NL2031061B1

NL2031061B1 - Multiple algorithms for controlling a cueing device.

Info

Publication number: NL2031061B1
Application number: NL2031061A
Authority: NL
Inventors: Waardenburg Floor; Van Der Ent Martijn; Leopold Marie Minnoye Alexander
Original assignee: Cue2Walk Int B V
Priority date: 2022-02-24
Filing date: 2022-02-24
Publication date: 2023-09-06

Abstract

A cueing device for producing a cue in assisting a user during a motion. The cueing device comprises a control unit operatively connected with a motion sensor and a cueing unit. The 5 control unit is programmed with multiple algorithms providing multiple outcomes to determine a gait irregularity during the motion. A cue is only produced in case outcomes of at least two algorithms positively indicate for a gait irregularity. Cue production should be minimised to situations in which a cue is really necessary and desired by the user. Different situations and personal specifics can be better distinguished by using the at least two outcomes of the at least 10 two gait irregularity algorithms. Unnecessary cue production can be reduced which prevents habituation and increases the effectiveness of the cueing device.

Description

P35401NL01/KHO

Title: Multiple algorithms for controlling a cueing device.

The invention relates to a cueing device, in particular a Parkinson cueing device comprising a control unit which is programmed by an algorithm providing an outcome to determine a gait irregularity during a movement of the user.

Gait irregularities, such as freeze of gait (FoG) or festination, can occur in people who suffer from neurological diseases such as Parkinson's disease, cerebral palsy, and multiple sclerosis.

Particularly for people who suffer from Parkinson’s disease, gait irregularity typically occurs when walking and is reflected as a brief unexpected inability to walk which might cause a fall.

US10251611 discloses a wearable apparatus which automatically activates a cue. The apparatus has a housing that contains electronics including at least one processor and a motion sensor. The electronics include a power circuit utilising a battery for power storage and peripherals, such as a power switch, an indicator light, and the like. When turned on, the motion sensor of the apparatus starts to detect movements and outputs motion data to the processor.

The processor uses a gait analysis algorithm to differentiate between normal walking and a gait irregularity such as a freeze of gait (FoG). If a gait irregularity is determined then a laser is activated. If no gait irregularity is determined, the laser is de-activated and the processor continues to receive motion data from the motion sensor.

A gait irregularity is determined by the algorithm performing a fast Fourier transformation on the motion data and by dividing it into predetermined frequency bands. A low frequency band relates to walking and a high frequency band relates to a gait irregularity. A mid-frequency band separates the high and low frequency bands to avoid false readings. The automatically activation of the cue is beneficial to prevent the user being concerned with having to activate the cue at a time when they are likely anxious, and ensures that a cue is provided for gait irregularities which the user may not have realised a gait irregularity was occurring or thought that the start of a gait irregularity wasn't of importance until too late. In the algorithm, a predetermined sensitivity number is used as a threshold to improve an accuracy in a determination of a gait irregularity. The predetermined sensitivity number is selectable and adjustable to provide a suitable sensitivity for a user wearing the apparatus.

A problem to this known apparatus is that a cue is produced in too many situations.

Unnecessary cueing adversely affects an effectiveness of the cueing device in overcoming gait irregularities. Too much cueing may lead to habituation. To prevent habituation, cue production should be minimised to situations susceptible to gait irregularities in which a cue is really necessary and desired by the user.

Regarding the above-mentioned prior art, it is remarked that any discussion of documents, acts, materials, devices, articles or the like included in the present specification is for the purpose of providing a context for the present invention, and is not to be taken as an admission that any such matters form part of the prior art or were before the priority date of each claim of this application common general knowledge in the field relevant to the present invention.

The general object of the present invention is to at least partially eliminate the above mentioned drawback and/or to provide a usable alternative. More specific, it is an object of the invention to improve an effectiveness of a cueing device by better distinguishing for situations requiring a cue, such that a cue is produced in situations which are susceptible to a gait irregularity and superfluous cueing is omitted in a situation not requiring a cue.

In a first aspect, according to the invention, this object is achieved by a cueing device according to claim 1. The cueing device is arranged for producing a cue for assisting a user during a motion. The cueing device comprises a control unit. The control unit is programmed by a first algorithm providing a first outcome to determine a gait irregularity, in particular a freeze of gait.

Here, a gait irregularity determination means a recognition of an occurring gait irregularity or a prediction of an upcoming gait irregularity. The control unit is operatively connected to at least one motion sensor for receiving a sensor signal. Based on the received sensor signal, motion data is obtained by the control unit. By running the first algorithm, a gait irregularity can be determined and is indicated by a positive outcome. If a gait irregularity is determined, the control unit may communicate an output signal to activate a cueing unit. The cueing unit is operatively connected to the control unit to provide a cue to the user to assist during the user's movement.

The cueing device according to the invention is improved in that the control unit is programmed by at least one additional algorithm providing an additional outcome to determine a gait irregularity in addition to the first outcome of the first algorithm. At least two outcomes are rendered by the control unit to indicate for a gait irregularity. The additional algorithm is different from the first algorithm. The additional algorithm has a different configuration. The additional algorithm deals with the motion data in a different manner and has different program rules. The additional algorithm may be based on a different formula for processing the motion data and/or include different parameters. The additional algorithm may include additional parameters to determine a gait irregularity.

Preferably, the additional algorithm is running simultaneously with the first algorithm. At least one additional algorithm may be running in parallel with the first algorithm.

A cue is only provided in case an outcome of both the first algorithm and a second algorithm positively indicate by rendering positive outcomes in determining a gait irregularity. It is required that also the additional algorithm provides a positive indication of a gait irregularity. By running at least two gait irregularity algorithms, unnecessary cue production can be reduced which may prevent or defer habituation and increase the effectiveness of the cueing device. The additional algorithm is carried out to control the outcome of the first algorithm. By the additional algorithm, a gait irregularity is determined more accurately and reliable to generate a cue when it is desired. The more accurate determination of a gait irregularity is beneficial in allowing a user to receive a cue in more detailed distinguished situations. The motion data is analysed by both the first and second algorithm, which results in a plurality of markers indicative of different types of motion. Hence, the different types of motion are distinguished more precisely, such that a determination to provide a cue to the user can be based on more detailed supporting data.

In an embodiment of the cueing device according to the invention, the control unit is programmed with at least three different algorithms for each determining a gait irregularity.

Each algorithm provides an own outcome in determining a gait irregularity. The control unit may be programmed to include a first, second and third algorithm. The control unit may be programmed to provide an output signal to the cueing unit to provide a cue to the user only in case that all algorithms positively indicate for a gait irregularity. A required amount of algorithms which have to positively indicate for a gait irregularity before providing a cue may be adjustable.

In an embodiment of the cueing device according to the invention, the control unit is programmed with at least three different algorithms for each determining a gait irregularity. At least one algorithm is selectable to be taken into account or cancelled, such that a variety of combinations of algorithms are available. A select value can be programmed in dependence of an occurring circumstance or to be disabled by a user or a physician to indicate which algorithms are to be taken into account in determining a gait irregularity. Herewith, the cueing device can be adapted to a personal situation. A personalised cueing device is provided. In a particular circumstance -e.g. a noisy or dark situation- it may be indicated that the first and third algorithm should run to provide a cue, while in another circumstance -e.g. a silent or visual bright situation- , it may be indicated that the first and second algorithm should run to provide acue. A particular algorithm may run for a selected time period, e.g. an afternoon, while another algorithm may run for another time period, e.g. an evening. Herewith, particular combinations of algorithms are available to be carried out in particular circumstances. Beneficially, dedicated combinations of algorithms may contribute in more accurate determined gait irregularities typically occurring in particular circumstances.

In an embodiment of the cueing device according to the invention, at least one of the algorithms is programmed to include at least one environmental parameter, e.g. shoe type, walking aid, hard or soft floor-surface, indoor, outdoor, a location, temperature and/or humidity. The environmental parameter may typically change on a daily basis. The cueing device may include a sensor for measuring the environmental parameter. All algorithms may include an environmental parameter, but preferably, at least one of the algorithms is distinguished from another algorithm by including an environmental parameter while the other algorithms lacks such environmental parameter.

In an embodiment of the cueing device according to the invention, the control unit is programmed with at least three different algorithms for each determining a gait irregularity. At least one algorithm is selectable to be taken into account or cancelled, such that a variety of combinations of algorithms are available. A select value can be programmed in dependence of personal parameters, e.g. a person's length, BMI, stride length, gait cycle, or arranged to be disabled by a user or a physician to indicate which algorithms are to be taken into account in determining a gait irregularity for a particular person. Such a selection of personal parameter may typically be constant over an amount of time, in particular for at least some weeks. For a specific person -e.g. a tall person - it may be indicated that the first and third algorithm should run to provide a cue, while for another person -e.g. a small person - , it may be indicated that the first and second algorithm should run to provide a cue. Herewith, the cueing device provides particular combinations of algorithms which are available to be carried out to better distinguish for different users. Beneficially, dedicated combinations of algorithms may contribute in more accurate determined gait irregularities typically relating to personal parameters.

In an embodiment of the cueing device according to the invention, at least one of the algorithms is programmed to include at least one user dependent parameter. At least one of the algorithms includes a user profile. The user profile contains at least one user dependent parameter. A user dependent parameter may include at least one of the parameters: a body length, a body weight, a BMI, and a stride length. Preferably, at least one of the user dependent parameters is adjustable by the user.

In an embodiment of the cueing device according to the invention, the cueing device has a user interface which is operatively connected to the control unit to allow an adjustment of an amount of algorithms being necessary to positively indicate for a gait irregularity to provide a cue to the user. The user interface may be accessible by the user or by a physician. Preferably, the user interface is accessible by the physician only. The user interface may be formed by a switch button or include a display to allow the user to switch between operative modes corresponding with a certain amount of algorithms being involved in determining a gait irregularity. The user interface of the cueing device may be formed by a remote connected device, like a smart phone or smart watch, to allow a remote adjustment of a set minimum amount of algorithms positively indicating a gait irregularity required to provide a cue.

In an embodiment of the cueing device according to the invention, the control unit is further programmed by a routine algorithm to determine a type of motion of the user based on a received sensor signal. The routine algorithm is distinguished from other algorithms by a classification step in which motion data is classified as relating to a routine or a non-routine motion. Routine motion considers a motion which is carried out by the user without or with less attention of the user, e.g. distance walking. In a routine motion, a user has no particular focus in carrying out the motion. A non-routine motion considers a motion of the user which is more incidental and which requires more attention of the user, e.g. stair climbing. The routine algorithm is preferably carried out before running any of the available algorithms to determine a gait irregularity. Preferably, the control unit is programmed that if a non-routine motion is determined, a running of any of the available gait irregularity algorithms is cancelled. A cancellation of the gait irregularity algorithms may be beneficial in saving battery of the device by not processing motion data in a situation which is less vulnerable to a gait irregularity.

In an embodiment of the cueing device according to the invention, the cueing device is a wearable cueing device which is formed by a system comprising a combination of a smartwatch and a smartphone, wherein the smartphone functions as a housing provided with at least one motion sensor for providing a sensor signal, and wherein the smartwatch functions as a cueing unit for producing a cue to the user in case of a gait irregularity. The smart phone and the smartwatch are each programmed with an app which is configured to establish a cooperation of the combination of the smart phone and smartwatch to determine and to produce a cue in case of a gait irregularity. In use, the smartwatch is worn on a wrist and the smartphone is positioned at one of the legs of the user. The smart phone may be simply inserted in a hose pocket, or may be attached to the leg of the user by a limb band. By forming the wearable cueing device by commonly available devices, a wearable cueing device is easily available for a person with Parkinson. Beneficially, only a smartwatch app and a smartphone app needs to be downloaded and installed by the user to obtain the wearable cueing device.

In an embodiment of the cueing device according to the invention, the cueing device is a wearable cueing device comprising a strap and a housing for housing the motion sensor, wherein the housing is connectable to the strap.

In an embodiment of the cueing device according to the invention, the cueing device further comprises an alarming unit to generate an alarm to attention the user and/or a remote alarm to attention a user -related person, like a relative or a physician. The cueing device may be a wearable fall alarming device including a cueing unit and an alarming unit which is configured to produce an alarm in case of a determination of a risk of falling instead of producing a cue as described above regarding the cueing device. The wearable fall alarming device may comprise multiple algorithms to provide multiple outcomes in determining a risk of falling by determining a gait irregularity. The control unit is programmed by at least a first and second algorithm providing at least a first and second outcome to determine a gait irregularity, wherein an alarm is only produced in case that the at least first and second outcomes of the at least first and second algorithms positively indicate for a gait irregularity.

Further, the invention relates to a computer-implemented method for controlling a cueing device configured to provide a cue during a motion of the user. According to the invention, the method comprises a step of receiving a sensor signal from at least one motion sensor. A step of acquiring motion data based on the received sensor signal. A step of running a first algorithm on the motion data to determine a gait irregularity. A step of running at least a second algorithm on the motion data to determine a gait irregularity. And a step of providing a cue only if at least two algorithms each have an outcome positively indicating for a gait irregularity.

Further, the invention relates to a computer program product comprising a computer readable medium. The computer readable medium as computer readable code embodied therein. The computer readable code is configured such that, on execution by a suitable computer or processor, the computer processor is caused to perform the method according to the invention.

Further, the invention relates to a method, in particular a computer-implemented method for controlling a cueing device formed by a combination of a smartwatch and a smartphone which combination is configured to provide a cue during a motion of a user, the method comprising: - receiving a sensor signal from at least one motion sensor of the smartphone; - acquiring motion data based on the received sensor signal by a control unit being formed by a processor of one of the smartwatch and smartphone; - running a gait irregularity algorithm on the motion data to determine a gait irregularity;

- in particular, running at least a second gait irregularity algorithm on the motion data to determine a gait irregularity; - in particular, providing a cue only if at least two algorithms have an outcome positively indicating for a gait irregularity.

In an embodiment of the cueing device according to the invention comprising a smartphone and a smartwatch, the method further comprises the step of producing a cue by one of the smartwatch or smart phone. Preferably, a cue is produced by the smartwatch, wherein the smartphone is used to obtain motion data.

Thus, a cueing device is provided for producing a cue in assisting a user during a motion. The cueing device comprises a control unit operatively connected with a motion sensor and a cueing unit. The control unit is programmed with multiple algorithms providing multiple outcomes in determining a gait irregularity during the motion. A cue is only produced in case outcomes of at least two algorithms positively indicate for a gait irregularity. Cue production should be minimised to situations in which a cue is really necessary and desired by the user. Different situations and personal specifics can be better distinguished by using the at least two outcomes of the at least two gait irregularity algorithms. Unnecessary cue production can be reduced which prevents habituation and increases the effectiveness of the cueing device.

In a second aspect according to the invention, instead of a cueing device, a wearable fall alarming device is provided which is configured to produce an alarm in case of a determination of a risk of falling instead of producing a cue as described above regarding the cueing device.

The wearable fall alarming device has an alarming unit to generate an alarm to attention the user and/or a remote alarm to attention user related persons, like a relative or a physician. The wearable fall alarming device according to the second aspect of the invention comprises multiple algorithms to provide multiple outcomes in determining a risk of falling. Embodiments of the wearable fall alarming device are described in the claims in which a cueing unit is to be considered as substituted by an alarm unit.

The invention will be explained in more detail with reference to the appended drawings. The drawings show a practical embodiment according to the invention, which may not be interpreted as limiting the scope of the invention. Specific features may also be considered apart from the shown embodiment and may be taken into account in a broader context as a delimiting feature, not only for the shown embodiment but as a common feature for all embodiments falling within the scope of the appended claims, in which:

Fig. 1 shows a flow diagram presenting operational steps of the cueing device according to the invention.

Fig. 1 shows a flow diagram of an embodiment of the computer-implemented method according to the invention.

In a first step of the method, sensor data is obtained. A cueing device provided with the method comprises at least one motion sensor to obtain motion data based on a received sensor signal.

Based on obtained sensor data, it can be established whether or not the user is active. When the user is inactive, no gait irregularity will occur, and the cueing device may remain in a standby status. Based on the sensor data, it can be determined when the user becomes active.

When the user becomes active, in a second step of the method, a routine algorithm is run to determine whether or not an occurring motion of the user is a routine motion. In case that it is determined by the routine algorithm that the occurring motion of the user is a non-routine motion, a risk on an occurrence of a gait irregularity may be neglectable and the cueing device may remain in the standby status. However, in case that it is determined by the routine algorithm that the occurring motion of the user is a routine motion, a gait irregularity might occur and the cueing device is switched to an operational mode in which multiple algorithms are processed to determine a gait irregularity.

Here, the control unit is programmed with a first algorithm, a second algorithm and a third algorithm indicated by Alg1, Alg2 and Alg3. To produce a cue, it is required that at least two outcomes of these algorithms determine a gait irregularity. The control unit is programmed with a weight factor ( >1) to indicate an amount of algorithms required to positively determine a gait irregularity before activating a cueing unit. If an amount of positive outcomes is equal or larger than the weight factor, the cue is activated. The weight factor may be adjustable by a user or physician. Alternatively, the weight factor may be dependent on an occurring circumstance or a personal parameter.

Numerous variants are possible in addition to the embodiment shown in the figure. For example, in a variant of the illustrated embodiment of the system, an alternative embodiment of the system may lack a routine algorithm to distinct between a non-routine and routine motion.

Although the present invention has been described in detail, it will be apparent to those skilled in the art that various changes and modifications can be made without departing from the scope of the invention as hereinafter claimed. It is intended that all such changes and modifications be encompassed within the scope of the present disclosure and claims.

Further, it is remarked that any feature of the device according to the invention which is described in the embodiments and/or mentioned in the dependent claims is in itself considered patentable without any dependency to another presented feature. In particular, any measure presented in a dependent claim is also considered patentable without dependency of the independent claim.

Thus, the invention provides a cueing device for producing a cue in assisting a user during a motion. The cueing device comprises a control unit operatively connected with a motion sensor and a cueing unit. The control unit is programmed with at least a first and second algorithm providing each an outcome to determine a gait irregularity during the motion. A cue is only produced in case outcomes of at least two algorithms positively indicate for a gait irregularity.

Herewith cue production can be minimised to situations in which a cue is really necessary and desired by the user. By running at least two gait irregularity algorithms and activating a cue based on at least two positive outcomes, unnecessary cue production can be reduced which may prevent or defer habituation and thus increase an effectiveness of the cueing device.

Claims

CONCLUSIONS

1. Stimulus device (1) for assisting a user with a generation of a stimulus during a movement when an irregularity in a gait is determined, the stimulus device comprising: - a control unit programmed with a first algorithm for providing a first outcome to determine an irregularity in a gait, in particular a freeze of gait (FoG), during the movement; - at least one motion sensor operatively connected to the control unit to obtain motion data based on a received sensor signal; and - a stimulus unit operatively connected to the control unit to provide a stimulus to the user, characterized in that the control unit is programmed with at least a second algorithm for providing a second outcome in addition to the first algorithm to detect an irregularity in a loop, wherein a stimulus is only generated if the at least first and second results of the at least first and second algorithms give a positive indication of an irregularity in a loop.

The stimulation device of claim 1, wherein the control unit is programmed with at least three different algorithms to provide at least three outcomes to determine an irregularity in a loop.

Stimulus device according to claim 1 or 2, wherein a stimulus is only generated in the event that all outcomes of the algorithms give a positive indication of an irregularity in a loop.

Stimulus device according to claim 1 or 2, wherein the required number of algorithms that must have a positive indication of an irregularity in a loop before a stimulus is delivered is adjustable.

The stimulation device of claim 4, wherein at least one algorithm is programmed or manually selectable to be taken into account or ignored, such that a variety of combinations of algorithms are available.

Stimulus device according to claim 4 or 5, wherein the control unit has a user interface that allows a person to disable at least one of the available algorithms.

7. Stimulus device according to any of the preceding claims, wherein at least one algorithm is based on a usage profile comprising at least one user-dependent parameter.

Stimulus device according to claim 7, wherein at least one of the algorithms is distinguished from another algorithm by including a user-dependent parameter, for example a BMI or body height, wherein the other algorithm then has no user-dependent parameter.

9. Stimulus device according to any of the preceding claims, wherein at least one algorithm comprises an environmental parameter, for example a hard or soft surface, a shoe type, a brightness, a temperature and/or humidity.

10. Stimulus device according to any one of the preceding claims, wherein the control unit is further programmed with a routine algorithm that is designed to distinguish between a routine and a non-routine movement of the user.

The stimulation device of claim 10, wherein the routine algorithm is executed before any of the algorithms for determining an irregularity in a loop are executed.

Stimulus device according to claim 11, wherein the control unit is programmed such that if a non-routine movement is determined by the routine algorithm, an execution of one of the available algorithms for determining an irregularity in a loop is canceled.

13. Stimulus device according to any of the preceding claims, wherein the stimulus device is formed by a system comprising a smartphone and a smartwatch that are provided with cooperating apps and wherein the smartphone comprises the at least one motion sensor for obtaining motion data from the user , and wherein the smartwatch acts as a stimulus unit for generating a stimulus to the user in the event of an irregularity in a loop.

Stimulus device according to any one of the preceding claims, wherein the stimulus device is a portable stimulus device, wherein the stimulus device comprises a strap and a housing for housing the motion sensor, the housing being connected to the strap.

15. A computer-implemented method for controlling a stimulus device that is adapted to provide a stimulus during movement of a user, wherein the method comprises: - receiving a sensor signal from at least one motion sensor; - obtaining movement data based on the received sensor signal; - executing a first algorithm on the movement data to provide a first outcome to determine an irregularity in a loop; - running a second algorithm on the movement data to provide a second outcome to determine an irregularity in the gait; - only providing a stimulus if at least two algorithms both have a positive indication for determining an irregularity in the loop.

A computer program product comprising a computer-readable medium, the computer-readable medium incorporating a computer-readable code, the computer-readable code being configured such that when executed by a suitable computer or processor, the method of claim 15 is executed by the computer or processor is running.