Method and device for reducing the health burden caused by a user's sitting and movement behavior

Abstract

Methods for reducing the health burden on a user (14), characterized by a) the measurement of three-dimensional coordinates (17a, 17b, 17c) of selected key points (18a, 18b, 18c) of the musculoskeletal system (20) of the user (14) in a body posture (12) of the user (14) by a sensor (16) in the form of a depth sensor (40) and / or a camera in relation to a predetermined reference point f) Assigning the three-dimensional coordinates (17a, 17b, 17c) of the key points (18a, 18b, 18c) to an index value of a stress index by an analysis unit (34) and determining the time period in which the stress index assumes the index value, wherein the assignment of the three-dimensional coordinates (17a, 17b, 17c) of the key points (18a, 18b, 18c) to an index value is carried out using a neural network (44), (g) the output of feedback to the user (14) to reduce the health burden by an output unit (38), wherein the output of the feedback takes place after a shorter or a longer period of time in which the user assumes a posture (12), wherein the output of the feedback takes place after the shorter period of time if the deviation of the posture (12) of the user (14) from the reference posture (36) is greater than a predetermined first reference deviation.

Description

BACKGROUND OF THE INVENTIONThe invention relates to a method for reducing the health burden on the user caused by the sitting and moving behavior of a user.The health burden arises in particular from the posture and / or the movement behavior of the user, particularly if the user assumes the same posture over a long time, and / or generally from muscle tensions. Methods for correcting posture are known from the prior art, in which the posture of a user is determined with the aid of data from a depth sensor. The posture is then assigned to one of several classes of healthy or unhealthy postures. The postures, in particular seating postures, of the user are determined over a period of several days. Then, the user receives a report with proposals for improving his seating posture.US 2017 / 020438 A1 discloses pressure sensors on a seating furniture item in order to determine the posture of a user on the seating furniture item from the pressure information item.WO 2017 / 137852 A2 discloses a device for determining a head posture which is worn on the upper body and neck of the user. A sensor on the device determines head posture by a distance measurement to the neck of the user, and the measured data is analyzed by a processor.The reference US 2016 / 0183687 A1 relates to sensors which are arranged on furniture items, in particular on seating furniture, in order to determine the posture of a user. With the aid of the sensors, in particular forces acting on different locations of the seating furniture, and movements and positions of the seating furniture, are determined.US 2019 / 0239813 A1 discloses a sensor equipped cloth that can be attached to a chair to determine a posture of a user sitting on the chair. The sensors are in particular designed as load sensors. To determine the posture of the user, the measured loads are divided into different levels.US 2015 / 0366350 A1 relates to a method for changing a posture of a user, comprising the steps of: generating a posture model which represents the posture of the user and comparing the posture with a reference posture in order to determine changes suggestions for the posture. Generating the posture model comprises evaluating sensor data relating to positions of body parts of the user.US 2018 / 0120420 A1 relates to measuring movements of a user with the aid of radio frequency sensors.US 2019 / 0159719 A1 relates to evaluating signals of an inertial sensor which is carried on the body.US 2019 / 0200919 A1 relates to a method for determining posture, in which image data are processed, wherein a depth is assigned to each pixel of a recorded image by measurements of a depth sensor.US 2009 / 0324024 A1 discloses a system and a method for correcting a posture of a user of the system, wherein a camera regularly records an image of the user. In each image, the face of the user is determined and compared with a predetermined reference image of the face in a posture of the user classified as good. If the user takes a posture classified as good several times, a corresponding feedback is given to him. In particular, the user is also given a feedback if he constantly assumes a posture classified as poor several times or over a predefined period of time.In the known methods, it is not taken into account that healthy postures can also have a stressful effect over a longer time. To reduce the health burden, only a correction of the posture is proposed in order to assume a posture that is considered healthy.Object of the InventionIt is an object of the present invention to specify a method for reducing the health burden, which also takes account of burdens caused by postures classified as healthy and more effectively reduces the burden on the musculoskeletal system.DESCRIPTION OF THE INVENTIONThis object is achieved according to the invention by a method according to claim 1. Advantageous embodiments are evident from the dependent claims which refer back.The method according to the invention comprises the following steps: a) measuring three-dimensional coordinates of selected key points of the musculoskeletal system of the user in a posture of the user by a sensor with respect to a predetermined reference point; f) assigning the three-dimensional coordinates of the key points to an index value of a stress index by an analysis unit and determining the time duration during which the stress index assumes the index value; g) outputting a feedback to the user for reducing the health stress by an output unit.Postures which are considered healthy can also have a burdensome effect on the body, in particular the musculoskeletal system, if they are taken for too long a time. Measuring the time duration in which the stress index assumes the index value has the effect that a possible overload of body parts can be determined and the user can be given a corresponding feedback for reducing the health stress, which relates in particular to the stress on the body or on the musculoskeletal system, in particular in the form of a correction of his posture, the insertion of a pause for a predetermined time duration or the performance of a predetermined physical activity, in particular a exercise of the body, over a predetermined time period. The output of the feedback or of instructions takes place in particular after a predefined time period in which the user assumes a posture, wherein the time period is defined depending on the posture. By measuring the time duration during which the user assumes a specific posture, the movement behavior of the user can also be determined.The selected key points of the body are located, for example, on the neck, on the back, on the arms, on the shoulders, on the head, in particular at positions on the head, by means of which the orientation of the head can be determined, for example on the forehead or the ears. Among other things, the measurements are taken at key points at selected joints.The load index is in particular an index for specifying the load. Preferably, ranges of load values are each assigned to a value of the load index. In order to determine the load during posture, the muscle tension is determined, in particular measured, for example.Alternatively or additionally, it is determined whether a user performs the proposed change in posture. The user then receives a feedback in the form of a confirmation when the change in posture is successfully carried out, a feedback in the form of a proposal for a correction when the change in posture is incorrectly carried out or a further feedback in the form of a later reminder when the previous feedback is not taken into account.Alternatively or additionally, data on the postures of the user are stored for a user over a predetermined period of time in this period of time and a behavior profile of the user is created therefrom, which the user can see. This allows the identification of user's ownities that are presented to him as feedback, in particular in conjunction with health-related information.Preferred Embodiments and DevelopmentsOne embodiment of the method is characterized by the following steps: b) the determination of a reference posture, c) the determination of selected parameter values for the determination of the posture of the user on the basis of the measurement data of the sensor by the analysis unit, e) the assignment of the difference between the parameter values of the posture and the respectively associated parameter values of the reference posture to an index value of the load index by the analysis unit.The reference posture is in particular a predetermined posture, in which in particular the load of the musculoskeletal system is the smallest under a selection of predetermined postures, preferably an upright posture classified as ergonomic. The measurement data of the sensor are in particular the aforementioned measured three-dimensional coordinates of the selected key points of the user's body. The three-dimensional coordinates are determined in particular by a coordinate measuring machine. The coordinate measuring machine has in particular the sensor. The three-dimensional coordinates of the key points, in particular Cartesian coordinates, are preferably determined relative to a predetermined reference point, in particular the position of the sensor, inter alia by the determination of angles and / or distances of the key points with respect to the sensor. The measured three-dimensional coordinates are assigned to the selected parameter values of the posture of the user. The assignment of the difference between the parameter values of the posture and the respectively associated parameter values of the reference posture to an index value of the load index then brings about an assignment of the three-dimensional coordinates measured by the sensor to the index value of the load index.The postures can be defined as parameter values inter alia by the three-dimensional coordinates of the key points, the relative distances and / or angles between the key points. These parameter values make it possible to distinguish the different postures from one another.The sensor, the analysis unit and the output unit can be present in an independent device or can also be integrated in other working devices, for example in a computer, a desk or a mobile radio device.By determining the posture on the basis of the selected parameters and the associated time duration, actions of the user such as sitting, telephone making and / or carrying out exercise of movement can be determined. These actions and their duration may be taken into account in the feedback for correcting posture, for example, by issuing recommendations for selection of the actions and how long the selected actions are to be performed to optimally reduce health burden.An advantageous embodiment of the method is characterized in that the method has the following method step:d) determining the health burden of the user from the deviation of the posture of the user from the reference posture;wherein the health burden or burden on the musculoskeletal system of the user determined in method step d) is included in the assignment to an index value according to method step e).In this case, a deviation is understood to mean, in particular, one or more differences in the parameter values of the posture and of the reference posture. In this case, in particular, the deviation is divided into ranges by dividing the associated parameter values and a load is assigned to each range of the deviations. This can be done on the basis of models and / or by measurements. From the deviation of the posture from the reference posture, in particular the stress index, in particular the index values that the stress index can assume, is determined.In a further embodiment of the method, the creation of a virtual body model is used to determine the load and relief of the musculoskeletal system of the user in selected postures. The virtual body model has in particular the key points and preferably the selected joints with the associated degrees of freedom, so that different postures of users can be simulated by the virtual body model. In particular, the virtual body model is used to define the load index, in particular the value range of the load index, for example by determining deviations of the various postures from the reference posture.A preferred development is characterized in that forces, torques, pulses and / or rotary pulses are determined at selected reference points of the virtual body at selected postures. In particular, the coordinates of their center of gravity, a selected volume and / or a selected weight can be assigned to individual body parts. This allows the force, the torque, the impulse and / or the angular impulse with which body parts act on other body parts in different postures to be estimated. For example, the head of the virtual body may be represented as a sphere having a selected weight acting with its weight force on the spine, which is modeled as a cylinder, for example. Alternatively or additionally, the coordinates of the center of gravity of the head and the weight of the head, as well as the coordinates of the center of gravity of the spine and the weight of the spine, may be established to determine, for example, the torque acting on the spine.From the forces, torques, pulses and / or rotational pulses determined, the load of the musculoskeletal system is preferably determined by an algorithm.The determination of the load is preferably carried out by assigning tensions and / or relaxations of muscles of test persons measured by muscle tension sensors for optimizing the algorithm. In a test phase, the muscle voltage sensors are arranged in particular at the selected key points of the test persons at selected postures. As a result, the tension or relaxation of the muscles in a specific posture can be determined directly. These measurements of the tension of the muscles allow a more accurate determination of the necessary correction of the posture in order to counteract unhealthy muscle tension or even unhealthy muscle relaxation. The measured muscle tensions or tensions of the muscles allow a correction of the muscle tensions determined with the virtual body model. In particular, the muscle stress sensors are used to define the stress index, in particular the value range of the stress index, by measuring the muscle stresses.According to the invention, the three-dimensional coordinates of the key points are assigned to an index value in method step f) using a neural network, in particular a trained neural network. The neural network is trained in particular by the neural network associating postures of test persons with predetermined categories of postures. The neural network is then signalled whether the assignment of the posture by the neural network to the predetermined category of posture is correct. This improves the decision process of the neural network.The categories of postures are defined in particular on the basis of ranges of the parameter values. Examples of such categories are "forward-bent", "upright", "rearward-tilted".The test subjects differ in particular with regard to height, age, weight and / or sex.Alternatively or additionally, the training of the neural network takes place by assigning an index value of the load index to the measured distances of key points of bodies of test persons to the sensor and / or to the three-dimensional coordinates of key points of bodies of test persons by the neural network. The neural network is then signalled whether its assignment has been carried out correctly. This trains the neural network to correctly determine the index value of the load index.According to the invention, the output of the feedback takes place after a shorter or a longer period of time in which the user assumes a posture, wherein the output of the feedback takes place after the shorter period of time if the deviation of the posture of the user from the reference posture is greater than a predefined first reference deviation. As a result, a high load on the body due to posture is corrected comparatively quickly. A low load does not require such a quick correction, so that the user can remain in the corresponding posture for a longer time. Within the scope of the method, the feedback can also take place after more time periods than the two time periods in the form of the shorter and the longer time periods, wherein decreasing time periods correspond to increasing deviations of the posture from the reference posture. The deviations of the posture from the reference posture are determined by the deviations of the associated parameter values from the parameter values in the reference posture.For example, the output of the feedback takes place after 4 minutes to 5 minutes if, with respect to a key point on the genick, the deviation of the posture of the user from the reference posture is greater than a predefined first reference deviation. The output of the feedback takes place after 9 minutes to 10 minutes if, with respect to the key point on the genick, the deviation of the posture of the user from the reference posture is less than the predefined first reference deviation. The deviation and the first reference deviation are defined, for example, by an angle, one leg of which runs from a key point on the spinal column through the key point on the genick. The other leg extends, for example, starting from the key point on the spinal column through a key point on the ears, on the nose or on the forehead.Correspondingly, the output of the feedback takes place, for example, after 9 minutes to 10 minutes if, with respect to a key point at the shoulders, the deviation of the posture of the user from the reference posture is greater than a predefined first reference deviation. The output of the feedback takes place after 19 minutes to 20 minutes if, with respect to the key point at the shoulders, the deviation of the posture of the user from the reference posture is less than the predefined first reference deviation.A further advantageous embodiment is characterized in that a greater or a smaller change in the posture is output in the feedback, wherein the greater change in the posture is output if the deviation of the posture of the user from the reference posture is greater than a predefined second reference deviation. The smaller change in the posture or the seating posture is output when the deviation of the posture of the user from the reference posture is smaller than the predetermined second reference deviation. The greater change compensates for a high load on the body to a greater extent due to the posture. The changed posture may itself deviate from the reference posture again for the purpose of compensation. Within the scope of the method, the feedback can also output more changes than the two changes in the form of the smaller and the larger change, wherein increasing changes correspond to increasing deviations of the posture from the reference posture. The changes in posture when transitioning from an original to a new posture are determined by the difference of the parameter values at the key points in the original and new postures. The second reference deviation may be identical to the first reference deviation.Alternatively or additionally, a longer or shorter predetermined time duration for a pause is output in the feedback, wherein the longer time duration is output if the deviation of the posture of the user from the reference posture is greater than the predefined second reference deviation. Alternatively or additionally, a first or second predetermined exercise is output in the feedback, wherein the first exercise is output if the deviation of the posture of the user from the reference posture is greater than the predetermined second reference deviation.The method is characterized according to the invention by using a depth sensor and / or a camera for determining the three-dimensional coordinates of the key points and / or the distances of the user's musculoskeletal system from the depth sensor and / or the camera. The depth sensor or the camera enable a contactless determination of the posture of the user. In the case of a depth sensor, a better user acceptance has been shown.An apparatus for carrying out the method according to one of the preceding embodiments comprises: a) a sensor for measuring three-dimensional coordinates of selected key points of the musculoskeletal system of the user in a posture of the user with respect to a predetermined reference point, b) an analysis unit for assigning the three-dimensional coordinates of the key points to an index value of a load index and determining the time duration during which the load index assumes the index value, and c) an output unit for outputting a feedback to the user in order to reduce the health load.By such a device, the time duration in which the stress index becomes the index value can be measured. This allows a possible overload of the body to be determined and a corresponding feedback to be given to the user.One embodiment of the device is characterized in that the analysis unit is configured to determine the deviation of the posture of the user from the reference posture and the duration of the posture of the user. The analysis unit has a means for determining the deviation of the posture of the user from the reference posture and the time duration during which the user takes up the posture.In addition, the device can have a temperature sensor, a light sensor, a noise sensor, a skin heat sensor and / or a moisture sensor.Further advantages of the invention will become apparent from the description and the drawing. Likewise, the features mentioned above and those set out further below can each be used individually on their own or together in any desired combinations. The embodiments shown and described are not to be understood as a final enumeration, but rather have exemplary character for describing the invention.DETAILED DESCRIPTION OF THE INVENTION AND DRAWINGFIG. 1 shows a schematic view of a device for carrying out the method according to the invention for reducing the health burden of a user, wherein the user is shown in a posture deviating from a reference posture; FIG. 2 shows a schematic view of the device for carrying out the method according to the invention with the user in the reference posture; and FIG. 3 shows a schematic illustration of the method according to the invention.The device 10 shown in FIG. 1 for reducing the health burden of a user 14, in particular for improving the posture 12 of the user 14, has a sensor 16 for measuring distances of key points 18 a, 18 b, 18 cof the musculoskeletal system 20 from the sensor 16 and / or coordinates 17 a, 17 b, 17 cof key points 18 a, 18 b, 18 c. The key points 18 a, 18 b, 18 cof the body or musculoskeletal system 20 are located at the hip 22, in the lower back section, in particular at the lower half 24 of the spinal column, and at the head 26, in particular at the ears, the forehead or at the nose. The three-dimensional coordinates 17 a, 17 b, 17 cserv to determine a parameter value 28 in the form of an angle 30, the legs 32 a, 32 bthereof extending, starting from the key point 18 aat the hip 22, on the one hand through the key point 18 bin the lower back section 24 and on the other hand through the key point 18 con the head 26. The posture 12 is determined by this parameter value 28, for example. The parameter value 28 is determined by an analysis unit 34 of the apparatus 10. In addition, the analysis unit 34 determines the deviation of the posture 12 of the user 14 from a reference posture 36 (see FIG. 2 ). The deviation is determined by way of example in the form of the difference between the angle 30 in the body posture 12 shown and in the reference posture 36 shown in FIG. 2. The deviation is in particular assigned to an index value of a stress index which measures the stress of the musculoskeletal system 20. The analysis unit 34 likewise determines the duration of the posture 12 of the user 14. An output unit 38 of the device 10 outputs a feedback to the user 14 in order to correct his posture 12. The sensor 16 is in particular designed as a depth sensor 40 or camera. The device 10 for improving the posture 12 of the user 14 is located in FIG. 1 by way of example on a table 42.FIG. 2 schematically shows the device 10 for reducing the health burden (see FIG. 1 ) of the user 14, wherein the user 14 is in the reference posture 36. A neural network 44 connected or integrated to the analysis unit 34 serves for the determination of the parameter values 46 of the reference posture 36, here for example in the form of the angle 30 at the reference posture 36. A muscle tension sensor 48 at the key point 18 b(cf. FIG. 2 ) on the lower half 24 of the spine of the user 14 measures the tension of the muscles of the user 14 at this key point 18 bin a test phase or training phase. The data thus determined are then later available in the evaluation of the body load.FIG. 3 schematically shows the method 100 for reducing the health burden of the user 14 (see FIG. 1 ). In a first step 102, selected parameter values 28 are measured for determining the posture 12 of the user 14. Then, in a second step 104, the difference between the parameter values 28 and the respectively associated parameter values 46 of the reference posture 36 is determined by an analysis unit 34. In a third step 106, the deviation of the posture 10 of the user 14 from the reference posture 36 is determined from the differences of the parameter values 28 of the posture 10 and the parameter values 46 of the reference posture 36 and the duration of the posture 10 of the user 14. The deviation is in particular assigned to an index value of a stress index which measures the stress of the musculoskeletal system 20. In step 106, alternatively or additionally, the body load is determined on the basis of a virtual (mechanical) body model and / or on the basis of muscle stresses previously measured on test persons. In a fourth step 108, a feedback is output to the user 14 for the correction of his posture 10 on the basis of the deviation of the posture 10 from the reference posture 36 and the duration of this deviation and / or on the basis of the determined body load by an output unit 38.By examining all the figures of the drawing together, the invention relates to a method 100 for reducing the body load of a user 14, for which purpose three-dimensional coordinates 17 a, 17 b, 17 cof selected key points 18 a, 18 b, 18 cof the musculoskeletal system 20 are determined by a sensor 16 with respect to a predefined reference point. By means of the three-dimensional coordinates 17 a, 17 b, 17 c, parameter values 28 are determined with an analysis unit 34, which parameter values define the posture 12 of the user 14. These parameter values 28 are compared with parameter values 46 at the corresponding key points 18 a, 18 b, 18 cin a reference posture 36 by the analysis unit 34. The reference posture 36 is a predetermined posture. The difference of the parameter values is in particular assigned to an index value of a load index which measures the load of the musculoskeletal system 20. In addition, the period of time during which the user 14 takes the posture 12 with the measured parameter values 28 is recorded by the analysis unit 34. A recommendation for the exertion of movement and / or for the reduction of the body load is derived from the time duration and the deviation from the reference posture 36, which the user 14 receives as feedback from an output unit 38, wherein the feedback is effected in particular depending on the strength of the deviation of the body posture 12 from the reference posture 36. Alternatively or additionally, the feedback can be effected on the basis of the determined load of the musculoskeletal system 20.

Claims

Method for reducing the health burden of a user (14), characterized by a) measuring three-dimensional coordinates (17a, 17b, 17c) of selected key points (18a, 18b, 18c) of the musculoskeletal system (20) of the user (14) in a posture (12) of the user (14) by a sensor (16) in the form of a depth sensor (40) and / or a camera with respect to a predetermined reference point f), assigning the three-dimensional coordinates (17a, 17b, 17c) of the key points (18a, 18b, 18c) to an index value of a burden index by an analysis unit (34) and determining the time duration in which the burden index assumes the index value, wherein the assignment of the three-dimensional coordinates (17a, 17b, 17 c) of the key points (18 a, 18 b, 18 c) to an index value is effected using a neural network (44), g) the output of a feedback to the user (14) for reducing the health burden is effected by an output unit (38), wherein the output of the feedback is effected after a shorter or a longer period of time in which the user assumes a posture (12), wherein the output of the feedback is effected after the shorter period of time if the deviation of the posture (12) of the user (14) from the reference posture (36) is greater than a predefined first reference deviation.Method according to claim 1, characterised by b) the determination of a reference posture (36); c) the determination of selected parameter values (28) for the determination of the posture (12) of the user (14) on the basis of the measurement data of the sensor (16) by the analysis unit (34); e) the assignment of the difference between the parameter values (28) of the posture (12) and the respectively associated parameter values (46) of the reference posture (36) to an index value of the load index (36) by the analysis unit (34).Method according to Claim 2, characterized in that the method has the following method step: d) the determination of the health burden of the user (14) from the deviation of the posture (12) of the user (14) from the reference posture (36); wherein the health burden of the user (14) determined in method step d) is included in the assignment to an index value according to method step e).Method according to one of the preceding claims, characterized in that the creation of a virtual body model is used to determine the load and relief of the musculoskeletal system (20) of the user (14) in the case of selected postures (12).Method according to Claim 4, characterized in that the forces, torques, pulses and / or rotary pulses are determined on selected reference points of the virtual body at selected postures (12).Method according to one of the preceding claims, characterized in that the determination of the load is carried out by the assignment of tenses of muscles of test persons measured by muscle voltage sensors (48).Method according to one of the preceding claims, characterized in that a greater or a smaller change in the posture (12) is output in the feedback, wherein the greater change in the posture (12) is output if the deviation of the posture (12) of the user (14) from the reference posture (36) is greater than a predefined second reference deviation.

Images