Measurement method and device

Abstract

Disclosed are an apparatus and method for determining motion information of a user carrying an accelerometer while moving. The apparatus receives acceleration data from the accelerometer, which is defined relative to the accelerometer's frame of reference. A transformation is determined and applied to the acceleration data, or data derived from the acceleration data, to determine acceleration data that is a user's frame of reference that includes the user's forward direction and a lateral direction transverse to the user's forward direction. The acceleration data, or data derived from the acceleration data, is analyzed to determine a period corresponding to either the user's stride period or step period when the user is walking or running. Information about lateral acceleration is used to clarify whether the determined period corresponds to the user's stride period or the user's step period.

Description

[Technical Field]

[0001] The present invention relates to the field of measuring user movements, and in particular, but not exclusively, to methods and devices for measuring and / or analyzing user movements to determine the user's step period and stride period. [Background technology]

[0002] Devices such as activity trackers or pedometers are used to measure a user's movement and can be used to determine when a user is walking, thereby determining the number of steps the user takes, i.e., step count.

[0003] Current devices are generally aimed at the leisure market, where repeatability is emphasized over accuracy. These devices may be dedicated devices designed to monitor a user's steps, or they may take the form of a software application that runs on a user device such as a mobile phone or smartwatch. As anyone who has used these devices knows, different devices often produce significantly different step counts for the same distance walked.

[0004] Existing methods and devices analyze the magnitude of data generated by an accelerometer mounted on a user device when a user walks. Specifically, existing methods and devices typically calculate the autocorrelation of this magnitude data over a period of time or roughly detect spike waveforms in the magnitude data corresponding to heel strikes to determine periodic motion corresponding to the user's steps, which are then counted. However, these analyses also capture other periodic motions, such as the user's stride period (the time interval between the first heel strike and the second heel strike), which should be approximately twice the step period (the time interval between the first heel strike and the second heel strike). Typically, when a user's step period during walking is less than approximately 0.8 seconds, existing technologies typically compare the measured period to this threshold to attempt to distinguish between step and stride periods. However, the inventors have recognized that existing methods that include this thresholding process introduce errors into the calculations they perform.

[0005] There is a need for devices and methods that can more accurately determine a user's movement. Such devices and methods would, of course, be used in the leisure market, where users would appreciate more accurate information, but they could also help open up new markets for this type of analysis. For example, athletes are constantly seeking devices and methods that can accurately analyze their movement to enable them to improve their skills and gain an advantage over their competitors. Devices that can track and accurately monitor a user's movement could also be used in the medical field, either for remote diagnostic purposes or to collect data that may be relevant to clinical research. For example, movement information may be needed for correlation with other sensors or time-specific measurements, and the absence of this data could adversely affect treatment efficacy determinations. In medical applications, the requirement for accuracy is particularly important, as it could affect treatment decisions and / or drug trial results, thereby significantly impacting health.

[0006] Some conditions, such as central nervous system disorders, may also be used in subjects with atypical walking styles, making devices and algorithms optimized for the general (i.e., healthy) population inappropriate.

[0007] Because some subjects may be unable or unwilling to wear a device in a particular location (e.g., ankle) or a particular device (e.g., a watch), the device and algorithm should ideally be location-independent and valid for a variety of hardware devices to address these issues. Summary of the Invention [Problem to be solved by the invention]

[0008] Aspects of the invention are set out in the independent claims, with preferred features set out in the dependent claims. [Means for solving the problem]

[0009] According to a first aspect, an apparatus and method for determining motion information of a user carrying an accelerometer while moving are provided. The apparatus receives acceleration data from the accelerometer defined relative to a frame of reference of the accelerometer. A transformation is determined and applied to the acceleration data, or data derived from the acceleration data, to determine acceleration data in the user's frame of reference, including a direction of the user's forward movement and a lateral direction transverse to the user's direction of movement. The acceleration data, or data derived from the acceleration data, is analyzed to determine a period corresponding to either a user's stride period or a user's step period when the user is walking or running. Information about lateral acceleration is used to clarify whether the determined period corresponds to a user's stride period or a user's step period.

[0010] In some embodiments, the processor and memory are configured to use information about the lateral and forward accelerations to clarify whether the determined period corresponds to a user's stride period or a user's step period.

[0011] The processor and memory may determine a first autocorrelation function to determine the period corresponding to either the user's stride period or the user's step period, and may process the first autocorrelation function to identify a peak in the first autocorrelation function at an autocorrelation lag corresponding to the user's stride period or the user's step period. In some embodiments, the processor and memory process the first autocorrelation function to identify a highest peak in the first autocorrelation function after the zero-lag peak, and determine the period corresponding to the user's stride period or the user's step period as the autocorrelation lag associated with the identified highest peak.

[0012] Typically, the processor and memory determine a second autocorrelation function of the lateral acceleration and determine whether the period corresponds to a stride period or a step period based on whether the second autocorrelation function includes a peak near an autocorrelation lag corresponding to the step period or stride period.

[0013] A second autocorrelation function of the lateral acceleration and a third autocorrelation function of the heading acceleration may be determined and used to clarify whether a period corresponds to a stride period or a step period based on whether the second autocorrelation function contains a peak near an autocorrelation lag corresponding to a step period or a stride period. The first, second, and third autocorrelation functions may also be used to determine whether the user is walking or not.

[0014] A first autocorrelation function is calculated for the acceleration data or for transformed acceleration data that defines acceleration in the user's frame of reference.

[0015] In one embodiment, the processor and memory are configured to determine and apply a first transformation that aligns a first axis of the acceleration data or data derived from the acceleration data with a vertical axis, a second transformation that aligns a second axis of the acceleration data or data derived from the acceleration data with the direction of travel, and a third axis of the acceleration data or data derived from the acceleration data with the lateral direction. These transformations typically include a rotation.

[0016] In one embodiment, the processor and memory are configured to determine that the determined period corresponds to the user's stride period if the information about the lateral acceleration matches the information about the forward acceleration, and to determine that the determined period corresponds to the user's step period if the information about the lateral acceleration does not match the information about the forward acceleration.

[0017] The user's frame of reference typically includes a vertical direction transverse to both the forward and lateral directions.

[0018] In one embodiment, the processor and memory are configured to process the acceleration data to identify walking periods within the acceleration data, and to use the acceleration data from within the identified walking periods to determine the periods corresponding to either a stride period or a step period of the user.

[0019] The forward and lateral directions may be determined as directions in the horizontal plane that have the greatest and least variability in the received acceleration data. Alternatively, a compass or global positioning system (e.g., GPS) onboard the user device may provide the direction of movement information.

[0020] The processor and memory may be configured to use the identified step period or stride period to determine the user's step count for exercise corresponding to walking or running. This step count information may be stored and / or output to the user (e.g., on a display of the user device). The step count information may also be transmitted to a remote computer.

[0021] The present invention also provides an apparatus for determining motion information of a user carrying an accelerometer while moving, the apparatus comprising one or more processors and a memory configured to: receive acceleration data from the accelerometer defining accelerations experienced by the accelerometer due to the user's motion, the accelerations being defined relative to a frame of reference associated with the accelerometer; apply a transform to the acceleration data or data derived from the acceleration data to convert the frame of reference to a user's frame of reference including a forward direction of the user and a lateral direction transverse to the user's forward direction; determine a first autocorrelation function of the acceleration data or data derived from the acceleration data, determine a second autocorrelation function of acceleration in the forward direction, and determine a third autocorrelation function of acceleration in the lateral direction; and determine whether the user is walking or not using the first, second, and third autocorrelation functions.

[0022] The present invention also provides an apparatus for determining motion information of a user carrying an accelerometer while moving, the apparatus comprising one or more processors and a memory configured to: receive acceleration data from the accelerometer, for each of a plurality of time points, the acceleration data including acceleration values ​​with respect to a plurality of first orthogonal directions defined by an orientation of the accelerometer, the acceleration values ​​each indicating acceleration of the accelerometer in one of the first orthogonal directions at a given time point; convert the acceleration data into transformed acceleration data including acceleration values ​​with respect to a plurality of second orthogonal directions defined by an orientation of the user, the acceleration values ​​each indicating acceleration movement of the accelerometer in one of the second orthogonal directions including a direction of travel of the user and a lateral direction transverse to the direction of travel of the user; analyze at least a portion of the acceleration data or the transformed acceleration data to determine a period corresponding to either a user stride period or a step period; and use the transformed acceleration data related to the user's motion in at least the lateral direction to determine whether the determined period corresponds to either a user stride period or a user step period.

[0023] The apparatus as outlined above may form part of a user device (such as a mobile phone, smartwatch, etc.) carried by a user, and the accelerometer may form part of the user device or may be in a separate device that communicates with the user device. The apparatus as outlined above may also form part of a central server that receives acceleration data from the user device and processes the received acceleration data to determine movement information.

[0024] The present invention also provides a method for determining motion information of a user carrying an accelerometer while moving, the method comprising receiving acceleration data from the accelerometer defining acceleration experienced by the accelerometer due to the user's movement, the acceleration being defined relative to a frame of reference associated with the accelerometer, applying a transformation to the acceleration data or data derived from the acceleration data that converts the frame of reference to a user's frame of reference that includes a direction of travel of the user and a lateral direction transverse to the direction of travel of the user, analyzing the acceleration data or data derived from the acceleration data to determine a period of time that corresponds to either the user's stride period or step period when the user is walking or running, and using information about the lateral acceleration to determine whether the determined period of time corresponds to either the user's stride period or the user's step period.

[0025] The present invention also provides a computer program product (which may be a tangible computer readable medium or a carrier wave signal) comprising computer implementable instructions that cause a programmable computing device to be configured as an apparatus as outlined above.

[0026] The present invention also provides a clinical trial system and method including a central computer in communication with a plurality of user devices, each configured to collect acceleration data relating to the movements of a user associated with the user device, wherein the central computer or at least one of the user devices comprises an apparatus as outlined above. [Brief explanation of the drawings]

[0027] Exemplary embodiments of the present invention will now be described with reference to the accompanying drawings. [Figure 1A] 1 illustrates a schematic diagram of a clinical trial in which the physical activity of users participating in the clinical trial is determined by user devices worn or carried by the users and reported to a central server for collection and analysis. [Figure 1B] FIG. 1B is a block diagram showing the major electronic components of the system shown in FIG. 1A. [Figure 2]FIG. 1C is a block diagram illustrating the main components of the user device shown in FIG. 1B. [Figure 3] FIG. 1 is a flow diagram illustrating a conventional technique for determining a user's step or stride period. [Figure 4] 1 is a plot showing an autocorrelation function calculated from acceleration data acquired while a user is walking. [Figure 5A] 1 is a flow chart illustrating a preferred technique for determining and characterizing a user's step or stride period while walking. [Figure 5B] 1 is a flow chart illustrating a preferred technique for determining and characterizing a user's step or stride period while walking. [Figure 6] 1 is a plot showing autocorrelation calculated from acceleration data acquired while a user is walking and used to characterize step and stride periods. [Figure 7] 1 shows a flow chart illustrating a preferred method for determining whether a period of exercise corresponds to a period of walking or a period of non-walking.

[0028] In the drawings, like reference numbers are used to indicate like elements. DETAILED DESCRIPTION OF THE INVENTION

[0029] overview As outlined above, the present invention provides an alternative method for analyzing a user's movement. The methods and devices provided by the present invention may be used in a variety of applications, such as fitness trackers. However, the present invention may also be used in a medical setting, as described below.

[0030] 1A and 1B show how the present invention may be used in a clinical trial system 10 in which several subjects (hereinafter also referred to as users) 30a-30e use respective user devices 100a-100e to monitor the corresponding subjects' ambulatory movements. Information gathered by the user devices 100 is transmitted via a communications network 120 (represented in FIG. 1A by dashed lines 40a-40e) to a central server 140, with results that may be displayed within the clinic 20.

[0031] Clinic 20 may be a medical center, such as a hospital or a physician's operating room. Clinic 20 may consist of a single center or may consist of several centers located in several different geographic locations. Subjects 30a-30e are patients at clinic 20 and are participating in clinical trials organized by clinic 20. Patients in the trial are cohorted into groups with the same medical condition.

[0032] Each subject 30a-30e is provided with a user device 100 that is dedicated to the clinic and can be returned to the clinic after the trial is completed. Alternatively, the clinic may provide the subject with a software application that can run on the subject's own user device, such as a mobile phone or smartwatch. In either case, each subject is asked to wear or carry their user device so that an accelerometer associated with the user device can capture the user's movements during the clinical trial. As shown in FIG. 1B , some user devices 100 have a built-in accelerometer 102, while some user devices 100 (in this example, user device 100a) do not. If the user device 100 does not have an accelerometer, a separate actigraphy-measuring device 101 is provided with an accelerometer 102-a that captures the subject's movements. The actigraphy measurement device 101 may be worn or carried by the subject, for example, around the subject's wrist, around the ankle, in a pocket, on a belt, held in the hand, in a bag worn by the subject, or worn as a pendant, for example, around the subject's neck.

[0033] The accelerometer typically provides acceleration information in three orthogonal directions depending on the orientation of the accelerometer. By analyzing the acceleration data, the user device 100 can determine movement information about the subject that is transmitted (wirelessly or via a wired connection) as subject data to a central server 140 for further analysis as part of a clinical trial.

[0034] In one embodiment, the subject data provided to the central server 140 includes gait data and identification data identifying the subject associated with the gait data. The gait data may include one or more of step counts, walking or activity periods, and walking distances over a period specified by the clinical trial, such as a day, a week, a month, or a year. The subject data may be acquired from the user device 100 when the subject visits the clinic, or the subject data may be transmitted to the clinic via a mobile phone, landline phone, or computer network (wireless or wired). The subject data collected at the clinic may be supplemented by physical observations or tests that are only performed at the clinic 20 and cannot be monitored remotely. The accuracy of the data provided to the clinic 20 regarding the subject's activities outside the clinic 20 or at home is important to ensure that the clinical trial captures a true picture of the subject's activities during the monitoring period. This may help determine the effectiveness of the clinical trial's treatments.

[0035] In another embodiment, the subject data provided to the central server 140 includes the subject's identification data along with the acceleration data, so that the central server 140 processes the acceleration data for each subject, from which the central server 140 calculates the gait data itself for each subject. Although not shown in Figure 1B, in this case the central server 140 further comprises a user interface with a user input device such as a keyboard, and / or software for processing data collected from the user devices of the system.

[0036] Subject data indicating a subject's activity, such as gait data, can be a good indicator of a subject's health and fitness level. For example, because step count is an indicator of overall health, such subject data can be used as an indicator of recovery. An increase in step count can indicate increased mobility and indicate patient improvement, while a decrease or plateau in step count can indicate that the patient is not responding to treatment, is not improving, or is becoming more ill. An increase in step count during periods of maximal treatment benefit compared to periods of diminishing benefit can provide an indication of treatment effectiveness. In some cases, step counts can indicate the need for a patient to be referred to a clinic or that the patient needs to be hospitalized for a short period of time. In some instances, collected patient data can be used by clinics to help schedule patient appointments with doctors or clinicians as needed.

[0037] The gait data provided by the user device may also be used to provide one or more of the subjects 30a-30e with a personalized exercise plan tailored to the individual needs and / or abilities indicated by the data. A prompt to get active may be sent to a subject if their step count is too low.

[0038] Gait data is particularly useful in studying patients with one or more medical conditions known to affect walking ability. Temporary gait or balance complications may be caused by injury, trauma, inflammation, or pain. Certain conditions may also cause gait, balance, and coordination problems. Conditions for which measuring gait activity may be particularly important include, but are not limited to, arthritis, multiple sclerosis (MS), Meniere's disease, brain injury (e.g., from hemorrhage or tumor), Parkinson's disease, orthopedic surgery of the lower back or lower body, cancer and related treatments, cerebral palsy, obesity, gout, muscular dystrophy, stroke, spinal cord injury, and deformities.

[0039] Additionally, step count and stride length data can be utilized to measure and monitor an athlete's performance. Detailed analysis of step count and stride length during target assessments of an athlete's activities can be provided to the athlete or their trainer or coach. The data can then be used to inform training plans to improve the athlete's performance.

[0040] Additionally, step count and stride length data can be utilized to measure and manage physical therapy performance. Detailed analysis of step count and stride length during targeted assessments of supervised and unsupervised therapy session activities can be provided to patients or their therapists or physicians. The data can then be used to inform treatments to improve recovery programs.

[0041] User Device 2 is a block diagram of a typical user device 100 used in the above-described system. As shown, in this case, the user device 100 includes an accelerometer 102 that provides acceleration data to at least one central processing unit (CPU) 108. The operation of the CPU 108 is controlled by software instructions stored in memory 106. As shown, the software instructions include an operating system 106-1 and a motion analysis application 106-2. The acceleration data from the accelerometer 102 is processed by the motion analysis application 106-2 to calculate gait data for the subject.

[0042] User device 100 further comprises a communication interface 110 that communicates subject data determined by motion analysis application 106-2 to central server 140, and a user interface 112 that includes a keypad 112-1 and a display 112-2 and enables subject interaction with user device 100. Display 112-2 may display one or more icons configured to provide information to the user, and / or one or more of the time, date, number of steps, activity specific icon (walking, running, cycling, etc.), duration of the activity, reminder messages and / or instructions regarding the activity, network connection status, remaining battery power, and any other useful information to be displayed to the user.

[0043] Acceleration Data Analysis Before describing how motion analysis application 106-2 processes acceleration data, a conventional method is described with reference to FIG. 3, in which a fitness tracker or the like processes acceleration data to determine steps taken by a user.

[0044] In step 310, data is received from the accelerometer 102. The acceleration data includes a series of data points indexed by time. A data point (reading) from the accelerometer at time t represents acceleration measurements (A) in three orthogonal directions Ax, Ay, and Az. Ax (t), A Ay (t), A Az (t)). Ax, Ay, and Az are aligned with (defined by) the orientation of the accelerometer 102, not the orientation of the person carrying the accelerometer or any other geographic coordinate system. Typically, readings from the accelerometer 102 are provided in units of g, where g is the acceleration due to gravity at the Earth's surface (9.8 m / s 2 ). The sampling rate (the rate at which accelerometer 102 provides acceleration readings) varies between accelerometers and is often configurable, but to be useful for gait analysis, the sampling rate should be at least 20 Hz, and preferably higher (e.g., 30 Hz or 100 Hz).

[0045] After receiving the acceleration data, conventional devices low-pass filter the data to remove high-frequency fluctuations in the acceleration data that are not associated with the user's walking movements. The low-pass filter typically has a cutoff frequency of approximately 10 Hz. In step 320, the time series acceleration data is processed to identify periods of walking from other periods in which the user is not walking. There are various ways in which this determination can be made. Typically, a conventional method for separating periods of walking from other periods compares the magnitude of the acceleration data to a threshold to identify periods of activity that may correspond to walking. The magnitude of the acceleration data provided at time t is calculated as follows: A mag (t)=sqrt(A Ax (t) 2 +A Ay (t) 2 +A Az (t) 2 ) The periods thus identified are then analyzed to determine whether the periodic pattern corresponds to a gait pattern (i.e., coincides with a typical stride or step period). In step 325, an autocorrelation analysis is performed to detect periodic patterns in the time-series magnitude data calculated in step 320. Specifically, the autocorrelation unit 106-2-3 calculates the autocorrelation of the time-series magnitude data M(t) obtained in each separated gait period (or gait section) calculated in step 320. That is, the autocorrelation unit 106-2-3 calculates the following: JPEG0007819122000001.jpg27110where AC(k) is the autocorrelation at lag k, and A mag where (n) is the accelerometer magnitude at time n within an isolated gait section, and T is the number of magnitude values ​​within the isolated gait section. An autocorrelation function is calculated for each gait section. Thus, if step 320 separates 20 gait sections, then in step 325 autocorrelation unit 106-2-3 calculates 20 autocorrelation functions, i.e., one for each isolated gait section.

[0046] In step 330, each autocorrelation function calculated in step 325 is analyzed to determine the lag at which the highest peak after the zero-lag peak is found in the autocorrelation function. The calculated lag corresponds to either the user's stride period or the user's step period. To illustrate this analysis, FIG. 4 is a plot representing the autocorrelation function determined for one of the gait sections isolated in step 325. The autocorrelation function is symmetric about zero lag (k=0), and only the portion corresponding to non-negative lags is shown in the plot. Peaks corresponding to the stride period and step period are indicated by circles and squares, respectively. A typical stride period is between 1.0 and 1.2 seconds (100-120 steps per minute), with a typical step period being half this value.

[0047] The portion of the autocorrelation function calculated between zero lag and the first point where the autocorrelation function is less than zero is considered to be the zero-lag peak. The delay of the highest peak in the autocorrelation function after the zero-lag peak is considered to be either the step period or stride period. Because the autocorrelation function is calculated at a defined number of lags, the calculated autocorrelation value may not include the autocorrelation value at exactly the peak. A more accurate estimate of the delay corresponding to the peak of the autocorrelation function can be determined using interpolation. For example, a more accurate calculation of the delay value corresponding to the highest peak can be achieved by fitting a second-order polynomial to the calculated peak value and its neighbors on either side, and taking the peak of the polynomial function as the peak of the autocorrelation function.

[0048] In step 340, it is determined whether the delay corresponding to the identified highest peak corresponds to the user's step period or stride period. Depending on the symmetry of the user's gait and the location of the accelerometer, the delay calculated in step 330 (or step 335) may correspond to a step period or a stride period. For example, assuming a subject's gait is symmetrical, if the accelerometer is worn / held centrally by the user, e.g., a phone held in front of the chest or a device attached to the user's waist, a left-foot step and a right-foot step will produce very similar acceleration magnitudes in the accelerometer, and the calculated delay may correspond to a step period. On the other hand, if the accelerometer is worn on the ankle or wrist, left and right steps will produce significantly different acceleration data, and the calculated delay may correspond to a stride period.

[0049] In the example autocorrelation function shown in FIG. 4, the peak of the autocorrelation function at a 0.5 second delay is approximately the same height as the peak of the autocorrelation function at a 1.0 second delay; slight variations in the acceleration data may change which peak is tallest and therefore which peak is identified as the highest peak in step 330.

[0050] To determine whether the highest peak found in the autocorrelation function corresponds to a stride period or a step period, conventional fitness devices compare the determined delay with a threshold. For a particular individual at a particular time, the step period is half the stride period (assuming the step period of the right foot and the step period of the left foot are the same). Therefore, if the determined delay is below a threshold (e.g., 0.8 seconds), the highest peak can be assumed to correspond to a step period. If the determined delay is above the threshold, the highest peak can be assumed to correspond to a stride period.

[0051] The determined step period / stride period calculated for each isolated walking period is used to calculate various characteristics of the user's walking, such as the number of steps taken, the length of time the user has walked, etc., and this information is output (typically displayed) to the user and / or to a central server.

[0052] However, across a given population, there is overlap between stride periods and step periods, and some individuals' step periods may be longer than others'. Therefore, using thresholding to attempt to determine whether a calculated delay period corresponds to a step period or a stride period is imperfect and introduces errors. Calibrating the user device to the individual carrying the device or providing additional information about the individual (e.g., height) can help reduce these errors. However, even for a particular individual, there is overlap between the individual's stride period and step period, depending on the individual's gait (e.g., running vs. walking) at any given time. Thus, in relevant cases, using threshold calculations to determine whether a calculated delay period corresponds to a step period or a stride period can lead to erroneous conclusions, which affect the accuracy of the resulting step count. For example, if a calculated delay period corresponds to a step period but is actually determined to correspond to a stride period, the calculated number of steps will be half the true value, which may have ramifications for estimations of other parameters, such as travel speed and travel distance. Conversely, if the calculated delay period corresponds to a stride period, while it is determined that it actually corresponds to a step period, then the calculated number of steps is twice the true value.

[0053] Measurement and Analysis Applications In order to reduce at least some of these errors in conventional systems and to determine more accurate step and / or stride information from acceleration data, measurement analysis application 106-2 has been developed. The manner in which measurement analysis application 106-2 operates in this embodiment is described in detail below.

[0054] 2 and 5, the measurement analysis application 106-2 receives time series measurement data from the accelerometer 102 in step 505. As previously mentioned, acceleration is calculated along three orthogonal directions Ax, Ay, and Az defined by the orientation of the accelerometer 102, and the acceleration data from the accelerometer at time t is expressed as (A Ax (t),A Ay (t),A Az (t)). Thus, each acceleration data point effectively defines a vector that defines the resultant direction of acceleration experienced by the accelerometer 102 at measurement time t. An optional low-pass filter 106-2-1 filters the time series measurement data points received from the accelerometer in step 510 to remove high frequency fluctuations in the accelerometer measurements that are not associated with the user's walking motion. The cutoff frequency of the low-pass filter is typically between 8 Hz and 20 Hz, and preferably about 10 Hz.

[0055] In step 515, walking period detection unit 106-2-2 processes the acceleration data to detect periods when the user is walking or running. As mentioned above, there are various ways in which these periods can be detected. In a typical situation, isolated walking periods are approximately 10 to 20 seconds in length. If a longer walking period is detected, the longer period is generally divided into several sections, each of which is typically 10 to 20 seconds in length.

[0056] Transformation unit 106-2-3 then processes the acceleration data to predict measurements with respect to a coordinate reference system defined by the user's walking direction. Specifically, the z-axis is aligned with the vertical direction, the y-axis is aligned with the user's direction of travel, and the x-axis is aligned with the horizontal direction transverse to the direction of travel. In this embodiment, this is achieved as follows:

[0057] 1) The average acceleration vector over a period of time (a few seconds) is determined in step 520. JPEG0007819122000002.jpg32101 where A(n) is the acceleration data point at time n, and N is defined by the accelerometer sample rate and the time over which the average is calculated. Gravity is the largest static component of acceleration measured by accelerometer 102. Other accelerations experienced by the accelerometer include forward, backward, and sideways accelerations that, when averaged over time, cancel each other to some extent. As a result, the average vector calculated in step 520 identifies a vector direction.

[0058] 2) In step 525, the transform unit 106-2-3 uses the determined mean vector to perform a first transform that predicts each acceleration data point from the accelerometer (A(t) - post-low-pass filter, if low-pass filtered) on the horizontal plane as follows: A proj (t)=A(t)-(A(t)·A mean U )A mean U In the formula, A mean U is the unit vector of the average acceleration vector determined in step 520. The z-axis of the acquired predicted data points is aligned with the vertical axis, while the predicted y-axis of the accelerometer is poorly aligned with the direction of travel (forward and backward), and the predicted x-axis of the accelerometer is poorly aligned with the direction transverse to the direction of travel (sideways).

[0059] 3) In step 530, the transform unit 106-2-3 effectively calculates the rotation that needs to be applied to the predicted acceleration data to align the accelerometer's predicted x-axis and y-axis with the desired lateral and forward / backward directions, respectively. This rotation angle can be determined in a variety of ways. In this embodiment, the transform unit 106-2-3 performs a principal component analysis (PCA) on the predicted data (after setting the z-axis values ​​at the predicted data points to zero). The PCA analysis identifies two orthogonal directions in the horizontal plane that have the greatest and least variability. The direction with the greatest variability typically corresponds to forward / backward (y-) movement, and the direction with the least variability typically corresponds to lateral (x-) movement. The orthogonal directions identified by the PCA analysis effectively define the rotation in the horizontal plane that needs to be applied to the predicted data points to align the accelerometer's predicted x-axis and y-axis with the desired lateral and forward / backward directions, respectively.

[0060] 4) In step 535, the transform unit 106-2-3 applies the rotation determined in step 530 to the predicted acceleration data obtained in step 525, thereby generating transformed acceleration data points A that identify the vertical (z-axis), forward-backward (y-axis), and lateral (x-axis) accelerations for the acceleration data at time t. rot proj (t) is generated.

[0061] In step 540, the autocorrelation unit 106-2-4 calculates the autocorrelation function of the vertical acceleration data (z-axis data), the autocorrelation of the forward-backward direction acceleration data (y-axis data), and the autocorrelation of the lateral direction acceleration data (x-axis data) for each of the separated walking periods identified by the walking period determination unit 106-2-1. That is, the following autocorrelations are calculated: JPEG0007819122000003.jpg29146JPEG0007819122000004.jpg26146JPEG0007819122000005.jpg24146JPEG0007819122000006.jpg25164

[0062] It should be noted that the above processing may result in switched x and y data, i.e., the y axis data may actually correspond to lateral acceleration measurements and the x axis data may correspond to forward / reverse acceleration measurements. However, this is not a problem and will become clear from the following discussion.

[0063] Figure 6 illustrates plots of three autocorrelation functions calculated for one of the isolated gait periods for delays (k) between 0 and 4 seconds. In general, the autocorrelation function (AC) in the z (vertical) direction z ) is the autocorrelation function (AC x ) and the autocorrelation function (AC y ), each autocorrelation function shown in FIG. 6 is adjusted to unity at zero lag for each comparison.

[0064] As can be seen from Figure 6, the autocorrelation function (AC z ) has a plot similar to the magnitude autocorrelation of the original acceleration data (as shown in Figure 4), with strong peaks at both 0.5 seconds (step period) and 1.0 seconds (stride period). However, the AC after the zero-delay peak z The highest peak in may correspond to the step period or stride period. x and A.C. y ) has a prominent peak at 1.0 seconds (the stride period). The data used to create the example autocorrelation function shown in FIG. 6 was obtained from a wrist-worn user device, and the autocorrelation function AC y For devices worn on the body, the autocorrelation function AC yHowever, the peak of the autocorrelation function AC x The peak is missed in 0.5 seconds (step period) regardless of how the user device is carried or worn.

[0065] Table 1 below shows the vertical autocorrelation function (AC) at delays corresponding to stride period and step period for various user device / accelerometer wearing positions. z ) and the forward-backward autocorrelation function (AC y ) and the horizontal autocorrelation function (AC x ) and whether or not peaks may exist.

[0066] [Table 1]

[0067] AC x may exhibit troughs at the step period. However, AC x The absence of a peak in the AC after the zero delay peak without using a threshold. z can be used to distinguish whether the highest peak in corresponds to a stride period or a step period.

[0068] Specifically, in step 545, the analyzer 106-2-5 analyzes the autocorrelation values ​​AC obtained for the vertical direction to identify the delay corresponding to the maximum peak after the zero delay peak. z As is conventional, the optional interpolator 106-2-6 may utilize interpolation using a polynomial to determine a more accurate estimate of the delay corresponding to this maximum peak. Then, in step 550, the step / stride determiner 106-2-7 calculates the autocorrelation function (AC) for the x and y directions. x and A.C. y ) also has a peak at the delay identified in step 545. x and AC yIf both AC and AC also have (surrounding) peaks at this delay, the step / stride determining unit 106-2-7 determines that the delay identified in step 545 corresponds to the user's stride period. x and AC y If only one of AC and AC has a peak at the identified delay (or neither), the step / stride determining unit 106-2-7 determines in step 545 that the identified delay corresponds to the user's step period. x and AC y There are a variety of different ways to determine whether the delays identified in step 545 exhibit a "peak" (i.e., a sample with a higher autocorrelation than its neighbors on either side) that is often at or near the delay identified in step 545. In other embodiments, the AC at the determined delay is x or AC y If is higher than a threshold, which can be related to the autocorrelation at zero or zero lag, then AC x or AC y is considered to be the peak. Note that this approach does not depend on the assumption that y corresponds to the forward-backward direction and x corresponds to the lateral direction, but the approach is still valid if x corresponds to the forward-backward direction and y corresponds to the lateral direction.

[0069] Once the step / stride determiner 106-2-7 determines whether the delay identified in step 545 corresponds to the user's step period or the user's stride period, the motion analysis application 106-2 can calculate the number of steps taken by the user during the walking period in step 555. This information is then output in step 560. The step count may be output to the user on the display 112-2 and / or the step count may be transmitted to the user along with an identifier that identifies the user in association with the central server 140 and other relevant gait data for use in the clinical trial.

[0070] Variations and Alternatives The detailed embodiment has been described above. Note that various modifications and changes are possible to the above embodiment. Some of these alternative examples will be described below.

[0071] In the above-described embodiment, there is an implicit assumption that the orientation of the accelerometer 102 remains the same (constant) within each isolated gait period determined in step 515 and over the period over which the autocorrelation function is calculated. It also assumes that gait characteristics (particularly step period / stride period) remain relatively constant over the isolated gait periods. These assumptions may not be correct, especially during longer isolated gait periods. To address this issue, the isolated gait periods may be divided into smaller subsections or epochs (which may or may not overlap in time), and the above-described analysis from step 520 is performed on each smaller subsection of acceleration data. The duration of each subsection must be at least 3 seconds long to encompass several strides. If rotations between subsections are calculated in step 530, the PCA analysis may cause the determined rotation to change abruptly from one subsection to the next. Interpolation may be used (e.g., using a 4D display or other means) to smoothly transition between the rotations of adjacent subsections. Data from other sensors (particularly gyroscopes that may be included on the user device) may help determine changes in the user device's orientation, i.e., the rotational changes required to align the acceleration data with the user's direction of travel.

[0072] In the above embodiment, measurements from the accelerometer were resolved into a vertical direction (z) and x and y directions corresponding to the user's forward and lateral directions. Autocorrelation functions were then calculated for the x, y, and z measurements. In an alternative embodiment, instead of determining the autocorrelation function of the acceleration data in the z direction, autocorrelation may be performed on the magnitude of the acceleration data (either before or after transformation). To resolve the ambiguity of whether the highest peak after the zero-lag peak in the autocorrelation function of the magnitude data corresponds to the user's stride period or the user's step period, the autocorrelation functions for the x and y directions are still calculated and used as usual.

[0073] In the above-described embodiment, the analyzer 106-2-5 used principal component analysis (PCA) to calculate the rotation required to align the accelerometer's predicted x-axis and predicted y-axis with the user's walking direction and lateral direction. Instead of using PCA to determine this rotation, a satellite navigation system (such as a GPS system) installed in the user device may provide the geographic direction the user is walking, and a compass within the user device may provide the device's orientation relative to the Earth's axes. From this, the analyzer 106-2-5 may calculate the rotation required to map the acceleration data from the accelerometer onto the user's walking frame of reference (where y corresponds to the direction the user is walking, x is transverse to y in the horizontal plane, and z is vertical).

[0074] Alternatively, if the device has a fixed, known orientation relative to the user's direction of travel, for example, if the device is held pointing in the direction of travel, the rotation required to map acceleration data from the accelerometer onto the user's frame of reference may already be known.

[0075] The x, y, and z autocorrelation functions calculated in the above embodiments can also be useful for distinguishing between walking and other activities. For example, a user with a wrist-worn user device may swing their arm, fooling a walking detection algorithm and misinterpreting the arm swing as walking if the swing period is comparable to a typical stride period. The autocorrelation data determined in the x, y, and z directions can be used to confirm that the walking periods are in fact walking periods, rather than the user moving the device in an attempt to mimic walking motion.

[0076] FIG. 7 is a flow chart illustrating how the system can more accurately determine whether a period of exercise corresponds to a period of walking or some other user exercise that attempts to mimic walking.

[0077] In step 710, the device z 5. Determine whether the highest peak after the zero-lag peak in (or in the autocorrelation of the magnitude acceleration data) corresponds to the user's stride period or step period (this effectively corresponds to the determination in step 550 or step 555 in FIG. 5). If the highest peak corresponds to the user's stride period, processing proceeds to step 715; if the highest peak corresponds to a step period, processing proceeds to step 740.

[0078] In step 715, the gait determination unit 106-2-2 calculates the autocorrelation function (AC) in the z (vertical) direction to determine whether it has a peak at a delay that is half the delay corresponding to the stride period. z In step 720, the peak is AC z If the peak is found in the AC z If not, the walking determination unit 106-2-2 determines in step 735 that the user is not actually walking during this period.

[0079] In step 725, the walking determination unit 106-2-2 determines the maximum x and ACy The autocorrelation functions (AC) in the x and y directions are then calculated to determine whether one of the two also contains a peak at the half-step period. x and A.C. y ) Check the AC x and AC y If both AC and AC contain peaks in the half-stride period, the process proceeds to step 735 where the walking determination unit 106-2-2 again determines that the movement in the current period is not actually walking. x and AC y If only one of the two has a peak in the half-stride period (or neither), the process proceeds to step 730 where the walking determination unit 106-2-2 confirms that the user is actually walking in the current period.

[0080] AC z If the highest peak after the zero-delay peak in corresponds to the user's step period, in step 740, the gait determination unit 106-2-2 calculates the autocorrelation function (AC z ) is processed. The peak is twice the step period. z If not, then in step 755, walking determination unit 106-2-2 determines that the user is not actually walking during this period.

[0081] In step 750, the walking determination unit 106-2-2 determines the AC x and AC y The autocorrelation functions (AC) in the x and y directions are then calculated to determine whether both also contain a peak at twice the step period. x and A.C. y ) Check the AC x and AC y If neither AC nor AC contains a peak at twice the step period, the process proceeds to step 755 where the walking determination unit 106-2-2 again determines that the movement in the current period is not actually walking. x and AC yIf both of the step period and the step period have peaks at twice the step period, the process proceeds to step 760 where the walking determination unit 106-2-2 confirms that the user is walking in the current period.

[0082] In the above-described embodiment, the step / stride determiner considered the presence or absence of peaks in the forward / backward direction (y-direction) and the lateral direction (x-direction) to determine whether the delay period of the highest peak after the zero-delay peak corresponds to a step period or a stride period. A preferred technique counts the peaks of various autocorrelation functions at the same delay. This helps avoid any errors caused by mixing the forward / backward direction with the lateral direction. In other embodiments, the device analyzing the acceleration data may simply assume that the determined lateral direction (x-direction) is correct and may determine whether the identified delay corresponds to a step period or a stride period based on whether the autocorrelation function for the x-direction contains a peak at the identified delay. If the autocorrelation function for the x-direction contains a peak at the identified delay, the identified delay corresponds to a stride period; if the autocorrelation function for the x-direction does not contain a peak at the identified delay, the identified delay corresponds to a step period.

[0083] In the above-described embodiment, acceleration data obtained from the accelerometer was analyzed by looking at the autocorrelation function of the data in various directions. Autocorrelation analysis is suitable for highlighting periodic changes in acceleration data caused by repetitive movements such as walking or running. Other types of analysis may be performed to identify these periodic changes (and their duration). For example, a Fourier transform (or other frequency analysis such as a discrete cosine transform) may be determined and analyzed to identify peaks in the frequency domain that represent step or stride periods.

[0084] Similarly, in the above-described embodiment, acceleration data from the accelerometer was transformed from the accelerometer's coordinate reference frame to the user's coordinate reference frame, and autocorrelation was performed on the transformed acceleration data. In an alternative embodiment, this transformation of coordinate systems may be performed after the autocorrelation function is calculated. Thus, the original acceleration data defining the accelerometer's acceleration in the Ax, Ay, and Az directions may first be autocorrelated, and the autocorrelation transformed to account for the change in reference frame.

[0085] Additionally, if a user device includes multiple accelerometers, data from each accelerometer may be analyzed and the results combined (e.g., averaged) to calculate a more accurate or less noisy step and / or stride period. Similarly, if a user carries multiple devices (such as a mobile phone) and an actigraph device, and both devices have accelerometers, the system can use data from both accelerometers to determine the step and / or stride period. The measurements from the two (or more) accelerometers may then be averaged again to improve the signal-to-noise ratio, or measurements from one accelerometer may be used to verify or validate the step and / or stride period determined from acceleration data obtained from the other accelerometers.

[0086] As mentioned above, the calculated peak of the autocorrelation function depends on whether the accelerometer is worn or carried on the user's core, or on the user's wrist or ankle. The inventors have realized that by comparing the amplitude of the autocorrelation function peak at the step delay and the stride delay, it is possible to determine how the user is holding or wearing the accelerometer. Specifically, the user device or central server can compare the peak values ​​of the autocorrelation function at delays corresponding to the step period and the stride period (which can be determined using the above-mentioned invention or other prior art). If the two peak values ​​are identical or similar to each other, or if the peak at the step period is larger than the peak at the stride period, the user device or central server can determine whether the accelerometer is worn or carried on the core. The inventors have found that in this situation, the peaks can be considered similar to each other if they are 10% to 15% (or less) of each other. However, if the peaks in the stride period are 20% or more greater than the peaks in the step period, the user device or a central server can determine that the accelerometer is likely worn or carried peripherally by the user (on the user's hand or on the user's wrist or ankle). Similarly, if a frequency analysis is performed on the acceleration data instead of an autocorrelation analysis, a comparison of the magnitude of the peaks in the frequency plot at frequencies corresponding to the step frequency and the stride frequency can be used to determine whether the accelerometer is worn / carried centrally or peripherally by the user (e.g., on the user's wrist or ankle). Knowing whether the accelerometer device is worn centrally or peripherally is useful because knowledge of this information can aid in the interpretation of the acceleration data and the adjustment of thresholds used. For example, if the acceleration data is used to determine a user's activity level, different thresholds may be used depending on whether the device is worn centrally or peripherally by the user.In particular, if a user is sitting writing or typing at a desk, a wrist-worn device may experience more movement than a belt-worn device, and this difference needs to be taken into account when the central server attempts to determine how active the user is. While wearing positions vary, it is reasonable to assume that the wearing positions determined while the user is walking are the same as on days when the user is likely sedentary. Furthermore, some algorithms used to determine walking distance and other gait parameters (such as parameters characterizing freezing of gait in Parkinson's disease patients) assume that the accelerometer device is worn on the ankle, and would be less accurate or require recalibration if the device were worn centrally. Therefore, knowledge of the wearing position of the accelerometer device allows such recalibration to be performed, improving the consistency and accuracy of the scale and allowing data to be recognized as less accurate in the alternative.

[0087] The accelerometer's wearing position may be determined using the autocorrelation function (frequency function) of the original acceleration data (its components) or the magnitude of the acceleration data. If the aforementioned technique is used to decompose the acceleration data into the user's frame of reference (vertical, lateral, forward, and backward directions), the accelerometer's wearing position can also be determined by comparing the dominant period of lateral (sideways) acceleration with the dominant period of the vertical component of the acceleration data or acceleration magnitude. If the dominant periods are identical (or within an acceptable tolerance level), the user device or central server can determine that the accelerometer is likely worn / carried on the user's limb (e.g., the user's wrist or ankle). The inventors have found that acceptable tolerance levels in this case are within 10% to 15%, respectively. The dominant period is the delay corresponding to the maximum peak of the autocorrelation function (after the zero-lag peak), which in this scenario corresponds to the stride period. A similar analysis can be performed on a frequency plot of the decomposed acceleration data.

[0088] However, if the dominant period of the lateral acceleration is approximately twice the dominant period of the vertical acceleration component or acceleration magnitude, and the lateral acceleration component has little or no energy during the dominant period of the vertical acceleration component or acceleration magnitude, the user device or central server can determine that the accelerometer is likely carried / worn in a more central location on the body, e.g., on the user's chest or waist. (In this scenario, the dominant period of the lateral component is the stride period, while the dominant period of the vertical component / magnitude is the step period.)

[0089] In the above-described embodiment, the software application for processing the acceleration data resides within the user device. The same or similar software may be provided to a central server computer so that the central server performs the above-described step / stride analysis. This software application may be provided as computer-implementable instructions on a carrier signal or on a tangible computer-readable medium. Alternatively, the functionality of the software application may be defined in hardware circuitry, such as within an FPGA or ASIC device.

[0090] It will be understood from the above description that many features of the different embodiments are interchangeable and combinable. The present disclosure extends to additional embodiments that include features from different embodiments combined together in a manner not specifically mentioned. Indeed, it will be apparent to those skilled in the art that there are many features presented in the above embodiments that can be advantageously combined with each other.

[0091] In the above-described embodiments, various delays or periods were calculated. As will be appreciated by those skilled in the art, determining such periods occurs once the corresponding frequencies are determined. Specifically, in the above-described embodiments, a step period and a stride period are determined. Clearly, when a step frequency or stride frequency is determined, this implies that the corresponding step period or stride period is determined as well, and the claims are intended to cover determining such frequencies instead of or in addition to the periods / delays.

[0092] This application also includes the following numbered sections defining various aspects of the present invention:

[0093] Item 1 1. An apparatus for determining motion information of a user carrying an accelerometer while moving, comprising: comprising one or more processors and a memory; The processor and memory receiving acceleration data from the accelerometer defining acceleration experienced by the accelerometer due to movement of the user, the acceleration being defined relative to a frame of reference associated with the accelerometer; applying a transformation to the acceleration data or data derived from the acceleration data that converts the frame of reference to a user's frame of reference that includes the user's direction of travel and a lateral direction transverse to the user's direction of travel; analyzing the acceleration data or data derived from the acceleration data to determine a period corresponding to either a stride period or a step period of the user; using information about the lateral acceleration to clarify whether the determined period corresponds to a user stride period or a user step period; It is configured as follows: Device.

[0094] Section 2 the processor and memory are configured to use information about the lateral and forward accelerations to determine whether the determined period corresponds to a user stride period or a user step period. Item 1. The device according to item 1.

[0095] Section 3 the processor and memory are configured to determine a first autocorrelation function to determine the period corresponding to either the stride period or the step period of the user. Item 1 or 2. The device according to item 1 or 2.

[0096] Section 4 the processor and memory are configured to process the first autocorrelation function to identify a peak in the first autocorrelation function at an autocorrelation delay corresponding to a user stride period or a user step period; Item 3. The device according to item 3.

[0097] Section 5 the processor and memory are configured to process the first autocorrelation function to identify a highest peak in the first autocorrelation function after the zero lag peak and determine a period corresponding to the user's stride period or the user's step period as the autocorrelation lag associated with the identified highest peak; Item 4. The device according to item 4.

[0098] Section 6 the processor and memory are configured to determine a second autocorrelation function of the lateral acceleration and to define whether a period corresponds to a stride period or a step period based on whether the second autocorrelation function includes a peak near an autocorrelation lag corresponding to a step period or a stride period. Item 4 or 5. The device according to item 4 or 5.

[0099] Section 7 the processor and memory are configured to determine a second autocorrelation function of the lateral acceleration and a third autocorrelation function of the headway acceleration, and to define whether a period corresponds to a stride period or a step period based on whether the second autocorrelation function includes a peak near an autocorrelation lag corresponding to a step period or a stride period. Item 4, 5, or 6. The device according to item 4, 5, or 6.

[0100] Item 8 the processor and memory are configured to use the first autocorrelation function, the second autocorrelation function, and the third autocorrelation function to determine whether the user is walking or running, or whether the user is not walking or running. Item 7. The device according to item 7.

[0101] Section 9 the first autocorrelation function is calculated for the acceleration data or for data derived from the acceleration data; Item 9. The device according to any one of items 3 to 8.

[0102] Section 10 the first autocorrelation function is calculated for transformed acceleration data defining acceleration in a user's frame of reference; Item 10. The device according to any one of items 3 to 9.

[0103] Section 11 the processor and memory are configured to determine and apply a first transformation that aligns a first axis of the acceleration data or data derived from the acceleration data with a vertical axis; Item 11. The device according to any one of items 1 to 10.

[0104] Section 12 the processor and memory are configured to determine and apply a second transformation that aligns a second axis of the acceleration data or data derived from the acceleration data with the direction of travel and a third axis of the acceleration data or data derived from the acceleration data with the lateral direction; Item 12. The device according to item 11.

[0105] Section 13 the second transformation includes a rotation; Item 13. The device according to item 12.

[0106] Section 14 The processor and memory determining that the determined period corresponds to a stride period of the user if the information about the lateral acceleration matches the information about the forward acceleration; determining that the determined period corresponds to a user step period if the information about the lateral acceleration does not match the information about the forward acceleration; It is configured as follows: Item 14. The device according to any one of items 1 to 13.

[0107] Section 15 The user's frame of reference includes a vertical direction transverse to both the forward and lateral directions. Item 15. The device according to any one of items 1 to 14.

[0108] Section 16 the processor and memory are configured to process the acceleration data to identify periods of walking or running within the acceleration data, and to use the acceleration data from within the identified periods of walking or running to determine the periods corresponding to either a stride period or a step period of the user; Item 16. The device according to any one of items 1 to 15.

[0109] Section 17 The heading and lateral directions are identified as the directions in the horizontal plane having the greatest and least variability in the received acceleration data; Item 17. The device according to any one of items 1 to 16.

[0110] Section 18 The processor and memory are further configured to use the identified step period or stride period to determine a user's step count for exercise corresponding to walking or running. Item 18. The device according to any one of items 1 to 17.

[0111] Section 19 1. An apparatus for determining motion information of a user carrying an accelerometer while moving, comprising: comprising one or more processors and a memory; The processor and memory receiving acceleration data from the accelerometer defining acceleration experienced by the accelerometer due to movement of the user, the acceleration being defined relative to a frame of reference associated with the accelerometer; applying a transformation to the acceleration data or data derived from the acceleration data that converts the frame of reference to a user's frame of reference that includes the user's direction of travel and a lateral direction transverse to the user's direction of travel; determining a first autocorrelation function of the acceleration data or data derived from the acceleration data; determining a second autocorrelation function of the forward acceleration; determining a third autocorrelation function of the lateral acceleration; configured to determine whether the user is walking, not walking, running, or not running using the first autocorrelation function, the second autocorrelation function, and the third autocorrelation function; Device.

[0112] Section 20 1. An apparatus for determining motion information of a user carrying an accelerometer while moving, comprising: comprising one or more processors and a memory; The processor and memory receiving acceleration data from the accelerometer, the acceleration data including acceleration values ​​for each of a plurality of first orthogonal directions defined by an orientation of the accelerometer, the acceleration values ​​each indicating an acceleration of the accelerometer in one of the first orthogonal directions at a given time point; converting the acceleration data into converted acceleration data including acceleration values ​​with respect to a plurality of second orthogonal directions defined by the orientation of the user for each of a plurality of time points, the acceleration values ​​each indicating an acceleration movement of the accelerometer in one of the plurality of second orthogonal directions including a direction of travel of the user and a lateral direction transverse to the direction of travel of the user; analyzing at least a portion of the acceleration data or the transformed acceleration data to determine a period corresponding to either a stride period or a step period of the user; using the transformed acceleration data relating to the user's movement in at least the lateral direction to clarify whether the determined period corresponds to a user's stride period or a user's step period; It is configured as follows: Device.

[0113] Section 21 The apparatus forms part of a user device carried by a user, The accelerometer forms part of or is configured to communicate with the user device; Item 21. The device according to any one of items 1 to 20.

[0114] Section 22 The at least one processor and memory are configured to acquire acceleration data from a plurality of accelerometers carried by the user when walking or running; The step period or stride period is determined using acceleration data from the plurality of accelerometers. Item 22. The device according to any one of items 1 to 21.

[0115] Section 23 the at least one processor and memory are configured to determine each step period or each stride period using acceleration data from each accelerometer, and to i) average the acquired step periods or stride periods, or ii) verify the step period or stride period determined from the acceleration data from one accelerometer using acceleration data or data derived from acceleration data acquired from the other accelerometer; Item 23. The device according to item 22.

[0116] Section 24 The accelerometers are mounted on the same user device carried by the user, or the accelerometers are mounted on different user devices carried by the user; Item 24. The device according to item 22 or 23.

[0117] Section 25 The accelerometers are mounted in different user devices carried by the user at different wearing positions; Item 25. The device according to item 24.

[0118] Section 26 1. An apparatus for determining motion information of a user carrying an accelerometer while moving, comprising: means for receiving acceleration data from the accelerometer defining acceleration experienced by the accelerometer due to movement of the user, the acceleration being defined relative to a frame of reference associated with the accelerometer; means for applying a transformation to the acceleration data or data derived from the acceleration data to convert the frame of reference to a user's frame of reference that includes the user's direction of travel and a lateral direction transverse to the user's direction of travel; means for analyzing the acceleration data or data derived from the acceleration data to determine a period corresponding to either a stride period or a step period of the user; means for using information about the lateral acceleration to clarify whether a determined period corresponds to a user stride period or a user step period; Equipped with Device.

[0119] Section 27 1. A method for determining motion information of a user carrying an accelerometer while moving, comprising: receiving acceleration data from the accelerometer defining acceleration experienced by the accelerometer due to movement of the user, the acceleration being defined relative to a frame of reference associated with the accelerometer; applying a transformation to the acceleration data or data derived from the acceleration data that converts that frame of reference to a user's frame of reference that includes the user's direction of travel and a lateral direction transverse to the user's direction of travel; analyzing the acceleration data or data derived from the acceleration data to determine a period corresponding to either a stride period or a step period of the user when the user is walking or running; using information about the lateral acceleration to determine whether the determined period corresponds to a user stride period or a user step period; Including, method.

[0120] Section 28 comprising computer-implementable instructions that cause a programmable computing device to be configured as the apparatus of any one of clauses 1 to 26; Tangible computer readable medium.

[0121] Section 29 1. A clinical trial system including a central computer in communication with a plurality of user devices, Each of the user devices is configured to collect acceleration data for movement of a user associated with the user device; the central computer or at least one user device analyzes the acceleration data; Item 27. The device according to any one of items 1 to 26 is provided. Clinical trial system.

[0122] Item 30 1. An apparatus for determining motion information of a user carrying an accelerometer while moving, comprising: comprising one or more processors and a memory; The processor and memory receiving acceleration data from the accelerometer defining acceleration experienced by the accelerometer due to movement of the user, the acceleration being defined relative to a frame of reference associated with the accelerometer; applying a transformation to the acceleration data or data derived from the acceleration data to convert its frame of reference to a user's frame of reference that includes a direction of travel of the user and a lateral direction transverse to the direction of travel of the user to provide transformed acceleration data or transformed data derived from the acceleration data; processing the transformed acceleration data or transformed data derived from the acceleration data to determine information about the lateral acceleration; analyzing the acceleration data or data derived from the acceleration data to determine a period corresponding to a maximum peak in an autocorrelation function or frequency function of the acceleration data or data derived from the acceleration data, wherein the determined period is ambiguous and corresponds to a stride period or step period of the user; using the information about the lateral acceleration to clarify whether the determined period corresponds to a user stride period or a user step period; It is configured as follows: Device.

[0123] Section 31 1. An apparatus for determining motion information of a user carrying an accelerometer while moving, comprising: comprising one or more processors and a memory; The processor and memory receiving acceleration data from the accelerometer defining acceleration experienced by the accelerometer due to movement of the user, the acceleration being defined relative to a frame of reference associated with the accelerometer; applying a transformation to the acceleration data or data derived from the acceleration data to convert its frame of reference to a user's frame of reference comprising i) acceleration data relating to acceleration in a direction of travel of the user, and ii) acceleration data relating to acceleration in a lateral direction transverse to the direction of travel of the user, to provide transformed acceleration data or transformed data derived from the acceleration data comprising said forward direction and said lateral direction; determining a first autocorrelation function of the acceleration data or data derived from the acceleration data; determining a second autocorrelation function of the acceleration data relating to the acceleration in the direction of travel included in the transformed acceleration data or the transformed data derived from the acceleration data; determining a third autocorrelation function of the acceleration data relative to the lateral acceleration included in the transformed acceleration data or the transformed data derived from the acceleration data; determining whether the user is walking, not walking, running, or not running using the first autocorrelation function, the second autocorrelation function, and the third autocorrelation function; It is configured as follows: Device.

[0124] Section 32 1. A method for determining motion information of a user carrying an accelerometer while moving, comprising: receiving acceleration data from the accelerometer defining acceleration experienced by the accelerometer due to movement of the user, the acceleration being defined relative to a frame of reference associated with the accelerometer; applying a transformation to the acceleration data or data derived from the acceleration data to convert its frame of reference to a user's frame of reference that includes a direction of travel of the user and a lateral direction transverse to the direction of travel of the user to provide transformed acceleration data or transformed data derived from the acceleration data; processing the transformed acceleration data or transformed data derived from the acceleration data to determine information about the lateral acceleration; analyzing the acceleration data or data derived from the acceleration data to determine a period of time corresponding to a maximum peak in an autocorrelation function or frequency function of the acceleration data or data derived from the acceleration data, the determined period of time being ambiguous and corresponding to either a stride period or a step period of the user when the user is walking or running; using said information about said lateral acceleration to clarify whether a determined period corresponds to a user stride period or a user step period; Including, method.

Claims

1. 1. An apparatus for determining motion information of a user carrying an accelerometer while moving, comprising: comprising one or more processors and a memory; The processor and the memory receiving acceleration data from the accelerometer defining acceleration experienced by the accelerometer due to movement of the user, the acceleration being defined relative to a frame of reference associated with the accelerometer; processing the acceleration data or data derived from the acceleration data, said processing comprising: applying a transformation to transform the frame of reference to the user's frame of reference that includes a direction of travel of the user and a lateral direction transverse to the direction of travel of the user to provide transformed data; processing the transformed data to determine information about the lateral acceleration; determining a period of time corresponding to a maximum peak in an autocorrelation function or a frequency function, the determined period of time corresponding to either a stride period of the user or a step period of the user; Including, using the information about the acceleration in the lateral direction to clarify whether the determined period corresponds to the stride period of the user or the step period of the user. It is configured as follows: An apparatus characterized in that

2. the processor and the memory are configured to use the information about the acceleration in the lateral direction and the forward direction to clarify whether the determined period corresponds to the stride period of the user or the step period of the user.

10. The apparatus of claim 1.

3. the processor and the memory are configured to determine a first autocorrelation function to determine the period corresponding to either the stride period or the step period of the user.

3. The device according to claim 1 or 2.

4. the processor and the memory are configured to process the first autocorrelation function to identify a peak in the first autocorrelation function at an autocorrelation delay corresponding to the stride period of the user or the step period of the user.

4. The apparatus of claim 3.

5. the processor and the memory are configured to process the first autocorrelation function to identify a highest peak in the first autocorrelation function after a zero-lag peak and determine the period corresponding to the stride period of the user or the step period of the user as the autocorrelation lag associated with the identified highest peak.

5. The apparatus of claim 4.

6. the processor and the memory are configured to determine a second autocorrelation function of the acceleration in the lateral direction and to define whether the period corresponds to the stride period or the step period based on whether the second autocorrelation function includes the peak near the autocorrelation lag corresponding to the step period or the stride period.

6. The device according to claim 4 or 5.

7. the processor and the memory are configured to determine a second autocorrelation function of the acceleration in the lateral direction and a third autocorrelation function of the acceleration in the direction of travel, and to define whether the period corresponds to the stride period or the step period based on whether the second autocorrelation function includes a peak near the autocorrelation lag corresponding to the step period or the stride period.

6. The device according to claim 4 or 5.

8. the processor and the memory are configured to use the first autocorrelation function, the second autocorrelation function, and the third autocorrelation function to determine whether the user is walking or running, or whether the user is not walking or running.

8. The apparatus of claim 7.

9. the first autocorrelation function is calculated for the acceleration data or for the data derived from the acceleration data; 9. Apparatus according to any one of claims 3 to 8.

10. the processor and the memory are configured to determine and apply a first transformation that aligns a first axis of the acceleration data or the data derived from the acceleration data with a vertical axis; 10. Apparatus according to any one of claims 1 to 9.

11. the processor and the memory are configured to determine and apply a second transformation to align a second axis of the acceleration data or the data derived from the acceleration data with the heading direction and a third axis of the acceleration data or the data derived from the acceleration data with the lateral direction.

11. The apparatus of claim 10.

12. The processor and the memory processing the transformed data to determine information about the forward acceleration; determining that the determined period corresponds to the stride period of the user if the information about the acceleration in the lateral direction matches the information about the acceleration in the direction of travel; configured to determine that the determined period corresponds to the step period of the user if the information about the acceleration in the lateral direction does not match the information about the acceleration in the forward direction.

12. Apparatus according to any one of claims 1 to 11.

13. The frame of reference of the user is: a vertical direction transverse to both the forward direction and the lateral direction; Including, 13. Apparatus according to any one of claims 1 to 12.

14. the processor and the memory are configured to process the acceleration data to identify periods of walking or running within the acceleration data, and to use the acceleration data from within the identified periods of walking or running to determine the periods corresponding to either the stride period or the step period of the user.

14. Apparatus according to any one of claims 1 to 13.

15. the heading direction and the lateral direction are identified as directions in a horizontal plane having maximum and minimum variability in the received acceleration data; 15. Apparatus according to any one of claims 1 to 14.

16. the processor and the memory are further configured to use the determined step period or stride period to determine a step count of the user for a movement corresponding to walking or running.

16. Apparatus according to any one of claims 1 to 15.

17. 1. An apparatus for determining motion information of a user carrying an accelerometer while moving, comprising: comprising one or more processors and a memory; The processor and the memory receiving acceleration data from the accelerometer defining acceleration experienced by the accelerometer due to movement of the user, the acceleration being defined relative to a frame of reference associated with the accelerometer; applying a transformation to the acceleration data or data derived from the acceleration data to convert the frame of reference to the user's frame of reference that includes a direction of travel of the user and a lateral direction transverse to the direction of travel of the user; determining a first autocorrelation function of the acceleration data or the data derived from the acceleration data; determining a second autocorrelation function of the acceleration in the direction of travel; determining a third autocorrelation function of the acceleration in the lateral direction; using the first autocorrelation function, the second autocorrelation function, and the third autocorrelation function to determine whether the user is walking, not walking, running, or not running; It is configured as follows: An apparatus characterized in that

18. The apparatus forms part of a user device carried by the user, the accelerometer forms part of the user device or is configured to communicate with the user device; 18. Apparatus according to any one of claims 1 to 17.

19. 1. A method for determining motion information of a user carrying an accelerometer while moving, comprising: receiving acceleration data from the accelerometer defining acceleration experienced by the accelerometer due to movement of the user, the acceleration being defined relative to a frame of reference associated with the accelerometer; processing the acceleration data or data derived from the acceleration data, said processing comprising: applying a transformation to convert the frame of reference to the user's frame of reference that includes a heading direction of the user and a lateral direction transverse to the heading direction of the user to provide transformed data; processing the transformed data to determine information about the lateral acceleration; determining a period of time corresponding to a maximum peak in an autocorrelation function or a frequency function, the determined period of time corresponding to either the user's stride period or the user's step period; Including, using the information about the acceleration in the lateral direction to clarify whether the determined period corresponds to the stride period of the user or the step period of the user; Including, A method characterized by:

20. An apparatus constituting part of a clinical trial system, a central computer in communication with a plurality of user devices; each of the user devices configured to collect the acceleration data for the user's movements associated with the user device or to communicate with the accelerometer; the central computer or at least one of the user devices; 10. The apparatus of claim 1, further comprising: Equipped with 18. Apparatus according to any one of claims 1 to 17.

21. 1. An apparatus for determining motion information of a user carrying an accelerometer while moving, comprising: comprising one or more processors and a memory; The processor and the memory receiving acceleration data from the accelerometer defining acceleration experienced by the accelerometer due to movement of the user; processing the acceleration data or data derived from the acceleration data to determine a step period or stride period of the user; Calculating a peak value of an autocorrelation function or a frequency function of the acceleration data or the data obtained from the acceleration data at the determined step period and stride period; comparing the peak value calculated at the step period with the peak value calculated at the stride period to determine whether the accelerometer is carried by the user centrally or peripherally; It is configured as follows: An apparatus characterized in that

22. the one or more processors and the memory are configured to determine that the accelerometer is carried in the central region of the user's body if the peak value calculated in the step period is greater than the peak value calculated in the stride period.

22. The device of claim 21.

23. the one or more processors and the memory are configured to determine that the accelerometer is carried on the periphery of the body of the user if the peak value calculated in the stride period is greater than the peak value calculated in the step period.

23. Apparatus according to claim 21 or 22.

24. 1. A method for determining motion information of a user carrying an accelerometer while moving, comprising: receiving acceleration data from the accelerometer defining acceleration experienced by the accelerometer due to movement of the user; processing the acceleration data or data derived from the acceleration data to determine a step period or stride period of the user; Calculating a peak value of an autocorrelation function or a frequency function of the acceleration data or the data obtained from the acceleration data at the determined step period and stride period; comparing the peak value calculated at the step period with the peak value calculated at the stride period to determine whether the accelerometer is carried by the user centrally or peripherally; using one or more processors and memory to Including, A method characterized by:

25. 1. An apparatus for determining motion information of a user carrying an accelerometer while moving, comprising: comprising one or more processors and a memory; The processor and the memory receiving acceleration data from the accelerometer defining acceleration experienced by the accelerometer due to movement of the user, the acceleration being defined relative to a frame of reference associated with the accelerometer; applying a transformation to the acceleration data or data derived from the acceleration data to convert the frame of reference to the frame of reference of the user that includes a vertical direction, a direction of travel of the user, and a lateral direction transverse to the direction of travel of the user; determining a time period corresponding to a maximum peak of an autocorrelation function or a frequency function of the acceleration data in the lateral direction; determining the time period corresponding to the maximum peak of the autocorrelation function or the frequency function of the vertical acceleration data or the acceleration magnitude data; using the determined period to determine whether the accelerometer is carried by the user centrally on the user's body or whether the accelerometer is carried by the user peripherally on the user's body; It is configured as follows: An apparatus characterized in that

26. the one or more processors and the memory are configured to determine that the accelerometer is carried on the core of the user's body when the period corresponding to the maximum peak of the autocorrelation function or the frequency function of the acceleration data in the lateral direction is approximately twice the period corresponding to the maximum peak of the autocorrelation function or the frequency function of the acceleration data in the vertical direction or the acceleration magnitude data.

26. The apparatus of claim 25.

27. the one or more processors and the memory are configured to determine that the accelerometer is carried in the periphery of the body of the user when the period corresponding to the maximum peak of the autocorrelation function or the frequency function of the acceleration data in the lateral direction is approximately the same as the period corresponding to the maximum peak of the autocorrelation function or the frequency function of the acceleration data in the vertical direction or the acceleration magnitude data.

27. Apparatus according to claim 25 or 26.

28. 1. A method for determining motion information of a user carrying an accelerometer while moving, comprising: receiving acceleration data from the accelerometer defining acceleration experienced by the accelerometer due to movement of the user, the acceleration being defined relative to a frame of reference associated with the accelerometer; applying a transformation to the acceleration data or data derived from the acceleration data to convert the frame of reference to the frame of reference of the user that includes a vertical direction, a direction of travel of the user, and a lateral direction transverse to the direction of travel of the user; determining a time period corresponding to a maximum peak of an autocorrelation function or a frequency function of the acceleration data in the lateral direction; determining a time period corresponding to the maximum peak of the autocorrelation function or the frequency function of the vertical acceleration data or the acceleration magnitude data; using the determined period to determine whether the accelerometer is carried by the user centrally on the user's body or whether the accelerometer is carried by the user peripherally on the user's body; using one or more processors and memory to Including, A method characterized by:

Images