Exercise evaluation system and method of use thereof

Abstract

The motion evaluation system includes a motion evaluation device including a plurality of sensors, a motion evaluation presentation device having a screen for displaying an image, and a processing device in communication with the motion evaluation device and the motion evaluation presentation device. The processing device receives displacement data from the motion evaluation device. In response to receiving the displacement data, the processing device identifies a plurality of features from the displacement data including at least one of a movement, an amplitude, and a speed variation of the detected motion, extracts a spectrum from the plurality of features to identify a characteristic variation over time, identifies a potential disorder from the spectrum based on a proportion of abnormal motion exceeding an identified likelihood threshold, and presents the potential disorder to a user on the motion evaluation presentation device.

Description

Technical Field

[0001] The present disclosure generally relates to an evaluation system and methods of using the same, and more particularly to an evaluation system for general motor evaluation for identifying and / or predicting neuromotor disorders.

Background Art

[0002] Early detection of neuromotor disorders in a neonatal intensive care unit (NICU) enables targeted evaluation of the infant and parental support. Precht's General Movement Assessment (GMA) allows visual recognition of multiple movement patterns, which has high specificity in predicting neuromotor disorders when cramped synchronized (CS). However, multiple challenges inherent in various medical settings and a rigorous GMA training process have hindered universal adoption for use in neonates.

[0003] Infants in the NICU environment are more likely to develop neuromotor disorders associated with other conditions such as prematurity, perinatal depression, congenital heart defects, genetic or congenital syndromes. The burden of physical disability for these children and their families is much greater than the outcome of movement. Throughout their lives, infants who develop neuromotor disorders suffer from and have preventable co-morbidities in cognitive, communication, sensory and socio-emotional domains, as well as in vision, hearing, feeding, pain, sleep and uncontrolled epilepsy.

[0004] Early and targeted observation, as well as medical, surgical, or developmental guidance, are crucial in altering their outcomes. However, early observation is highly variable due to inconsistencies in implementation, limited resources in many settings, and difficulties in accessing medical care after the child leaves the hospital. The time prior to discharge from the NICU is ideal for screening infants at high risk of neuromotor disorders such as cerebral palsy (CP).

[0005] For example, cerebral palsy (CP), the most common physical disability in the United States and globally, is undervalued. Many individuals with CP suffer from developmental disregard, or the inability of the brain to "see" the affected hand (e.g., neglect), which in turn leads to inadequate sensory and motor function in the affected hand. Similarly, neglect of this nature often affects the approximately 800,000 adults in the United States who have suffered a stroke.

[0006] Currently, there is no clinical assessment that accurately quantifies developmental neglect and / or abandonment of the limbs. Known research assessments are typically based on subjective assessments of behavior and / or are not quantitative, and therefore do not translate into clinical practice. Known research assessments often confuse cognitive abilities with visuomotor and / or tactile abilities.

[0007] Current research and evaluation of CP requires large, prohibitively expensive, and complex equipment. The lack of clinical evaluation makes it difficult to accurately measure progression in specific individuals, which in turn makes it difficult to determine whether a treatment is helping a particular individual (e.g., improving their outcomes). Furthermore, effective treatments for upper limb use in CP and / or other similar physical disorders often involve realignment of sensory / motor pathways in the affected hand, but also bilateral hand movements. To implement effective rehabilitation strategies such as bilateral intensive treatment, the condition (CP) and / or the affected hand must be identified. Currently, effective treatments are isolated from effective trials. [Overview of the project]

[0008] One aspect of this disclosure comprises a motion evaluation system including a motion evaluation device including a plurality of sensors, a motion evaluation presentation device having a screen for displaying images, and a processing device that communicates with the motion evaluation device and the motion evaluation presentation device. The processing device receives displacement data from the motion evaluation device. In response to receiving the displacement data, the processing device identifies a plurality of features from the displacement data, including at least one of the detected motion, amplitude, and velocity fluctuations, extracts a spectrum from the plurality of features to identify feature fluctuations over time, identifies potential diseases from the spectrum based on the percentage of abnormal motion exceeding an identified likelihood threshold, and presents the potential diseases to the user on the motion evaluation presentation device.

[0009] Another aspect of the present disclosure includes a non-temporary computer-readable medium storing a plurality of instructions executable by an associated processor for performing a method for implementing a motion evaluation system. The method comprises receiving motion data from a motion evaluation device having a plurality of sensors, the motion data being based on the motion of a person on the motion evaluation device as detected by the plurality of sensors; plotting the motion recorded based on the motion data acquired over a first period of time to generate displacement data; and normalizing the displacement data as first, second, third, fourth, and fifth quintets having first, second, third, fourth, and fifth regions having one or more sensors from the plurality of sensors, thereby generating normalized quintet data. The method further comprises generating two clusters per element as clustered data from the normalized quintet data; and generating extracted data. Generating extracted data includes identifying the first proportion of time when the first cluster of two clusters for each element is active relative to the second proportion of time when the second cluster of two clusters for each element is active; calculating the center of mass of each cluster for each element based on the first and second proportions; and calculating the distance between the center of mass of the core quintet and the centers of mass of the peripheral quintets. The method further comprises classifying the extracted data to identify disease probabilities.

[0010] A further aspect of the present disclosure comprises a motion evaluation system including a motion evaluation device including a plurality of pressure sensors, a motion evaluation presentation device having a screen for displaying a plurality of images, and a processing device communicating with the motion evaluation device and the motion evaluation presentation device, wherein the processing device receives displacement data from the motion evaluation device and responds to the receipt of the displacement data. The processing device identifies five elements (quintets) including one or more sensors from the plurality of sensors of the motion evaluation device, extracts a plurality of features from each of the five elements to generate a set of five features, generates a set of five spectra from the set of five features, each of the set of five spectra reflecting the displacement data detected in one of the five elements. Furthermore, the processing device identifies the proportion of normal movement and the proportion of abnormal movement from the set of five spectra, identifies potential diseases based on the proportion of abnormal movement exceeding an identified probability threshold, and presents the potential diseases to the user on the motion evaluation presentation device. [Brief explanation of the drawing]

[0011] The aforementioned and other features and advantages of this disclosure will become apparent to those skilled in the art, with reference to the accompanying drawings and considering the following description of this disclosure. In the accompanying drawings, similar reference numerals refer to the same parts throughout the drawings unless otherwise noted. [Figure 1] Figure 1 is a schematic diagram of a motion evaluation system for supporting a motion evaluation device according to a particular exemplary embodiment of the present disclosure. [Figure 2A] Figure 2A shows a top view of a motion evaluation device according to a particular exemplary embodiment of the present disclosure. [Figure 2B] Figure 2B shows a schematic diagram of multiple sensors in a motion evaluation device according to a particular exemplary embodiment of the present disclosure. [Figure 2C] Figure 2C shows a schematic diagram of five elements formed by a motion evaluation device according to a particular exemplary embodiment of the present disclosure. [Figure 2D] Figure 2D shows a schematic diagram of five elements formed by a motion evaluation device according to a particular exemplary embodiment of the present disclosure. [Figure 3]Figure 3 shows a first chart of the output of a motion evaluation device displayed on a motion evaluation display device according to a particular exemplary embodiment of the present disclosure. [Figure 4] Figure 4 shows a spectral diagram of the output of a motion evaluation device displayed on a motion evaluation display device according to a specific exemplary embodiment of the present disclosure. [Figure 5] Figure 5 shows a flowchart illustrating a method for using the exercise evaluation system according to a specific exemplary embodiment of the present disclosure. [Figure 6] Figure 6 shows a flowchart illustrating a method for acquiring displacement data and identifying potential diseases using a motion assessment system according to a specific exemplary embodiment of the present disclosure. [Figure 7] Figure 7 shows a flowchart of a method for acquiring and analyzing displacement data using a motion evaluation system according to a specific exemplary embodiment of the present disclosure. [Figure 8] Figure 8 shows a flowchart of a method for normalizing displacement data using a motion evaluation system according to a specific exemplary embodiment of the present disclosure. [Figure 9] Figure 9 shows a flowchart of a method for clustering displacement data using a motion evaluation system according to a specific exemplary embodiment of the present disclosure. [Figure 10] Figure 10 shows a flowchart of a method for extracting displacement data using a motion evaluation system according to a specific exemplary embodiment of the present disclosure. [Figure 11] Figure 11 shows a flowchart of a method for storing displacement data using a motion evaluation system according to a particular exemplary embodiment of the present disclosure. [Figure 12] Figure 12 shows a first view of the top of a motion evaluation device according to a particular exemplary embodiment of the present disclosure. [Figure 13] Figure 13 shows a second top view of a motion evaluation device according to a particular exemplary embodiment of the present disclosure. [Figure 14] Figure 14 shows a third view of the top of a motion evaluation device according to a particular exemplary embodiment of the present disclosure. [Modes for carrying out the invention]

[0012] Those skilled in the art will understand that the multiple elements in the drawings are shown for simplicity and clarity and are not necessarily drawn to scale. For example, the dimensions of some elements in the drawings may be exaggerated compared to other elements to assist in the understanding of the embodiments of the present disclosure.

[0013] In the drawings, the components of the apparatus and method are represented, where appropriate, by conventional symbols and only show those specific details relevant to the understanding of the embodiments of the present disclosure, so that the present disclosure is not obscured by details that would be readily apparent to those skilled in the art having the benefit of the description herein.

[0014] Referring now generally to the drawings, like numbered features shown in the drawings refer to like elements throughout, unless otherwise indicated. The present disclosure generally relates to an evaluation system and a method of using the same, and more specifically to an evaluation system that monitors and / or measures neural function, neural and / or muscle fatigue, and / or variations in cognitive and / or multiple motion variables.

[0015] FIG. 1 shows a schematic diagram of a motion evaluation system 100 according to one of the multiple exemplary embodiments of the present disclosure. The motion evaluation system 100 includes a processing device 112. In a particular exemplary embodiment, the processing device 112 includes a computing device 115 having computing capabilities (e.g., a database server, a file server, an application server, a computer, etc.) and / or a processor 114. The processor 114 includes a central processing unit (CPU) such as a programmable general-purpose or dedicated microprocessor, and / or other similar devices, or combinations thereof.

[0016] The processing device 112 generates an output based on inputs received from the motion evaluation device 200 and / or the motion evaluation presentation device 300, cloud storage, local input from the user, etc. Those skilled in the art will understand that in some embodiments, the processing device 112 includes various forms of data storage devices 117 such as non - transient memory, volatile memory, and non - volatile memory that store buffered data or persistent data as well as a plurality of compiled programming codes used to execute the plurality of functions of the processing device 112. In another exemplary embodiment, the data storage device 117 is external to the processing device 112 and can be accessed by the processing device 112. In yet another exemplary embodiment, the data storage device 117 includes an external hard drive, cloud storage, and / or other external recording devices 119.

[0017] In a particular exemplary embodiment, the processing device 112 includes one of a remote or local computer system 121. The computer system 121 includes desktops, laptops, tablets, handheld personal computing devices, IAN, WAN, WWW, etc. operating on any number of known operating systems and is accessible to communicate with remote data storage such as the cloud, host operating computers, etc. via the World Wide Web or the Internet.

[0018] In another exemplary embodiment, the processing unit 112 includes a processor, a microprocessor, data storage, computer system memory including random access memory ("RAM"), read-only memory ("ROM"), and / or an input / output interface. The processing unit 112 executes multiple instructions from either internal or external non-temporary computer-readable media through the processor and communicates with the processor from the motion evaluation device 200 and / or motion evaluation presentation device 300, etc., via the input interface and / or telecommunications. In yet another exemplary embodiment, the processing unit 112 communicates with networks such as the Internet, LAN, WAN, and / or the cloud, input / output devices such as flash drives, remote devices such as smartphones or tablets, and displays.

[0019] In a particular exemplary embodiment, the motion evaluation presentation device 300 includes an interactive display 304, the display which receives tactile input. In a particular exemplary embodiment, the motion evaluation presentation device 300 includes a secondary device such as a smartphone or tablet. In another exemplary embodiment, the processing unit 112, an SD card writer (e.g., for data acquisition), and the interactive display 304 (e.g., a touch-screen status-and-control LCD display) are housed in a separate module connected to the processing unit 112 and / or the motion evaluation device 200 via short-range wireless signals, Wi-Fi, and / or wired communication.

[0020] As shown in Figures 2A to 2D, the motion evaluation device 200 includes a flat surface or plane within the detection area 202 that supports multiple sensors 210. In a particular embodiment, the motion evaluation device 200 supports a grid of 32 × 32 sensors 210 (see Figure 2B). A particular example of the motion evaluation device 200 is the BodiTrak mat system from VistaMedical. The multiple sensors 210 are pressure sensors such as resistive sensors, capacitive sensors, piezoelectric sensors, and / or micro electro-mechanical system (MEMS) sensors. In a particular exemplary embodiment, the motion evaluation device 200 supports multiple sensors 210 arranged in an array where some or all of the sensors 210 are not evenly spaced apart from each other. In this exemplary embodiment, the orientation of the infant or child relative to the detection area 202 affects the quality of motion comparison between multiple infants and multiple children.

[0021] The motion evaluation device 200 is covered with a fabric such as plastic, a natural material (e.g., cotton or linen sheet), or an artificial material (e.g., polyester sheet), or other material that allows for the identification of pressure changes on multiple sensors 210. In another exemplary embodiment, the motion evaluation device 200 is covered with surface labeling such as surface labels 1200, 1300, and 1400 shown in Figures 12 to 14. Surface labeling 1200 shown in Figure 12 outlines the head, two arms, torso, and bottom half of an infant or child. Surface labeling 1300 shown in Figure 13 outlines the head, two arms, torso, and two legs of an infant or child. Surface labeling 1400 shown in Figure 14 shows text indicating where the head and feet of an infant or child should be positioned, as well as text and line figures indicating a centerline along which the infant or child should be positioned. The surface markings 1200, 1300, and 1400 maximize consistency and reproducibility of infant or child placement across multiple users, creating a more uniform orientation of the infant or child relative to the detection area 202.

[0022] Similar to the embodiment shown in Figure 1, the motion evaluation device 200 has a power supply 208. In a particular exemplary embodiment, the power supply 208 is either cordless (e.g., battery-powered) or corded. In another exemplary embodiment, the motion evaluation device 200 includes a communication device 212. In a particular exemplary embodiment, the communication device 212 communicates with the processing unit 112 via shortwave radio waves (Bluetooth®), Wi-Fi, and / or corded communication. The sensing area 202 transmits detected pressure information to the processing unit 112, which divides multiple portions of the sensing area into first, second, third, fourth, and fifth elements 202a, 202b, 202c, 202d, and 202e.

[0023] As shown in the exemplary embodiment of Figure 5, Method 500 includes several steps of using the motion evaluation device 200. In 502, the infant or child is positioned on the motion evaluation device 200 in a first orientation. In a particular exemplary embodiment, the first orientation aligns with the surface markings 1200, 1300, 1400 (see Figures 12-14), with the infant or child's head positioned on or near the first and second elements 202a, 202b, the infant's body positioned on the entire third element 202c, and the infant's lower body positioned on the entire fourth and fifth elements 202d, 202e. In a particular embodiment, the motion evaluation device 200 omits the surface markings 1200, 1300, 1400. It will be understood that other orientations relative to a first orientation, such as flips (e.g., using the opposite side of the motion assessment device 200), transfers, primary rotations (e.g., rotations that are 90-degree increments), secondary rotations (e.g., rotations that are between 0 and 90 degrees), or any combination thereof, may be contemplated. Advantageously, the surface markers 1200, 1300, 1400 provide a consistent visual cue to multiple providers. The use of surface markers 1200, 1300, 1400 on one side of the motion assessment device 200 eliminates variations in flip orientation between different infants and children. The use of surface markers 1200, 1300, 1400 eliminates variations in primary rotation. Furthermore, the use of surface markings 1200, 1300, and 1400 minimizes transfer and secondary rotation by including a centerline target or anthropomorphic graphic centered on multiple sensors 210.

[0024] In another exemplary embodiment, after the infant is positioned on the detection area 202, the processing unit 112 assigns elements 202a, 202b, 202c, 202d, and 202e based on the detected orientation of the infant, such that the infant's head is in or near the first and second elements 202a, 202b, the infant's torso is in the entirety of the third element 202c, and the infant's lower body is in the entirety of the fourth and fifth elements 202d, 202e. In this exemplary embodiment, the motion evaluation device 200 transmits sensor data collected by the multiple sensors 210 to the processing unit 112, which then divides the multiple sensors into elements 202a, 202b, 202c, 202d, and 202e, as shown in Figures 2A, 2C, and 2D.

[0025] In 504, the infant's movements are monitored and recorded by the multiple sensors 210 during a first period (e.g., 1 to approximately 3 minutes). In another exemplary embodiment, the first period is 2 minutes. In a particular exemplary embodiment, the processing unit 112 times the sensor data collected by the multiple sensors 210 in order to generate recorded movements (collected by the multiple sensors 210). In 506, the movement evaluation system 100 determines the likelihood of disease from the recorded movements (e.g., identifying normal movements versus abnormal movements). In a particular exemplary embodiment, the movement evaluation system 100 determines that the recorded movements exceed a likelihood threshold (e.g., are more likely to indicate abnormal movements than normal movements) and presents suggested further examinations / recommended evaluations for the infant to undergo to the movement evaluation presentation device 300.

[0026] As shown in the exemplary embodiment of Figure 6, Method 600 includes several steps using the motion evaluation system 100. In 602, pressure data is received from the motion evaluation device 200. In a particular exemplary embodiment, the processing unit 112 receives the pressure data from the motion evaluation device 200. In 604, the pressure data is converted into distance data by calculating the infant's barycenter 204 (see Figure 2C). In a particular exemplary embodiment as shown in Figure 2C, the barycenter 204 is calculated for each element 202a-202e, and the barycenter is the center of mass of two or more bodies orbiting each other, and is the point from which those bodies orbit. In this exemplary embodiment, the third center of gravity 204c is the center of the third element 202c (for example, the point around which multiple objects orbit), the first center of gravity 204a is calculated for the first element 202a, the second center of gravity 204b is calculated for the second element 202b, the fourth center of gravity 204d is calculated for the fourth element 202d, and / or the fifth center of gravity 204e is calculated for the fifth element 202e (see Figure 2C). Those skilled in the art will understand that the center of gravity of any element may be the point around which multiple objects orbit.

[0027] In 606, the displacement of the infant over time is determined by recording distance data over time (tracking). In a particular exemplary embodiment, as shown in Figure 2A, the displacement is the change in the center of gravity 202b–202e over time. In 608, several features are identified from the displacement data, including respiratory motion, amplitude, and / or velocity fluctuations. In a particular exemplary embodiment, respiratory motion is identified in a third element 202c. In another exemplary embodiment, amplitude and / or velocity fluctuations are identified by elements 202a–202e. Amplitude measures the change in pressure, where higher pressure from a given sensor results in a higher amplitude, and lower pressure results in a lower amplitude. In yet another exemplary embodiment, velocity fluctuations are identified by the change in the center of gravity 204a–204b, 204d–204e over time in the respective elements 202a–202b, 202d–202e. In a particular exemplary embodiment, as shown in Figure 3, the amplitude over time 308 is displayed on the motion evaluation display device 300. In this particular exemplary embodiment, the amplitude over time 308 is extracted for each element 202, where the first amplitude 302a represents the motion expressed as the amplitude over time in the first element 202a, the second amplitude 302b represents the motion expressed as the amplitude over time in the second element 202b, the third amplitude 302c represents the motion expressed as the amplitude over time in the third element 202c, the fourth amplitude 302d represents the motion expressed as the amplitude over time in the fourth element 202d, and the fifth amplitude 302e represents the motion expressed as the amplitude over time in the fifth element 202e. In this particular exemplary embodiment, the amplitude over time 308 is displayed on the motion evaluation display device 300 in real time and / or after the completion of the first period. In another exemplary embodiment, the amplitude 308 over time is stored and analyzed by the processing unit 112 and is not presented to the user.

[0028] In 610, the spectrum 310 is extracted from multiple features to identify the temporal feature variation. In a particular exemplary embodiment, as shown in Figure 4, the spectrum 310 is displayed on the motion evaluation display device 300. In a particular exemplary embodiment, the spectrum 310 is extracted for each temporal amplitude 302a to 302e, with the first spectrum 312a corresponding to the first temporal amplitude 302a in the first element 202a, the second spectrum 312b corresponding to the second temporal amplitude 302b in the second element 202b, the third spectrum 312c corresponding to the third temporal amplitude 302c in the third element 202c, the fourth spectrum 312d corresponding to the fourth temporal amplitude 302d in the fourth element 202d, and the fifth spectrum 312e corresponding to the fifth temporal amplitude 302e in the fifth element 202e. In 612, potential diseases and / or probabilities of potential diseases are identified from the spectrum 310 based on the normal versus abnormal movements of the identified infant. In 614, the potential diseases and / or probabilities of potential diseases based on the detection of the infant's normal versus abnormal movements are presented to the user on the movement assessment display device 300. In certain exemplary embodiments, in response to the identification of potential diseases and / or probabilities of potential diseases exceeding a probability threshold (e.g., from the identification of abnormal movements), additional tests necessary to confirm the disease are presented to the user on the movement assessment display device 300. In another exemplary embodiment, in response to the identification of potential diseases and / or probabilities of potential diseases exceeding a probability threshold, recommended assessments consistent with published guidelines are presented to the user on the movement assessment display device 300. In another exemplary embodiment, the spectrum 310 is presented to the user on the movement assessment display device 300 in real time or presented upon completion.

[0029] As shown in the exemplary embodiment of Figure 7, Method 700 includes several steps performed by the motion evaluation system 100. In 702, the infant's movements are monitored and recorded over a first period using the motion evaluation device 200 to generate recorded infant data. In this exemplary embodiment, the recorded infant data includes movements recorded from the motion evaluation device 200 over the first period. In another exemplary embodiment, in response to the infant being placed supine in a first orientation on the motion evaluation device 200, pressure data is sampled from the motion evaluation device at a first rate (e.g., 32 Hz or 32 frames per second) to generate comparable motion data. In another particular exemplary embodiment, the processing unit 112 stores the recording of pressure data for the first orientation acquired over the first period.

[0030] In a particular exemplary embodiment, pressure data from the motion evaluation device 200 is output to a CSV file in the form of multiple columns representing periods or time points and multiple rows containing pressure data for each sensor of multiple sensors 210 to generate amplitudes 208 and / or spectra 310 over time. The pressure data from each period or time point is reconstructed to obtain a sensor matrix that reflects the number of sensors present within the multiple sensors 210. The pressure data is plotted to verify that the infant's head is positioned in a first orientation (e.g., corresponding to surface markers 1200, 1300, 1400), in which the head is between the first element 202a and the second element 202b and above the third element 202c. The five elements 202 are transposed or rotated as necessary until the infant is in the first orientation. The midpoint of the pressure data collected over time is determined to select the central 3000 timepoint matrix. In a particular exemplary embodiment, the midpoint of the matrix at time 3000 is 1500. It is conceivable that additional time matrices, both greater than and less than 3000, could be considered.

[0031] In method 704, recorded infant data are plotted and adjusted based on the detected infant movement over a first period to generate displacement data. In method 800, shown in Figure 8 and described in detail below, the displacement data is normalized to generate normalized displacement data. Normalization addresses any differences such as sensor sensitivity and the thickness of clothing worn by the infant. In method 900, shown in Figure 9 and described in detail below, the normalized displacement data is data clustered to generate clustered normalized displacement data. In method 1000, shown in Figure 10 and described in detail below, the clustered normalized displacement data is feature extracted to generate a set of extracted features. In method 1100, shown in Figure 11 and described in detail below, the extracted features are categorized to identify the infant's normal vs. abnormal movement, disease type, and / or disease probability.

[0032] As shown in Figure 8, a method 800 for normalizing displacement data is illustrated. In 802, displacement data at each period or point in time received from the motion evaluation device 200 is referenced using steps 804, 806, and 808 in parallel. In a particular exemplary embodiment, pressure data is normalized to minimize differences due to the recording process and conditions (e.g., thickness or softness of the crib or mattress, infant placement, infant size, and / or weight). In 804, binary is assigned to the pressure detected from the motion evaluation device 200, where pressure = 1 and no pressure = 0 (e.g., binary normalization). Binary normalization yields binary normalized displacement data.

[0033] In step 806, the sensor pressure detected at each of the multiple sensors 210 is divided by the standard deviation of the detected sensor pressures across all the multiple sensors (e.g., standard deviation (STD) normalization). STD normalization yields STD normalized displacement data. In step 808, a histogram compensation program is used to assign the calculated values ​​to a range from "0" to the final value (e.g., histogram normalization). In a particular exemplary embodiment, the range is 0 to 255. A particular exemplary histogram compensation program is OpenCV, which is commonly used to improve grayscale images, resulting in pressure data where the calculated values ​​are assigned to a range from black to white, such as 0, 255. Histogram normalization yields histogram normalized displacement data. In step 810, the binary, STD, and histogram normalized displacement data generated in method steps 804, 806, and 808 are independently clustered in method 900, as shown in Figure 9. In a particular exemplary embodiment, method steps 804, 806, and 808 are performed simultaneously. In another exemplary embodiment, method steps 804, 806, and 808 are performed sequentially.

[0034] Figure 9 shows a method 900 for clustering normalized displacement data (e.g., binary, STD, and histogram-normalized displacement data). In a particular exemplary embodiment, the clustering of method 900 is performed independently for each of the binary, STD, and histogram-normalized displacement data. In another exemplary embodiment, the clustering of method 900 is performed independently for at least one of the binary, STD, and histogram-normalized displacement data.

[0035] In 902, K-means clustering is performed on the normalized displacement data. An example of K-means clustering is performed using the scikits learn module. In 904, the first cluster is assigned as motionless or less active, and the second cluster is assigned as motion state. In 906, two clusters per 202 elements are generated as clustered data. In this exemplary embodiment, K-means clustering reduces aggregate durations or time points with similar variances within the normalized displacement data. The clustered data includes independently generated binary clustered displacement data, which is clustered data based on binary normalized displacement data; STD clustered displacement data, which is clustered data based on STD normalized displacement data; and histogram clustered displacement data, which is clustered data based on histogram normalized displacement data, and is stored in the processing unit 112.

[0036] Figure 10 shows a method 1000 for extracting multiple features from clustered normalized displacement data. In a particular exemplary embodiment, method steps 1002–1012 are performed on binary clustered displacement data, STD clustered displacement data, and / or histogram clustered displacement data, collectively referred to as clustered data, to generate three independent extracted datasets. In 1002, multiple features are extracted from the clustered data. In 1004, the ratio of the time a first cluster is active to the time a second cluster is active is identified. In this embodiment, the first feature extracted is the ratio of time each cluster appeared during the first period. In other words, the fitting step is performed to generate a comparable signal processing analysis.

[0037] In 1006, in the first and second clusters, at least one of the following is calculated: total area of ​​activation (corresponding to the number of sensors among the multiple sensors 210 that record pressure), total mean pressure (e.g., the total pressure across the multiple sensors 210), and total standard deviation of pressure (STD). In this exemplary embodiment, the second feature extracted is at least one of the total area of ​​activation, total mean pressure, and total STD of pressure in the first and second clusters.

[0038] In 1008, the center of mass of each cluster of each element 202a-202e is calculated using the following Equation 1. In this exemplary embodiment, the center of mass represents the difference in the infant's weight distribution in the motion evaluation device 200. In a particular exemplary embodiment, the center of mass includes x and y coordinates (e.g., the two centers of mass of each element 202a-202e). The two centers of mass are combined into x and y coordinates to constitute the center of mass. Furthermore, each sensor of the multiple sensors 210 includes a center of mass coordinate, and the combination of the two center of mass coordinates (x,y) representing the center of mass generates a representation of the center of mass of the infant on the motion evaluation device 200, or the center of mass of the infant within a particular element 202a-202e, representing the heterogeneous distribution of the infant in the motion evaluation device or in the various five elements. In a particular exemplary embodiment, a third feature extracted is the center of mass of each cluster of each element 202a-202e. In a particular exemplary embodiment, the center of mass of each cluster is represented as a tuple (e.g., xb, yb) with coordinates (0,0). In this example, the center of mass is obtained by combining the centroids 204a to 204e of each element 202a to 202e along the x and y axes, as shown in Equation 1 below.

[0039]

number

[0040] In Equation 1, x or y is the position of an index, and v is the pressure value from the corresponding index. In certain exemplary embodiments, the fluidity, velocity, and engagement of multiple limbs of an infant during general movement are considered by indexing multiple tuples.

[0041] In 1010, the distance between the center of mass of the core element (the third element 202c) and the multiple peripheral elements (the first, second, fourth, and fifth elements 202a, 202b, 202d, and 202e) (see Figures 2A and 2C) is calculated. In a particular exemplary embodiment, the fourth feature extracted is the distance between the center of mass of the core element 202c and the multiple peripheral elements 202a, 202b, 202d, and 202e. In another exemplary embodiment, the five elements 202 are defined as a 3x3 sensor quintet with respect to the center of mass of the core element 202c centered at (0,0) to represent the trunk movement of the infant, and the multiple peripheral elements 202a, 202b, 202d, and 202e are skewed outside and above / below the multiple sensors with multiple peripheral angles of the core element. In this exemplary embodiment, the pressures and identified movements in the peripheral elements 202a, 202b, 202d, and 202e represent the fulcrums of the infant's proximal joint movements. In a particular exemplary embodiment, each of the five elements 202a, 202b, 202c, 202d, and 202e is assigned its own calculated center of mass having coordinates (x1, y1)...(x5, y5) as defined using Equation 1 above.

[0042] In 1012, the distances 206A, 206B, 206D, 206E (see Figure 2C) between the mass center 204c of the core element 202c and the mass centers 204a, 204b, 202d, 202e of the surrounding elements 202a, 202b, 202d, 202e are calculated to generate the extracted data. In a particular exemplary embodiment, the distances 206a, 206b, 206d, 206e between the mass centers 204a, 204b, 204d, 204e of the surrounding elements 202a, 202b, 202d, 202e and the mass center 204c of the core element 202c are calculated using vector products. These distances represent the movement of the pivot point relative to the infant's trunk. The distance 206c from the center of mass 204c, assigned as (0,0) of core element 202c, to the calculated center of mass of core element 202f is calculated to represent the respiratory movements and / or large positional changes of the infant on the motion evaluation device 200. In this exemplary embodiment, the fifth feature extracted includes the calculated distances 206a, 206b, 206d, and 206e.

[0043] Figure 11 shows a method 1100 that aggregates multiple extracted features using a decision tree. In a particular exemplary embodiment, method steps 1102-1106 are used to aggregate the first, second, third, fourth, and multiple fifth features extracted from the binary clustered displacement data, the standard clustered displacement data, and / or the histogram clustered displacement data, respectively, into extracted data. In 1102, a decision tree model is used to aggregate the extracted data generated in methods 900 and 1000 from the independent normalized data generated in method 800. In a particular exemplary embodiment, the decision tree model is a supervised decision tree classifier from the scikits-learn module, which predicts the results based on two classifications. In 1104, the decision tree is used to identify normal and abnormal motion. In this exemplary embodiment, detection of normal motion is achieved when abnormal motion is distinguishable from normal motion. For example, a movement that is simply an insufficient repertoire is distinguished from an abnormal movement that carries a high risk of motor impairment. In 1106, five-fold cross-validation was used to distinguish between normal and abnormal movements. In this exemplary embodiment, the decision tree was trained using a decision tree classifier in a five-fold cross-validation procedure. The decision tree output is a confusion matrix (true positive, true negative, false positive, and false negative). Sensitivity and specificity were then calculated.

[0044] Advantageously, the motor assessment system 100 allows for the earlier identification of abnormal movements in infants than current methods, enabling more time for treatment. Furthermore, the motor assessment system 100 enables more accurate identification and referral in hospitals, while also facilitating access to specialized medical care through early identification of abnormal movements / diseases. The use of the motor assessment system 100 improves the provision of targeted effective early interventions after infants with potential disorders / abnormal movements are discharged from the hospital. Targeted early interventions have known positive downstream impacts on the neurodevelopmental outcomes of such infants. The motor assessment system 100 does not require extensive training for users and allows users to identify or diagnose potential disorders / abnormal movements without frequently retaking expensive and infrequent courses. In addition, surface labeling 1200, 1300, and 1400 provides consistency and repeatability across multiple users, maximizing the effectiveness of the motor assessment system 100. Finally, the motor assessment system 100 helps to increase awareness of the importance of early detection of CP and other disorders to reduce preventable disorders, as well as to increase the ability of research institutions to develop new interventions to alter infant outcomes earlier than previously possible for systematically identified populations.

[0045] Specific embodiments are described in the foregoing specification. However, those skilled in the art will understand that various modifications and alterations can be made without departing from the scope of this disclosure as defined in the following claims. Accordingly, this specification and the figures should be interpreted as illustrative rather than restrictive, and all such modifications are intended to be within the scope of this teaching.

[0046] Any benefit, advantage, solution to a problem, or any one or more elements that could evoke or further highlight any benefit, advantage, or solution shall not be construed as a defining, required, or essential feature or element of any claim. This disclosure is defined solely by the appended claims, including any amendments made during the pendency of this application and all equivalents of those claims as granted.

[0047] Furthermore, in this specification, terms indicating relationships, such as first and second, above and below, may be used solely to distinguish one entity or action from another, and do not necessarily require or imply any actual relationship or order between such entities or actions. The terms “include,” “contains,” “have,” “possess,” “include,” “contain,” “include,” or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article or apparatus that includes, has, contains, or includes a list of elements may not only include those elements but may also include other elements that are not expressly listed or that are inherent to such process, method, article or apparatus. Elements preceded by “include,” “have,” “contain,” or “contain” do not, without further restriction, exclude the existence of additional identical elements in a process, method, article or apparatus that includes, has, contains, or includes that element. The terms “a” and “an” are defined as one or more unless otherwise specified herein. The terms “substantially,” “essentially,” “approximately,” “about,” or any other variation thereof are defined as close, as understood by those skilled in the art. In one non-limiting embodiment, these terms are defined as being, for example, within 10%, within 5% in another possible embodiment, within 1% in another possible embodiment, and within 0.5% in another possible embodiment. The term “coupled,” as used herein, is defined as being connected or in contact, whether temporary or permanent, but not essential, but not directly and not necessarily mechanically. A device or structure “configured” in a certain way is configured at least in that way, but may also be configured in other ways not listed.

[0048] Unless documentation for any of the embodiments described above or its components is specified, it should be understood that a person skilled in the art would be able to identify documentation suitable for the intended purpose.

[0049] A summary of this disclosure is provided so that readers can quickly ascertain the nature of this technical disclosure. This is submitted with the understanding that it will not be used to interpret or limit the claims or their meaning. In addition, in the detailed description above, various features are grouped into one in various embodiments for the purpose of simplifying the disclosure. This method of disclosure should not be interpreted as reflecting an intention that the claimed embodiments require more features than are explicitly described in each claim. Rather, as reflected in the following claims, the subject matter of the invention lies in features that are not all present in a single disclosed embodiment. Accordingly, the following claims are incorporated herein by reference into the detailed description, and each claim stands alone as individually claimed subject matter.

Claims

1. It is an exercise evaluation system, A motion evaluation device including a grid of multiple sensors, wherein the motion evaluation device includes a flat surface or plane, A motion evaluation presentation device having a screen that displays multiple images, The system comprises a motion evaluation device and a processing device that communicates with the motion evaluation presentation device, The processing device receives displacement data from the motion evaluation device, In response to receiving the displacement data, the processing device: Identifying multiple features from displacement data, wherein each of the multiple features includes at least one of motion, amplitude, and velocity fluctuations detected from the multiple sensors. Extracting spectra from the aforementioned multiple features to identify changes in features over time, From the aforementioned spectrum, the ratio of normal movement to abnormal movement is identified, Identifying potential diseases based on the percentage of abnormal movements that exceed a specified probability threshold, An exercise evaluation system that performs the following: presenting the potential disease to the user on the exercise evaluation presentation device.

2. The motion evaluation system according to claim 1, wherein the processing device identifies five elements, including one or more sensors from the plurality of sensors of the motion evaluation device.

3. The motion evaluation system according to claim 2, wherein the plurality of features are extracted from each of the five elements to generate a set of five features, the spectrum is generated from the set of five features to generate a set of five spectra, and each of the set of five features reflects displacement data detected in one of the five elements.

4. The motion evaluation system according to claim 1, wherein the plurality of sensors are a plurality of pressure sensors.

5. The motion evaluation system according to claim 1, wherein text or graphic representations of the orientation of a person using the motion evaluation system are superimposed on the plurality of sensors.

6. The motion evaluation system according to claim 1, wherein the processing device normalizes the displacement data in parallel using first, second, and third normalization structures to generate normalized displacement data, and the first, second, and third normalization structures have different structures.

7. The motion evaluation system according to claim 6, wherein the first normalization structure is a binary normalization structure, the second normalization structure is a standard deviation normalization structure, and the third normalization structure is a histogram compensation normalization structure.

8. The motion evaluation system according to claim 7, wherein the displacement data is received and normalized as first, second, third, fourth, and fifth elements including one or more sensors from the plurality of sensors, and normalized element data is generated by normalizing the displacement data.

9. The motion evaluation system according to claim 8, wherein the processing device clusters the normalized element data, and the processing device generates two clusters for each element as clustered data.

10. The motion evaluation system according to claim 6, wherein the processing device clusters the normalized displacement data using K-means clustering to generate clustered displacement data.

11. The aforementioned processing apparatus is Identify the ratio of the time when the first cluster is active for each element to the ratio of the time when the second cluster is active for each element, Calculate the center of mass of each cluster of each element. The motion evaluation system according to claim 9, which calculates the distance between the center of mass of a core element and the centers of mass of a plurality of peripheral elements and generates extracted data.

12. The aforementioned processing apparatus is A first feature is extracted that includes the ratio of the time the first cluster of the two clusters is active for each element, relative to the ratio of the time the second cluster of the two clusters is active for each element. A second feature is extracted, including the total active area, total mean pressure, and total standard deviation of pressure in the first and second clusters. A third feature is extracted, including the center of mass of each cluster of each element. A fourth feature is extracted, which includes the distance between the center of mass of the core element and multiple peripheral elements. The motion evaluation system according to claim 9, wherein a fifth feature is extracted, which includes the distance between the center of mass of a core element and the centers of mass of a plurality of peripheral elements, and the plurality of features include extracted data.

13. The motion evaluation system according to claim 12, wherein the processing device aggregates the extracted data from the normalized displacement data using a decision tree.

14. The motion evaluation system according to claim 11, wherein the processing device uses a decision tree to identify normal motion and abnormal motion from the extracted data and the normalized displacement data.

15. The motion evaluation system according to claim 14, wherein the processing device uses five-segment cross-validation to confirm the identified normal motion and abnormal motion.

16. The motion evaluation system according to claim 10, wherein the processing device aggregates the extracted data from the normalized displacement data using a decision tree.

17. An exercise evaluation system, A motion evaluation device including multiple sensors, A motion evaluation presentation device having a screen that displays multiple images, The system comprises a motion evaluation device and a processing device that communicates with the motion evaluation presentation device, The processing device receives displacement data from the motion evaluation device, In response to receiving the displacement data, the processing device: Identifying multiple features from displacement data, wherein each of the multiple features includes at least one of motion, amplitude, and velocity fluctuations detected from the multiple sensors. Extracting spectra from the aforementioned multiple features to identify changes in features over time, From the aforementioned spectrum, the ratio of normal movement to abnormal movement is identified, Identifying potential diseases based on the percentage of abnormal movements that exceed a specified probability threshold, The system presents the potential disease to the user on the exercise evaluation display device, and performs the following actions: A motion evaluation system in which text or graphic representations of the orientation of a person using the motion evaluation system are superimposed on the multiple sensors.

18. A movement evaluation system, A motion evaluation device including multiple sensors, A motion evaluation presentation device having a screen that displays multiple images, The system comprises a motion evaluation device and a processing device that communicates with the motion evaluation presentation device, The processing device receives displacement data from the motion evaluation device, In response to receiving the displacement data, the processing device: Identifying multiple features from displacement data, wherein each of the multiple features includes at least one of motion, amplitude, and velocity fluctuations detected from the multiple sensors. Extracting spectra from the aforementioned multiple features to identify changes in features over time, From the aforementioned spectrum, the ratio of normal movement to abnormal movement is identified, Identifying potential diseases based on the percentage of abnormal movements that exceed a specified probability threshold, The system presents the potential disease to the user on the exercise evaluation display device, and performs the following actions: The processing device normalizes the displacement data in parallel using first, second, and third normalization structures to generate normalized displacement data, and the first, second, and third normalization structures have different structures, thus forming a motion evaluation system.

19. The motion evaluation system according to claim 18, wherein the first normalization structure is a binary normalization structure, the second normalization structure is a standard deviation normalization structure, and the third normalization structure is a histogram compensation normalization structure.

Images