more machine learning models through the training data sets, causing adjustments in the model parameters, until the error between the at least one output and the detected issues and completed solutions falls below a suitable threshold, and then uploads at least one trained machine learning model to predictive model module 508 for application to new issue detection.

[0098] In some embodiments, the one or more machine learning models may include one or more neural networks, such as a convolutional neural network, a deep learning neural network, or the like. The neural network may have one or more layers of nodes, and the model parameters adjusted during training may be respective weight values applied to one or more inputs to each node to produce a node output. In other words, the nodes in each layer may receive one or more inputs and apply a weight to each input to generate a node output. The node inputs to the first layer may correspond to the model input data fields, and the node outputs of the final layer may correspond to the at least one output of the model, intended to predict the at least one result data field. One or more intermediate layers of nodes may be connected between the nodes of the first layer and the nodes of the final layer. As model trainer module 504 cycles through the training data sets, model trainer module 504 applies a suitable backpropagation algorithm to adjust the weights in each node layer to minimize the error between the at least one output and the corresponding result data field. In this fashion, the machine learning model is trained to produce one or more outputs which reliably detect issues and provide solutions when processed through generative models. Alternatively, the machine learning model has any suitable structure. In some embodiments, model trainer module 504 provides an advantage by automatically discovering and properly weighting complex, second- or third-order, and / or otherwise nonlinear interconnections between the model input data fields and the at least one47735-3output. Absent the machine learning model, such connections are unexpected and / or undiscoverable by human analysts.

[0099] Additionally, or alternatively, the one or more machine learning models may include one or more multilayer perceptron (MLP) classifiers. A MLP classifier may comprise input and output layers, and one or more hidden layers with many neurons stacked together.

[0100] Additionally, or alternatively, the one or more machine learning models may include one or more support vector machines (SVMs). SVMs are supervised learning models with associated learning algorithms that analyze data for classification and regression analysis. More particularly, a SVM constructs a hyperplane or set of hyperplanes in a high or infinite-dimensional space, which can be used for classification, identification, regression, or other tasks like outlier detection.

[0101] In some embodiments, predictive model module 508 compares the data with the output from the trained model, and routes the comparison result to a model updater module 506 of AI / DL module 510. Model update module 510 is programmed to derive a correction signal from the comparison results, and to provide correction signal to model trainer module 504 to enable updating or '‘re-training” of the at least one machine learning model to improve performance. The retrained machine learning model may be periodically re-uploaded to predictive model module 508.

[0102] In some embodiments, model trainer module 504 may update the training dataset by creating one or more new historical records which includes new data and re-training the operator model using the updated training dataset, further improving the accuracy of the operator model.

[0103] A Risk prediction module may employ AI / DL module 510 to use the trained model to predict the appropriate solutions to use. The issues, solutions, and other data may be viewable via the user computer device 220 (shown in Figure 2).

[0104] In some embodiments, the user computer device 220 and / or signal analysis computer device 225 (shown in Figure 2) includes an application47735-3module is configured to facilitate maintaining a data management application and providing the functionality thereof to users. The application module may store instructions that enable the download and / or execution of data management application at user computer devices 220 and / or signal analysis computer device 225. The application module may store instructions regarding user interfaces, controls, commands, settings, and the like, and may format data into a format suitable for transmitting to user computer devices 220 for display thereof.

[0105] Figure 6 depicts an exemplary' configuration of user computer device 602, in accordance with one embodiment of the present disclosure. In the exemplary embodiment, user computer device 602 may be similar to, or the same as, user computer device 220 (shown in Figure 2). User computer device 602 may be operated by a user 601.

[0106] User computer device 602 may include a processor 605 for executing instructions. In some embodiments, executable instructions may be stored in a memory' area 610. Processor 605 may include one or more processing units (e.g., in a multi-core configuration). Memory- area 610 may be any device allowing information such as executable instructions and / or transaction data to be stored and retrieved. Memory area 610 may include one or more computer readable media.

[0107] User computer device 602 may also include at least one media output component 615 for presenting information to user 601. Media output component 615 may be any component capable of conveying information to user 601. In some embodiments, media output component 615 may include an output adapter (not shoyvn) such as a video adapter and / or an audio adapter. An output adapter may be operatively coupled to processor 605 and operatively couplable to an output device such as a display device (e.g., a cathode ray tube (CRT), liquid crystal display (LCD), light emitting diode (LED) display, or “electronic ink” display) or an audio output device (e.g., a speaker or headphones).

[0108] In some embodiments, media output component 615 may be configured to present a graphical user interface (e.g., a web broyvser and / or a client47735-3application) to user 601. A graphical user interface may include, for example, an interface for viewing items of information provided by the signal analysis computer device 225 (shown in Figure 2). In some embodiments, user computer device 602 may include an input device 620 for receiving input from user 601. User 601 may use input device 620 to, without limitation, provide information either through speech or typing.

[0109] Input device 620 may include, for example, a keyboard, a pointing device, a mouse, a sty lus, a touch sensitive panel (e.g., a touch pad or a touch screen), a gyroscope, an accelerometer, a position detector, a biometric input device, and / or an audio input device. A single component such as a touch screen may function as both an output device of media output component 615 and input device 620.

[0110] User computer device 602 may also include a communication interface 625, communicatively coupled to a remote device such as signal analysis computer device 225. Communication interface 625 may include, for example, a wired or wireless network adapter and / or a wireless data transceiver for use with a mobile telecommunications netw ork.

[0111] Stored in memory area 610 are, for example, computer readable instructions for providing a user interface to user 601 via media output component 615 and, optionally, receiving and processing input from input device 620. A user interface may include, among other possibilities, a web browser and / or a client application. Web browsers enable users, such as user 601, to display and interact with media and other information typically embedded on a web page or a website from the signal analysis computer device 225. A client application may allow7user 601 to interact with, for example, the signal analysis computer device 225. For example, instructions may be stored by a cloud service, and the output of the execution of the instructions sent to the media output component 615.

[0112] Figure 7 depicts an exemplary configuration of a server computer device 701, in accordance with one embodiment of the present disclosure. In the exemplary embodiment, server computer device 701 may be similar to, or the same as, signal analysis computer device 225 (shown in Figure 2) and database server 30547735-3(shown in Figure 3). Server computer device 701 may also include a processor 705 for executing instructions. Instructions may be stored in a memory area 710. Processor 705 may include one or more processing units (e g., in a multi-core configuration).

[0113] Processor 705 may be operatively coupled to a communication interface 715 such that server computer device 701 is capable of communicating with a remote device such as another server computer device 701, signal analysis computer device 225, sensors 205 and 210 (shown in Figure 2), and user computer devices 220 (shown in Figure 2) (for example, using wireless communication or data transmission over one or more radio links or digital communication channels). For example, communication interface 715 may receive input from user computer devices 220 via the Internet, as illustrated in Figure 3.

[0114] Processor 705 may also be operatively coupled to a storage device 725. Storage device 725 may be any computer-operated hardware suitable for storing and / or retrieving data, such as, but not limited to, data associated with one or more models. In some embodiments, storage device 725 may be integrated in server computer device 701. For example, server computer device 701 may include one or more hard disk drives as storage device 725.

[0115] In other embodiments, storage device 725 may be external to server computer device 701 and may be accessed by a plurality' of server computer devices 701. For example, storage device 725 may include a storage area network (SAN), a network attached storage (NAS) system, and / or multiple storage units such as hard disks and / or solid-state disks in a redundant array of inexpensive disks (RAID) configuration.

[0116] In some embodiments, processor 705 may be operatively coupled to storage device 725 via a storage interface 720. Storage interface 720 may be any component capable of providing processor 705 w ith access to storage device 725. Storage interface 720 may include, for example, an Advanced Technology Attachment (ATA) adapter, a Serial ATA (SATA) adapter, a Small Computer System47735-3Interface (SCSI) adapter, a RAID controller, a SAN adapter, a network adapter, and / or any component providing processor 705 with access to storage device 725.

[0117] Processor 705 may execute computer-executable instructions for implementing aspects of the disclosure. In some embodiments, the processor 705 may be transformed into a special purpose microprocessor by executing computerexecutable instructions or by otherwise being programmed. For example, the processor 705 may be programmed with the instruction such as illustrated in Figure 4.MACHINE LEARNING AND OTHER MATTERS

[0118] The computer-implemented methods discussed herein may include additional, less, or alternate actions, including those discussed elsewhere herein. The methods may be implemented via one or more local or remote processors, transceivers, servers, and / or sensors (such as processors, transceivers, servers, and / or sensors mounted on vehicles or mobile devices, or associated with smart infrastructure or remote servers), and / or via computer-executable instructions stored on non-transitory computer-readable media or medium.

[0119] In some embodiments, the signal analysis computer device 225 is configured to implement machine learning, such that the signal analysis computer device 225 “learns” to analyze, organize, and / or process data without being explicitly programmed. Machine learning may be implemented through machine learning methods and algorithms (“ML methods and algorithms”). In an exemplary embodiment, a machine learning module (“ML module”) is configured to implement ML methods and algorithms. In some embodiments, ML methods and algorithms are applied to data inputs and generate machine learning outputs (“ML outputs”). Data inputs may include but are not limited to images, text data, and / or other t pes of data (i.e., multi-modal type of data). ML outputs may include, but are not limited to identified objects, items classifications, items identification, textual product, and / or other data extracted from the images or textual data. In some embodiments, data inputs may include certain ML outputs (i.e., overall convergence optimization parameters or multiple localized convergence points that lack an optimal convergence point).47735-3

[0120] In some embodiments, at least one of a plurality of ML methods and algorithms may be applied, which may include but are not limited to: linear or logistic regression, instance-based algorithms, regularization algorithms, decision trees, Bayesian networks, cluster analysis, association rule learning, artificial neural networks, deep learning, combined learning, reinforced learning, dimensionality reduction, and support vector machines. In various embodiments, the implemented ML methods and algorithms are directed toward at least one of a plurality of categorizations of machine learning, such as supervised learning, unsupervised learning, and reinforcement learning.

[0121] In one embodiment, the ML module employs supervised learning, which involves identifying patterns in existing data to make predictions about subsequently received data. Specifically, the ML module is “trained” using training data, which includes example inputs and associated example outputs. Based upon the training data, the ML module may generate a predictive function which maps outputs to inputs and may utilize the predictive function to generate ML outputs based upon data inputs. The example inputs and example outputs of the training data may include any of the data inputs or ML outputs described above. In the exemplary embodiment, a processing element may be trained by providing it with a large sample of text with known characteristics or features. Such information may include, for example, information associated with a plurality of text of a plurality of different towers, mounts, and / or radios.

[0122] In another embodiment, a ML module may employ unsupervised learning, which involves finding meaningful relationships in unorganized data. Unlike supervised learning, unsupervised learning does not involve user-initiated training based upon example inputs with associated outputs. Rather, in unsupervised learning, the ML module may organize unlabeled data according to a relationship determined by at least one ML method / algorithm employed by the ML module. Unorganized data may include any combination of data inputs and / or ML outputs as described above.47735-3

[0123] In yet another embodiment, a ML module may employ reinforcement learning, which involves optimizing outputs based upon feedback from a reward signal. Specifically, the ML module may receive a user-defined reward signal definition, receive a data input, utilize a decision-making model to generate a ML output based upon the data input, receive a reward signal based upon the reward signal definition and the ML output, and alter the decision-making model so as to receive a stronger reward signal for subsequently generated ML outputs. Other types of machine learning may also be employed, including deep or combined learning techniques.

[0124] In some embodiments, generative artificial intelligence (Al) models (also referred to as generative machine learning (ML) models) may be utilized with the present embodiments and may the voice bots or chatbots discussed herein may be configured to utilize artificial intelligence and / or machine learning techniques. For instance, the voice or chatbot may be a ChatGPT chatbot. The voice or chatbot may employ supervised or unsupervised machine learning techniques, which may be followed by, and / or used in conjunction with, reinforced or reinforcement learning techniques. The voice or chatbot may employ the techniques utilized for ChatGPT. The voice bot, chatbot, ChatGPT-based bot, ChatGPT bot, and / or other bots may generate audible or verbal output, text or textual output, visual or graphical output, output for use with speakers and / or display screens, and / or other types of output for user and / or other computer or bot consumption.

[0125] Based upon these analyses, the processing element may leam how to identify issues with mounted equipment and patterns to find solutions to rectify those issues. The processing element may also leam how to identify attributes of different issues and images. This information may be used to determine which categories to apply to those issues .ADDITIONAL CONSIDERATIONS

[0126] As will be appreciated based upon the foregoing specification, the above-described embodiments of the disclosure may be implemented using computer programming or engineering techniques including computer software,47735-3firmware, hardware or any combination or subset thereof. Any such resulting program, having computer-readable code means, may be embodied or provided within one or more computer-readable media, thereby making a computer program product, i.e., an article of manufacture, according to the discussed embodiments of the disclosure. The computer-readable media may be, for example, but is not limited to, a fixed (hard) drive, diskette, optical disk, magnetic tape, semiconductor memory such as read-only memory (ROM), and / or any transmitting / receiving medium such as the Internet or other communication network or link. The article of manufacture containing the computer code may be made and / or used by executing the code directly from one medium, by¬ copying the code from one medium to another medium, or by transmitting the code over a network.

[0127] These computer programs (also known as programs, software, software applications, “apps,” or code) include machine instructions for a programmable processor and can be implemented in a high-level procedural and / or object-oriented programming language, and / or in assembly / machine language. As used herein, the terms “machine-readable medium” “computer-readable medium” refers to any computer program product, apparatus and / or device (e g., magnetic discs, optical disks, memory, Programmable Logic Devices (PLDs)) used to provide machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The “machine-readable medium” and “computer-readable medium,” however, do not include transitory signals. The term “machine-readable signal” refers to any signal used to provide machine instructions and / or data to a programmable processor.

[0128] As used herein, the terms “processor” and “computer” and related terms, e.g., “processing device”, “computing device”, and “controller” are not limited to just those integrated circuits referred to in the art as a computer, but broadly refers to a microcontroller, a microcomputer, a programmable logic controller (PLC), a reduced instruction set circuit (RISC), an application specific integrated circuit (ASIC), logic circuits, and any other circuit or processor capable of executing the functions described herein. The above examples are example only and are thus not intended to limit in any way the definition and / or meaning of the term “processor.”47735-3

[0129] As used herein, the terms “software” and “firmware” are interchangeable, and include any computer program stored in memory for execution by a processor, including RAM memory . ROM memory, EPROM memory, EEPROM memory, and non-volatile RAM (NVRAM) memory7. The above memory ty pes are example only, and are thus not limiting as to the ty pes of memory usable for storage of a computer program.

[0130] As used herein, the term “database” can refer to either a body of data, a relational database management system (RDBMS), or to both. As used herein, a database can include any collection of data including hierarchical databases, relational databases, flat file databases, object-relational databases, object-oriented databases, and any other structured collection of records or data that is stored in a computer system. The above examples are example only, and thus are not intended to limit in any way the definition and / or meaning of the term database. Examples of RDBMS’ include, but are not limited to including, Oracle® Database, MySQL, IBM® DB2, Microsoft® SQL Server, Sybase®, and PostgreSQL. Elowever, any database can be used that enables the systems and methods described herein. (Oracle is a registered trademark of Oracle Corporation, Redwood Shores, California; IBM is a registered trademark of International Business Machines Corporation, Armonk. New York; Microsoft is a registered trademark of Microsoft Corporation, Redmond, Washington; and Sybase is a registered trademark of Sybase, Dublin, California.)

[0131] In another example, a computer program is provided, and the program is embodied on a computer-readable medium. In an example, the system is executed on a single computer system, without requiring a connection to a server computer. In a further example, the system is being run in a Windows® environment (Windows is a registered trademark of Microsoft Corporation, Redmond, Washington). In yet another example, the system is run on a mainframe environment and a UNIX® server environment (UNIX is a registered trademark of X / Open Company Limited located in Reading, Berkshire, United Kingdom). In a further example, the system is run on an iOS® environment (iOS is a registered trademark of Cisco Sy stems, Inc. located in San Jose, CA). In yet a further example, the system is run on a Mac OS® environment (Mac OS is a registered trademark of Apple Inc. located in Cupertino,47735-3CA). In still yet a further example, the system is run on Android® OS (Android is a registered trademark of Google, Inc. of Mountain View, CA). In another example, the system is run on Linux® OS (Linux is a registered trademark of Linus Ton aids of Boston, MA). The application is flexible and designed to run in various different environments without compromising any major functionality.

[0132] As used herein, an element or step recited in the singular and proceeded with the word “a” or “an” should be understood as not excluding plural elements or steps, unless such exclusion is explicitly recited. Furthermore, references to “example” or “one example” of the present disclosure are not intended to be interpreted as excluding the existence of additional examples that also incorporate the recited features. Further, to the extent that terms “includes,” “including,” “has,” “contains,” and variants thereof are used herein, such terms are intended to be inclusive in a manner similar to the term “comprises” as an open transition word without precluding any additional or other elements.

[0133] Furthermore, as used herein, the term “real-time” refers to at least one of the time of occurrence of the associated events, the time of measurement and collection of predetermined data, the time to process the data, and the time of a system response to the events and the environment. In the examples described herein, these activities and events occur substantially instantaneously.

[0134] In some embodiments, the system includes multiple components distributed among a plurality of computer devices. One or more components may be in the form of computer-executable instructions embodied in a computer-readable medium. The systems and processes are not limited to the specific embodiments described herein. In addition, components of each system and each process can be practiced independent and separate from other components and processes described herein. Each component and process can also be used in combination with other assembly packages and processes. The present embodiments may enhance the functionality and functioning of computers and / or computer systems.47735-3

[0135] The computer-implemented methods discussed herein can include additional, less, or alternate actions, including those discussed elsewhere herein. The methods can be implemented via one or more local or remote processors, transceivers, servers, and / or sensors (such as processors, transceivers, servers, and / or sensors mounted on vehicles or mobile devices, or associated with smart infrastructure or remote servers), and / or via computer-executable instructions stored on non-transitory computer-readable media or medium. Additionally, the computer systems discussed herein can include additional, less, or alternate functionality, including that discussed elsewhere herein. The computer systems discussed herein can include or be implemented via computer-executable instructions stored on non-transitory computer-readable media or medium.

[0136] As used herein, the term “non-transitory computer-readable media” is intended to be representative of any tangible computer-based device implemented in any method or technology for short-term and long-term storage of information, such as, computer-readable instructions, data structures, program modules and sub-modules, or other data in any device. Therefore, the methods described herein can be encoded as executable instructions embodied in a tangible, non-transitory, computer readable medium, including, without limitation, a storage device and / or a memory device. Such instructions, when executed by a processor, cause the processor to perform at least a portion of the methods described herein. Moreover, as used herein, the term “non-transitory computer-readable media” includes all tangible, computer-readable media, including, without limitation, non-transitory computer storage devices, including, without limitation, volatile and nonvolatile media, and removable and nonremovable media such as a firmware, physical and virtual storage, CD-ROMs, DVDs, and any other digital source such as a network or the Internet, as well as yet to be developed digital means, with the sole exception being a transitory-, propagating signal.

[0137] The patent claims at the end of this document are not intended to be construed under 35 U.S.C. § 112(f) unless traditional means-plus-function language is expressly recited, such as “means for” or “step for” language being expressly recited in the claim(s).47735-3

[0138] This writen description uses examples to disclose the disclosure, including the best mode, and also to enable any person skilled in the art to practice the disclosure, including making and using any devices or systems and performing any incorporated methods. The patentable scope of the disclosure is defined by the claims, and may include other examples that occur to those skilled in the art. Such other examples are intended to be within the scope of the claims if they have structural elements that do not differ from the literal language of the claims, or if they include equivalent structural elements with insubstantial differences from the literal language of the claims.

Claims

47735-3WHAT IS CLAIMED IS:

1. A computer device comprising at least one processor in communication with at least one memory device, wherein the at least one processor is programmed to:receive a plurality of sensor data including muscle data and inertial measurement unit (IMU) data from one or more sensors monitoring a subject;preprocess the plurality of sensor data to remove noise data from the plurality of sensor data;execute at least one artificial intelligence (Al) model trained with muscle data and IMU data, wherein the muscle data and IMU data of the plurality of sensor data are inputs for the at least one Al model, and wherein the at least one Al model outputs a classification and an identification of a subject movement; and provide feedback to the subject based on the classification and identification of the subject movement.

2. The computing device of Claim 1, wherein the muscle data includes one or more of Electromyography (EMG) data,-Vibromyography (VMG) data, and Mechanomyogram (MMG) data.

3. The computing device of Claim 1, wherein the IMU data includes movement data from video data from motion capture data.

4. The computing device of Claim 1 , wherein the plurality of sensor data is generated by pairs of sensors monitoring the subject.

5. The computing device of Claim 4, wherein the pairs of sensors are placed with one sensor monitors a muscle on a left side and another sensor monitors a matching muscle a right side of the subject.

6. The computing device of Claim 1, wherein the one or more sensors are attached to the epidermis above a muscle to be monitored.47735-37. The computing device of Claim 1, wherein at least one processor is further programmed to execute a trained Al model to preprocess the plurality of sensor data to remove noise data from the plurality of sensor data.

8. The computing device of Claim 7, wherein the trained Al model receives the plurality of sensor data as input and output locations of artifact regions of data in the plurality of sensor data.

9. The computing device of Claim 6, wherein the at least one processor is further programmed to remove the artifact regions of data from the plurality of sensor data.

10. The computing device of Claim 1, wherein the at least one processor is further programmed to identify sets of activity by the subject from the plurality of sensor data.

11. The computing device of Claim 1 , wherein the at least one processor is further programmed to identify cycles of motion from the plurality of sensor data.

12. The computing device of Claim 1, wherein the at least one processor is further programmed to extract muscle data metrics from the plurality of sensor data.

13. The computing device of Claim 1, wherein the at least one processor is further programmed to extract IMU metrics from the plurality of sensor data.

14. The computing device of Claim 1, wherein the plurality7of sensor data is normalized with MVC (maximum voluntary contraction) data collected at the beginning of a session of subject movement.

15. The computing device of Claim 1, wherein the at least one processor is further programmed to identify a plurality of activities performed by the47735-3subject based on the plurality of sensor data, wherein the identification of the plurality of activities is further based on an identified muscle.

16. The computing device of Claim 1, wherein the feedback to the subject includes real-time intensity information as subject movement is occurring.

17. The computing device of Claim 1, wherein the feedback to the subject includes a report of subject movements that were monitored by the one or more sensors.

18. The computing device of Claim 1 , wherein the feedback to the subject includes trend analysis over a plurality of sessions of subject movement.

19. A system comprising:one or more sensors monitoring a subj ect, wherein the one or more sensors monitor and provide muscle data and inertial measurement unit (IMU) data; andone or more computing devices each comprising at least one processor in communication with at least one memory device, wherein the one or more computing devices are in communication with the one or more sensors, and where the at least one processor is programmed to:receive a plurality of sensor data including muscle data and inertial measurement unit (IMU) data from the one or more sensors;preprocess the plurality of sensor data to remove noise data from the plurality of sensor data;execute at least one artificial intelligence (Al) model trained with the muscle data and IMU data, wherein the muscle data and IMU data of the plurality of sensor data are inputs for the at least one Al model, and wherein the at least one Al model outputs a classification and an identification of a subject movement; and47735-3provide feedback to the subject based on the classification and identification of the subject movement.

20. A method for monitoring subject movement implemented by a computing device comprising at least one processor in communication with at least one memory device, wherein the method comprises:receiving a plurality of sensor data including muscle data and inertial measurement unit (IMU) data from the one or more sensors monitoring a subject;preprocessing the plurality of sensor data to remove noise data from the plurality of sensor data;executing at least one artificial intelligence (Al) model trained with the muscle data and IMU data, wherein the muscle data and IMU data of the plurality of sensor data are inputs for the at least one Al model, and wherein the at least one Al model outputs a classification and an identification of a subject movement; and providing feedback to the subject based on the classification and identification of the subj ect movement.

Images