AT520485A1 - A method of automatically detecting a home position of a person lying on a mattress and generating a graphic representation of that person - Google Patents

Abstract

The present invention relates to a method for automatically detecting a basic position (41) of a patient (1) lying on a mattress (12) provided with pressure sensors (13). In this case, a print pattern (3) and severity points (59) is detected. A basic position of the patient is determined by means of predetermined rules based on the severity points. Furthermore, the present invention relates to a method for generating a graphic representation of a lying on a mattress (12) patient (1) based on the print pattern (3). The measured print pattern is compared to database print patterns from a print pattern database (42). A best-match database print pattern (5) is selected and the associated selected skeleton data (46) is displayed on a display device (6).

Description

11/10/2017 Austrian patent application

Huber, Dietrich

HUR1003PAT

Method and device for automatically detecting a basic position of a person lying on a mattress and for generating a graphic representation of this person

The present invention relates to a method and apparatus for automatically detecting a home position of a person lying on a mattress and for generating a graphic representation of that person.

From Rihar et al., Infant trunk posture and arm movement assessment using pressure mattress, inertial and magnetic measurement units (IMUs), [online] Journal of Neuroengineering and Rehabilitation 2014 11: 133, DOI: 10.1186 / 1743-0003-11-133 , 2014-09-06 downloaded on May 10, 2017 via http: //www.jneuroengrehab.biomedcentral.eom/article/10.1186/1743-003-11-133 shows a method in which data from a measuring mattress with built-in pressure sensors along with Magnetic sensors, which are placed on the body of an infant, is used for motion detection of an infant in three-dimensional space.

WO 2011/113070 A1 discloses a method for monitoring the medical condition of a patient, wherein u. a. the danger of formation of pressure ulcers should be recorded. On the patient's body sensors are attached, such. As acceleration sensors, gyros or magnetic sensors to detect the position and movement of the patient. It is calculated the orientation of the patient and from this a pressure distribution of the patient, with whom he rests on a pad determined.

In WO 2009/064788 A1 another method for detecting the movement of a patient lying in a bed is described. The patient wants with several different sensors, such. As a camera, a light beam matrix, an RFID grid system and the movement of the patient is recorded.

The techniques listed so far all use a combination of different sensors that need to be reconciled. However, most of these sensors can negatively affect the patient. Especially in sleep patients, it can be very annoying to have several sensors attached to the body. Light beam matrices have a similar disturbing effect. An everyday-like picture of sleep is difficult to carry out here. The effort per patient is also high insofar that the sensors must be individually adjusted and adjusted. Patients with epileptic seizures also have the risk of stripping off or even destroying the sensors during seizures. A hidden type of surveillance, which provides the patient as possible a familiar environment, would therefore be desirable. Furthermore, sleep aids, such as a blanket, can mask the sensors. In fact, a blanket not only obstructs sensors but also observers, such as sleep researchers. An analysis without a blanket may be useful, but it may be uncomfortable for many patients.

From Harada et al., Human motion tracking system based on skeleton and surface integration model, [online] Proceedings of the Workshop on Human Motion, Los Alamitos, Calif., USA DOI: 10.1109 / HUMO.2000.897378, 2000- 12-08 (on May 10, 2017, http://www.ieeexplore.ieee.org/document/897378/) a method is known in which a patient lying in a bed is detected solely by means of pressure sensors , From a model of the patient comprising a skeleton model and a surface model, a print pattern is calculated and compared with the measured print pattern. The calculated print pattern is matched to the measured print pattern to the extent that the patient's placement on the mattress is determined. By detecting successive arrangements of a patient on the mattress, the movement of the patient on the mattress can be determined. These procedures should be able to detect gross movements of the patient. With this method, it is not necessary to attach sensors to a patient's body. The model of the patient is adapted to this.

In the method according to Harada et al. For example, a data model of a patient consisting of a skeleton and a surface model is generated that is used to generate an artificial print pattern. This artificially generated print pattern is subsequently matched with a measured print pattern by adjusting the parameters of the patient data model. Creating the patient data model requires a great deal of experience and intuition to get a realistic print pattern. Especially in patients with abnormal body measurements and / or anomalies it is extremely difficult to produce a realistic model. In addition, manual initialization is necessary to equate the orientation and position of the data model with the real patient. Only then can changes in movement and position be displayed. In the event of a measurement interruption, the initialization must be repeated, which makes the method unsuitable for long-term measurements. Both the generation of the patient data model, as well as each individual initialization require considerable pre-processing before the actual measurement process.

In medicine, it is often desirable to measure and monitor the position and orientation of a patient on a mattress. In particular, the temporal monitoring of the lying position, the mobility of the patient and the type of movements can give conclusions about the clinical picture of the patient. For example, the data can be used to avoid pressure points that occur again and again, especially in long-term patients. But it can also track the sleep behavior. In addition, convulsions and epileptic seizures can be understood in this way.

The invention is thus based on the object to measure and represent the position and orientation of a patient in a way that does not hinder the patient.

Another task is long time stability of the measurement. The patient should be measured for at least one night, rather over several days, weeks and months, without having to recalibrate the system.

Another task is that each patient can be measured and monitored without much technical preparation.

Another object of the invention is the feasibility of the measurement, despite certain aids, such as a blanket.

One or more of the above objects are solved by the features of the independent claims. Preferred embodiments and advantageous developments form the subject of the dependent claims.

According to a first aspect of the present invention, a basic position of a person lying on a mattress provided with pressure sensors, hereinafter referred to as a patient, is automatically detected, wherein a print pattern is detected with the pressure sensors integrated in the mattress, severity points in the print pattern are determined, and a basic position of the patient is determined on the basis of the severity points on the basis of predetermined rules, the pressure values of the severity points and / or the course of the pressure values in the region of the center of gravity being taken into account with the rules.

Each gravity point is either a certain point on the mattress and / or a local maximum of the measured pressure. These severity points may be associated with a particular body part, such as head, shoulder, hip, leg, arm. For a leg or an arm, the severity points can be assigned to an elbow, a hand, a knee and / or a foot.

Such severity points can be easily determined from the print pattern. Based on these severity points, the basic position of the patient lying on the mattress can be detected reliably and easily. Basic positions are a supine position, a prone position and a lateral position. In the lateral position, a distinction can be made between a right and a left lateral position, wherein in the case of a right lateral position of the patient with its abdominal or front side, the mattress is directed to the right in the plan view and to the left in the case of a left lateral position. The basic position can also consist of a proportion of back, belly or side position.

By using pressure sensors to detect the home position, it is possible that the home position can be reliably detected even when the patient is covered with a blanket. For the patient lying on a mattress provided with pressure sensors, there is thus no difference to a conventional mattress, so that this patient does not have to restrict his sleep and lying behavior in any way.

The inventors of the present invention have recognized that it is very easy to determine the basic position of the patient lying on the mattress on the basis of a few points of gravity.

The mattress can be divided into different areas, in particular a head area, a shoulder area, a hip area and one or two leg areas. If two leg areas are provided, then the mattress is divided into a left and a right leg area. The pressure values measured in the different regions are taken into account differently in the rules for determining the basic position, in particular in order to assign the measured severity points to specific body parts. In principle, it is also possible to assign the severity points without subdividing the mattress into different areas of specific body parts. For this purpose, the orientation of the patient on the mattress can first be determined from the arrangement and the relative intensities of the points of gravity. The orientation of the patient determines where the patient's head or foot end lies on the mattress. Once the orientation of the patient has been determined, the individual points of gravity can simply be assigned to specific parts of the body, in which case a head-a hip and at least one leg-center of gravity are determined in particular. However, other severity points such as a shoulder gravity point and one or two hand gravity points may also be determined.

To determine the basic position of the patient lying on the mattress, the following rules can be used individually or in combination: the basic position is recognized as a lateral position if one of the two leg areas is unloaded, or if there is only a single gravity point in both leg areas, or if only a single leg area is detected, whereby the orientation of the lateral plies can also be determined; the basic position is recognized as a lateral position if the pressure value of the shoulder severity point and / or the hip severity point exceeds a predetermined value, the basic position is recognized as a lateral position if the the maximum slope of a pressure curve in the pressure pattern in the region of the shoulder gravity point and / or in the area of the hip center of gravity exceeds a predetermined value, - the basic position is recognized as a lateral position when the area occupied by a particular isobar in the shoulder area is closed, falls below a predetermined value. - The home position is recognized as a prone position, when at least two leg gravity points are measured, the pressure value of the head and / or shoulder severity point is above a predetermined value and / or the pressure value of the hip severity point is below a predetermined value, and / or Pressure profile in the pressure pattern in the region of the shoulder gravity point and / or in the region of the hip center of gravity is below a predetermined value, - The home position is detected as a supine position when at least two leg centers of gravity are measured, the pressure value of the head and / or shoulder Center of gravity is below a predetermined value and / or the pressure value of the hip center of gravity is above a predetermined value and / or if the maximum slope of a pressure profile in the pressure pattern in the region of the shoulder center of gravity and / or in the region of the hip center of gravity below a predetermined Value lies, - the basic position is recognized as supine, if at least two feet-heavy The baseline position is recognized as a prone position when at least two knee severity points are measured.

According to a second aspect of the present invention, a method is proposed, which comprises: - recording a digital print pattern, preferably of pressure sensors integrated in a mattress - comparing this print pattern with a database of stored print patterns - assigning the measured print pattern to similar print patterns of the database, which Skeletal data are again associated with the representation of the skeleton data assigned to the printing pattern on a display device, this representation reproducing the position and orientation of the patient.

For the purposes of the invention, a print pattern is a digital file in which a plurality of three-dimensional vectors is stored. These vectors contain two coordinates and a pressure value. This file is thus similar to a BMP image file and could also be stored as such and read out. The printing patterns are ideally produced by printing mattresses, as described for example in AT 518046 A. For this reason, reference is made in full to AT 518046 A.

For each print pattern in the database, there are corresponding skeleton data representing a human skeleton in a simplified form. This simplified form includes an array of so-called joints that represent the joints and bones that form joints between the joints and represent the bones.

One way to create print patterns with linked skeleton data is to use a motion capture method. Subjects who place themselves on a pressure sensor mattress are additionally equipped with motion capturing sensors, the so-called markers. A camera captures the subjects and an analysis software recognizes the markers in the image, whereby a three-dimensional image of the subject, or the skeleton of the subject can be calculated. On the computer, the generated print patterns can be compared with the skeleton data provided by the Motion

Capturing were generated. Furthermore, other techniques are conceivable, such as GPS systems or gyromagnetic systems.

If a print pattern of a patient corresponds to the print pattern from a database, the skeleton data of the database also correspond to the patient. The skeletal data contains all the necessary information about the position and orientation of the patient.

The determination of the similarity of the measured print patterns to the print patterns of the database can be done by different methods. For example, the mean square distance value may be used as a measure of similarity. Before comparing to determine similarity, the measured print patterns can be normalized, as explained in more detail below. In the comparison for determining the similarity, the measured pressure patterns and the pressure patterns of the database can be positioned relative to one another, whereby, for example, the distance of the local maxima or the distance of the center of gravity is minimized. In particular, the severity of the shoulder and the hip are compared. Only after normalization is the similarity of the printed patterns to be compared calculated.

For the purposes of the invention, a basic position is a classification of the position of the patient. Thus, the lead position can be divided into the following four basic positions: - prone - supine - left lateral - right lateral

The basic position can also consist of a proportion of back, belly or side position.

In a preferred embodiment it can be provided that a plurality of print patterns, each of which constitutes a data point, is located in the database. The dimension space of the database is spanned by the parameters of the patient data of the patient. For example, a print sample can contain the parameters basic position, height, total weight, gender, age and body fat.

Content. The dimension space may then be constrained by some or all of the parameters, thereby creating a subset to simplify or refine the comparison with the measured print pattern. For example, the pelvic structure is different in women than in males and it is therefore convenient to compare female patients only with patterns of print coming from female subjects.

Furthermore, it is useful if the dimension space can also be subsequently specified and extended. For example, it may be useful to extend the age classification for a small number of datasets and compare the measured pressure structure to a dataset generated by slightly older and / or slightly younger subjects. With a large number of data records, it may be useful to compare the print samples only with peers. In addition, it certainly makes sense to use more parameters for pre-sorting in a growing database. In addition, it may be that at a later date, parameters emerge that were previously not thought of or that were classified too specific. For example, subjects with certain (bone) malformations could be included in the database at some point in time, which would then benefit other patients with similar or similar characteristics.

In a further preferred embodiment it can be provided that the measured print pattern is normalized by transformation, such as scaling, mirroring, translation and / or rotation. Thus, the measured pressure pattern of a patient lying upside down or obliquely in bed can still be compared to the pressure pattern of a normally oriented lying patient. Also, the place where the patient lies on the mattress, then has no effect on the result. The required data record for recognizing the position is thereby reduced considerably.

In a further preferred embodiment it can be provided that, if several database print patterns match the measured print pattern, this is interpolated and weighted accordingly. This allows an increased accuracy in the representation of the body position.

It is useful if the interpolated print patterns with the associated skeleton data can be added as new data points to the database. In conjunction with the stored patient data, over time an ever denser network of data points in the multidimensional database can be developed, which increases the effectiveness of the invention.

In a further preferred embodiment it can be provided that certain patient data are determined automatically. For example, by calibrating the pressure mattress, the weight of the patient can be determined very easily. The body size is comparatively easily determined by the maximum distance between two pressure points on the mattress. As a result of these two measurements, the BMI and, under certain conditions, the body fat percentage can be determined. If the database of print patterns is powerful and the algorithm finely set, all necessary metainformation can be read from the print pattern. Even non-trivial parameters, such as age, should be read. A manual input is then no longer necessary and the reduction of the database is carried out automatically.

In a further preferred embodiment it can be provided that the skeleton data are used to generate a three-dimensional image of the patient. For this, similar to the "motion capture" method, the rigid positions and orientation of the bones and joints are used as a basis, which make up the skeleton data.

To create a three-dimensional arrangement of the bones and joints, a virtual elastic skin is drawn, which must necessarily adhere to the arrangement of the bones and joints. The position of this virtual body, also called avatar, then corresponds to that of the patient.

In a further preferred embodiment, provision may be made for the patient data of the patient to be used in order to render the avatar, that is to say the body wrap enclosing the bones and joints, even more realistic. This can range, for example, from the simple distinction between man and woman, to the physically correct behavior of the tissue. So the mass distribution in a human body with high body fat content is different, as with someone with low body fat content. There are no limits to the level of detail of the avatar, so that photorealistic avatars, which correspond to the patient very closely, also resemble the animated movement of the real person. While this may be of interest for cinematic effects, stylized avatars, such as those called body scanners, are probably more appropriate to not expose the patient.

In a further preferred embodiment it can be provided that the virtual three-dimensional body is displayed in a two-dimensional image so that it can be projected onto a monitor and / or beamer. Displaying three-dimensional objects, especially when they are moving, is by no means trivial. Since the dimensional reduction inevitably leads to a loss of information, it must be decided how to limit this loss. For example, it is usually not appropriate to project the patient's avatar onto the plane of the sole of the foot.

In a further preferred embodiment it can be provided that a plurality of chronologically successive printed patterns are measured. Each individual print pattern is now also compared to the database. This is especially helpful to avoid pressure sores in the patient or to detect hypermobility. Furthermore, temporal changes in patient data, such as weight, can be precisely measured. This is particularly useful in patients whose fluid balance needs to be monitored. Due to the frequent repetition of the measurement, the time course of the weight loss due to the elimination of water can be accurately determined.

If the temporally successive printed patterns and / or the associated skeleton data are stored, a sequence of movements can be created on the basis of these data. One possible use of this sequence of movements is the representation as a film at a sufficiently high measuring speed. For example, sleep patterns or convulsions, such as epileptic seizures, can also be analyzed. If the duration which the comparison module needs with the subsequent imaging method is equal to or less than the measurement duration of a print pattern, the virtual body can be displayed live. This live presentation may be especially helpful where constant monitoring of the patient is needed, but the supervising person can not be present in the same room. Furthermore, this form of monitoring can display movements under the ceiling, which remain hidden in other types of surveillance, such as by video camera. Also, a person who is right next to the bed can only guess the movements under the ceiling and may possibly draw the wrong conclusions. For example, a spasm can go undetected, whereas live monitoring via the pressure mattress can show it.

In a further preferred embodiment it can be provided that in the measurement of a chronological sequence of printed patterns, the printed patterns, the skeleton data and / or the body wrap are interpolated between two images in order to be able to represent the sequence of movements even less haltingly.

In a further preferred embodiment it can be provided that the movements and typical movement patterns are included in the print pattern adjustment. For example, the patient may lift his legs and then drop them onto the mattress. The print patterns could then be corrupted if the motion impulse was not taken into account. Furthermore, in the case of an abdominal patient it can be ruled out that the next lying position will be on the back. To get there, the patient first has to turn around. It is therefore expedient to put the database printing patterns in temporal relationship to each other.

In a further preferred embodiment it can be provided that medical details such as, for example, pressure ulcers, intervertebral discs, organs and the like are drawn into the illustrated three-dimensional body, the avatar. For example, ulcers or pressure points can be tracked in time. It also makes it easier to determine effects on the movement process.

In a further preferred embodiment, provision may be made for the mattress or parts of the mattress to be tilted, this tilting being measured indirectly by sensors on the mattress or indirectly by analysis of the pressure pattern and displayed accordingly in the representation of the patient.

In a further embodiment it can be provided that the pressure between the body and the mattress is measured in previously determined body parts. Patients with prolonged medical surgery, coma patients or obese patients are more prone to pressure sores. These are mainly caused by increased pressure on the corresponding body parts, which disturb the blood flow of the underlying tissue. With the present invention, it is possible to measure and monitor the pressure load at the individual body sites. Both a current pressure load and a temporally integrated pressure load can be measured. The above-described assignment to skeleton files, it is also possible to assign the pressure loads individual body parts. Even after a bed or a change in the lying position, the pressure load of the individual body parts is still properly assigned and measured. The effectiveness of the Umbettprozesses is thus directly measurable. Furthermore, this invention provides an additional control entity for an operating physician. During the operation, he can see the effects of certain actions and operations on the pressure load. Especially in frail patients or in operations where there is a large pressure transfer to the patient, or in very long-lasting operations, it is important to monitor the pressure load. In combination with a 3D camera, which is located above the patient and records the movements of the doctor, it can be used to measure precise information about how the doctor is doing. Previously, only a two- or three-dimensional image was transmitted, for example, a doctor who cuts through a bone, so now observers can also understand what pressure the doctor exerts. On the other hand, it can be determined by the video surveillance, where the starting point of the doctor. If the data has been recorded, this possibility is also interesting for the pathology in order to ensure quality assurance during the operation.

In a further embodiment it can be provided that the mattress and / or the bed in previously set events, such as too long

Pressure load of a particular body part, mechanically counteracts to minimize the pressure load. For example, a coma patient who has been in a position too long will be brought to a new position by automatic tilting of the mattress. Another possibility would be a variety of bolts, which are built into the mattress and adapt to the pressure loads in time. A massage device would also be conceivable here.

The invention will be explained in more detail by way of example with reference to the drawings. The drawings show schematically in:

FIG. 1 shows a method for measuring and displaying the orientation and position of a patient in a flow chart;

FIG. 2 shows a schematic representation of a key unit according to an exemplary embodiment of the present invention in a block diagram,

FIG. 3 shows a schematic illustration of the method for measuring and displaying the orientation and position of a patient in a block diagram,

FIG. 4 a schematic representation of gravity points and isobars (surface pressure lines) in a side view of a printing pattern,

Figure 5 is a schematic representation of skeleton data and isobars (surface pressure lines) on a printing pattern, and

Figure 6 is a schematic representation of an avatar on a 3D printed pattern in side view.

The present invention will now be described in detail by way of preferred embodiments with reference to the accompanying drawings. It should be understood that the pictorial representations are purely schematic and not necessarily to scale. Likewise, it should be noted that the drawings and the following description are intended for the understanding of the

Focus on helpful features, without thereby limiting the scope of the present invention, which is defined by the appended claims in the broadest sense.

In the context of this application, indications such as perpendicular, parallel, collinear, coplanar, the same or the like are not to be understood exclusively in the geometrically or mathematically exact sense, but may include an area within the scope of technically reasonable tolerances. The understanding of the technically meaningful can depend on the effect to be achieved. The same applies to figures or values.

In the measuring method according to the invention, the position and orientation of a human body, referred to hereinafter as patient 1, are measured with a measuring device (FIG. 1). The measuring device has at least one scanning unit 2 for generating printing patterns 3, a comparison unit 4 for matching the measured printing pattern 3 with other printing patterns 5, and a display unit 6 for displaying an avatar 7 of the patient 1.

As shown in FIG. 2, the button unit 2 comprises a bed 8, which may also be designed as an operating table or a stretcher, a control unit 9 and a pattern evaluation unit 10. The patient 1 is identified by patient data 11, such as size, weight and Gender, characterized. The bed 8 has a pressure sensor mattress 12. The pressure sensor mattress 12 is identified from a mattress and a matrix-like arrangement of sensors 13. The sensors 13 are designed as optical pressure sensors which respond to a change in a pressure force acting on the sensor with a change in a quality of a light supplied and discharged via optical fibers. Each sensor has for this purpose a first, incoming optical fiber and a second, outgoing optical fiber, which form a light guide pair 14. The pairs of optical fibers 14 are combined in the region of the bed 5 to form a light conductor collecting strand 15. The optical fiber collecting string 15 is divided into an input sub-string 16, which contains the incoming optical fibers of the sensors, and an output sub-string 17, which contains the outgoing optical fibers of the sensors. The sub-strands 16 and 17 lead to the control unit 9, which will be described in detail below. Such a key unit is described for example in AT 518046 A1.

The sensors 13 are configured and arranged to sense a pressure load on the mattress 12 at their respective locations in the matrix.

The control unit 9 is assigned to a specific bed 8 or a specific matrix of the sensors 13 of the mattress 12. If several beds are provided, each bed 8 can be assigned its own control unit 9. The control unit 9 has a processor unit 18, a power supply 19, a plurality of external interfaces 20, a light source 21, a light detection unit 22 and a memory unit 23, which interconnects in a conventional manner known per se, ie in terms of signaling (with respect to FIG a data line) and energy-technically (in terms of a power supply) are suitably interconnected. The control unit 9 has a power supply 19 and electrically active components. However, the control unit 9 can be placed at a safe distance (for example 1-5 m) from the bed 8 and the components installed in the mattress 12 (sensors 13, light guide pairs 11) are de-energized, so that there is a risk of electric currents on the Bed dormant patients can be avoided.

The pattern evaluation unit 10 is connected to the control unit 9 with a data connection and has an electronic data processing module for pattern analysis 24 with input interfaces for data records 25. The electronic data processing module 24 has a housing 26, a processor unit 27, external access interfaces 28, monitor interfaces 29, interfaces for a bidirectional interactive system 30 and interfaces for external data exchange 31. The processor unit 27 has visualization units 32, data management units 33, data storage units 34 and data transfer units 35. The input interface for data records 25 is designed to receive data records of a plurality of control units, such as the control unit 9.

With an evaluation program module 36 executed on the evaluation unit 10, the data records from the control unit 9 are converted into a print pattern 3 and displayed. The print patterns 3 are transferred via a data line 37 into the comparison unit 4.

The comparison unit 4 has an electronic data processing module for print pattern matching 38. It is designed like the electronic data processing module for pattern analysis 24. On the electronic data processing module for print pattern matching 38, a comparison program module 39 is stored and executable. The comparison program module 39 has a basic position module 40 for determining a basic position 41 of the patient 1, a print pattern database 42, a reduced print pattern database 43 and a comparison module 44 for matching the measured print pattern 3 with one or more database print patterns 5.

The print pattern database 42 comprises a plurality of database entries 45, each of which in turn consists of a database print pattern 5, associated skeleton data 46, and subject data 47 (FIG. 3). Each database entry 45 corresponds to a position and orientation of a subject 48.

A subject 48 is a human body whose print pattern 5 and skeleton data 46 were acquired prior to the patient 1 being measured. Skeletal data 46 is data representing the skeleton of a subject 48, so that the relationships of the individual body parts to each other are displayed in a simplified form. The skeletal data 46 comprises a collection of points, the so-called joints 49, which represent in a simplified form certain joints of the subject 48, and the joints of the joints 49 the so-called bones 50, which represent the bones of the subject 48 simplified.

The reduced print pattern database 43 is connected to the print pattern database 42 and takes over the database entries 45 from the print pattern database 42, whereby only entries with matching home position and with subject data 47 that match the patient data 11 are transmitted.

The skeleton data 46, which are selected by the comparison module 44, are transferred via a skeleton data line 51 into the presentation unit 6.

The presentation unit 6 has an electronic data processing module for visualization 52. It is embodied like the electronic data processing module for pattern analysis 24 and the electronic data processing module for print pattern matching 38. On the electronic data processing module for visualization 52, a presentation program module 53 is executed which generates an avatar 7 from the skeleton data 46.

The method for measuring and displaying the orientation and position of a patient 1 will be explained below with reference to FIG.

The first step of the method involves data acquisition. In the key unit 2, the patient 1 lays himself on the bed 8. With each of the sensors 13 in the mattress 12 of the bed 8, a pressure which the patient 1 exerts on the mattress 12 at the location of the respective sensor is measured. This happens because light beam exits from the transmitting element of the respective sensor 13 and is partially taken up again in the receiving element of the sensor 13. The amount of light received is proportional to the distance between the two elements. Since the distance changes in proportion to the pressure applied to the sensor, the amount of light received is also proportional to the pressure. The amount of light received is measured and digitized via light sensors. These data are transmitted via the control unit 9 to the pattern evaluation unit 10.

The pattern evaluation unit 10 analyzes the data records from the control unit 9 and composes them into a print pattern 3. Each pressure sensor 13 is assigned a position in a pressure matrix, which reproduces the arrangement of the sensors in the mattress. The value in this matrix corresponds to the measured pressure of the associated sensor 13. This pressure matrix is represented as a print pattern 3 and forwarded via the print pattern data line 37 to the electronic data processing module for print pattern matching 38.

The data acquisition is described in detail, for example, in AT 518046 A1. In this respect, reference is made in full to AT 518046 A1.

The second step of the method of comparing the measured print pattern 3 with the database print patterns 5 is performed in the comparison unit 4. In the electronic data processing module for print pattern matching 38, the print pattern 3 is first used to determine a basic position 41 of the patient 1 in the basic position module 40.

For this purpose, the print pattern 3 is divided into regions 54, for example in a head region 55 (upper 12.5%), a shoulder region 56 (the 25% adjoining the head region), a hip region 57 (the 25% adjoining the shoulder region), and two leg regions 58 (the remaining 37.5% in the longitudinal direction, divided in the transverse direction).

Gravity points 59 and isobars 60 are determined in these areas. The gravity points 59 represent the position of the largest measured pressure in the respective regions in the print pattern 3. The gravity points 59 are thus local maxima of the pressure. The isobars 60 are lines of equal pressure formed at discrete intervals around the gravity points 59. The home position is determined by rules that take into account the positions of the gravity points 59 and / or that take into account the size of the surfaces including the isobars 60 and / or take into account the slope of a pressure curve in the print pattern 3. For example, the following rules for determining the basic position 41 can be used individually or in combination: The basic position 41 is recognized as a lateral position if one of the two leg regions 58 is unloaded, or if only a single leg region 58 is detected, as a result of which the orientation of the Side layers can also be determined. The home position 41 is recognized as a lateral position when the pressure value of the shoulder severity point or the pressure value of the hip severity point exceeds a predetermined value. The basic position 41 is recognized as a lateral position if the slope of a pressure curve in the pressure pattern 3 in the region of the shoulder severity point and / or in the region of the hip severity point exceeds a predetermined value. The basic position 41 is recognized as a lateral position if the area enclosed by a particular isobar 60 in the shoulder area 56 falls below a predetermined value. The home position 41 is recognized as a prone position when a pressure in both leg regions 58 is measured, the intensity of the pressure at the point of gravity of the head / shoulder region is above a predetermined value and / or the intensity of the pressure at the severity point of the trunk / pelvic region below a predetermined value Value is. The home position 41 is detected as a supine position when a pressure in both leg regions 58 is measured, the intensity of the pressure at the point of gravity of the head / shoulder region is below a predetermined value and / or the intensity of the pressure at the center of gravity of the trunk / pelvic region above a predetermined one Value is. The basic position 41 is recognized as a supine position when the positions of the leg points of gravity are below a predetermined limit.

As a next step, the number of database printing patterns 5 to be compared is reduced within the comparison unit 4. For this purpose, database entries 45 of the print pattern database 42 are read out according to predetermined criteria and stored in the reduced database 43. These criteria preferably include the determined home position 41. For example, only database entries 45 with reserve print patterns are selected when the patient 1 is supine. If no pure basic position 41 but an intermediate position was found, database entries 45 with the same or similar intermediate position can be selected. Preferably, the proportions of the basic positions contained in the intermediate position should not deviate more than, for example, 10% from the measured intermediate position. If, for example, a basic position has been determined that consists of 80% supine and 20% lateral position, this can be compared to a printed pattern consisting of 82% supine and 18% lateral position. Alternatively, all involved pure basic positions can also be taken over into the reduced database 43.

However, additional or alternative criteria may also be selected. For example, patient data 11 may be taken into account so that only database entries 45 are selected whose subject data 47 matches or is similar to the patient data 11. For example, patient data 11 may describe gender, weight, body fat percentage, or height.

Thus, due to the different pelvic structure, if the sex matches, eg, only database entries 45 are selected by female subjects 48, the patient should be 1 female. It is advantageous if the required similarity in the age of the patient 1 scales with age. For example, print sample 3 may be very different from a 12 year old and a 14 year old, but there may be little difference between a 50 year old and a 52 year old boy if the other parameters match. A possible tolerance would be, for example, that the age of the subjects 48 differ by a maximum of 10% from the patient 1, so that the database entries 45 are transferred to the reduced database 43. Similarly, scaling of the similarity in weight would be advantageous. Ideally, only database entries 45 are taken over into the reduced database 43 whose subjects 48 have a weight that differs by a maximum of 5% from that of the patient 1. Another important feature of the patient is the body fat percentage, which can be expressed as body mass index (BMI), for example. Thus, for example, the pressure distribution of patients 1 with an increased BMI value is very different from subjects with a very low BMI value. It is therefore advantageous if the BMI value of the patient 1 does not differ more than 10% from that of the subjects 48. Although the size of patient 1 can be normalized as described below, ideally it should not differ more than 20% from the subject's data.

Another way of limiting the printing pattern database 42 is given by the angle of inclination of the mattress 12. Only database entries 45 whose print patterns were recorded with the same or similar tilting of the mattress are transferred to the reduced database 43. It is expedient if the tilt of the database entries 45 does not deviate more than 5 ° from the measured print pattern 3.

The comparison module 44 then compares the measured print pattern 3 with the database print patterns 5 in the reduce database 43. For the comparison, the print pattern 3 is first normalized by being shifted, scaled, mirrored and / or rotated. The normalization transforms the measured printing patterns 3 in such a way that they agree as closely as possible with the database printing patterns 5 in certain parameters, such as orientation, position and size. The number of required database printing pattern 5 is reduced considerably.

In order to scale the size, for example, the surface area of the surface enclosing certain isobars 60 is preferably adjusted. Furthermore, the measured print pattern 3 may be stretched or compressed such that the distance between two particular severity points 69 is equal to the distance of the same severity points in the database print pattern 5.

Also, to normalize the rotation, one or more particular axes in print pattern 3 may be determined. By rotating the print pattern, they are then parallelized to the corresponding axes in the database print pattern.

In addition, for example, the measured pressure pattern 3 can be adjusted such that the position of the patient 1 on the mattress has no influence on the measurement. One possibility would be to bring the absolute center of gravity of the print pattern 3 into agreement with the absolute center of gravity of the database print pattern 5.

It is also possible to combine several normalization steps. For example, in normalization, the middle between the shoulders can be identified as the first global gravity point. The second global gravity equals the hip gravity point. Perpendicular to one of these two points of gravity

Axis and shoulder axis and hip axis are formed at the shoulder and hip points of gravity. These axes are transformed by scaling, rotation, mirroring and / or translation with the corresponding axes of the database printing pattern 5 in such a way that, when the printing patterns are superimposed, these axes are at the same location with the same length, so that the deviation is small.

Preferably, each rotation / mirroring is performed such that the pattern is not distorted. Distortion can be carried out only in the case of a compression / extension, with the compression factor preferably being maintained in order to later reverse the compression.

Each normalized print pattern 3 is compared, for example, by the least squares method with each database print pattern 5 of the reduced database 43, and a corresponding difference value 61 is calculated. However, there are other methods by which difference features can be determined.

First, the best database print pattern 5, that is, the smallest difference value 61, is searched for. If this is below a predetermined threshold value, the matching print pattern 5 with the corresponding skeleton data 46 is selected. If further database printing patterns 5 with a similar difference value which, for example, only deviates by a maximum of 10% and in particular not more than 5% from the database printing pattern 5 with the greatest match, are found, these are likewise selected. If no database print pattern 5 with a difference value 61 below the previously determined threshold value has been found, then a predetermined number of database print patterns 5 are transmitted with the corresponding skeleton data 46. In this case, a warning can be output that no database print pattern 5 with a difference value 61 below the previously determined threshold value was found.

If several database print patterns 5 have been selected, these print patterns 5 can be interpolated to a new database print pattern 62 in order to ensure a better match. In this case, the given database print patterns 5 are composed in such a way that they reproduce the measured print pattern 3 as exactly as possible. The weighting factors are kept here. The linked skeleton data 46 are also interpolated with the weighting factors taken from the print patterns. The skeletal data 46 already contains the information on how the patient 1 is aligned in three-dimensional space. The skeleton data 46 is now transferred to the presentation unit 6. The interpolated print pattern 62 with the interpolated skeleton data forms a new data point and can be included as a new database entry 45 in the print pattern database 42.

The final step is the representation of the position and orientation of the patient 1 in the presentation unit 6. In this model, the presentation program module 53 from the skeletal data 46 models a three-dimensional image of the patient 1, a so-called avatar 7. The rigid framework of the skeletal data 46 is used. to arrange an elastic body wrap around the skeletal structure to create a simple stylized body figure, also called Mannequin or Avatar 7. The avatar 7 now occupies the same position and posture as the patient 1.

A two-dimensional projection of this avatar 7 is output via the monitor interface 29, which completes the process.

Movements can also be recorded and displayed using this procedure. Depending on the application, different measuring intervals can be set. In order to keep the amount of data generated and the corresponding evaluation capacities low, the repetition of data acquisition should not be less than 10 ms, 50 ms, 100 ms, 1 s, 5 s, 10 s or 30 s. In order to track the movements of patient 1, the repetition of data acquisition should not be more than 100 ms, 1 s, 5 s, 10 s, 30 s, 60 s or 120 s.

If the repetition rate of the display is above 10 Hz, this is taken as a movement. In the case of a live presentation, it is expedient to repeat the data acquisition correspondingly quickly.

With a short measuring interval, the printing pattern 3 can be adapted due to the movements. In the comparison of the print patterns, the movement of the patient 1 is included in the calculation.

Furthermore, a long-term measurement, which can take over hours, days, weeks, months and even years, provides information about the lying and especially the Liegelernverhalten of the patient 1. For this purpose, the positions of the patient 1 are recorded and stored time. The data can be evaluated differently. Thus, the movements and movement patterns, as well as the frequency and duration of a particular lying position can be determined. Similarly, the frequency of position changes and the frequency of getting up at night from bed 8 can be detected.

If a patient 1 is measured several times in succession, the scaling factor, if not specified, can be specified more precisely with each measurement. For example, body size will not change significantly and the print pattern will always be scaled in the same way. Rotations of the print pattern, however, have to be rechecked with each measurement. But here too, certain limitations may be beneficial to refining the reduced database 43. For example, the patient can only turn in bed at a certain speed. Certain changes in the rotation from measurement to measurement are not possible. Again, it is advantageous for other parameters to measure their time course. For example, the weight change can be measured precisely as the total pressure is measured several times per hour, or per minute.

Another possibility of the invention is to determine a medically relevant detail, such as pressure ulcers, intervertebral discs, organs and the like, and to draw it into the avatar 7 and store it as an element in the skeletal data 46. These can be linked to other medical details. For example, conclusions can be drawn on the influence of pressure on near-surface organs. Furthermore, for example, the levels of magnetic resonance tomography recordings can be entered so that the corresponding recordings are connected to the avatar 7 and can be displayed accordingly.

It is also possible for the mattress 12 or parts of the mattress 12 to be tilted, this tilt being measured indirectly by sensors 13 on the mattress 12 or indirectly by analysis of the print pattern 3. During the evaluation, this inclination is taken into account and displayed accordingly when displaying the patient 1 as the avatar 7.

Furthermore, it is possible to monitor the pressure load of previously determined body parts of a patient 1. The information contained in the print pattern 3 on the pressure exerted by the body on the mattress is equal to the pressure load on the patient 1. As described above, the skeleton data 46 are assigned to the print pattern 3, whereby an assignment to body sites is possible. As a result, a pressure load check for the individual body parts is possible. Here, both a current pressure load can be measured and displayed, as well as a temporally integrated pressure load. Especially the latter is important for the prevention of pressure ulcers. The pressure loads can be drawn in the form of color codes in the avatar 7. For example, if the patient 1 is lying too long on his back, after some time the back of the avatar 7 turns redder and redder as the integral pressure load increases.

The pressure load check may be extended with external monitoring, for example a 3D camera located above the patient 1. External influences can influence the development of pressure sores. A camera, in particular a 3D camera, which is arranged above the patient, allows access to information which takes place above or outside the patient, while the pressure sensors 13 can measure information below the patient. So it will open up another level of measurement. The use of a 3D camera makes it possible to precisely check the external influence on the pressure load. This can be, for example, a surgeon during an operation, whose actions are so even more understandable. The 3D camera can be used to determine which action generates which pressure. In the case of pressure-critical actions, the acute pressure load can be output by color signals. For example, under certain circumstances, a cardiac surgeon must open the thorax and often put pressure on the patient there. Now it may be that the patient has particularly unstable bones. Signals, such as signal colors or sounds, would warn the surgeon of critical actions here.

Likewise, it is possible that the mattress 12 and / or the bed 8 mechanically act against previously set events, such as the too long pressure load of a particular body part, to minimize the pressure load and counteract a long-lasting, or unfavorable pressure load , In particular, in this case, the integral pressure load of individual body parts can be considered, wherein the mechanical counteracting is triggered automatically, for example, if a certain threshold value of the integral pressure load has been exceeded. This can be achieved, for example, by automatically tilting the mattress so that patients are relocated, or by bolts that are built into the mattress and can be adjusted to relieve overstressed body regions.

In an alternative embodiment, the basic position module 40 is part of the pattern evaluation unit 10. Instead of the print pattern 3, the mattress 12 is now divided into regions 54. The evaluation of these areas and the associated severity points 59 and isobars 60 follows analogously to the exemplary embodiment described above.

In a further alternative embodiment, the gravity points 59 are determined independently of the regions 54 as local maxima of the pressure in the print pattern 3.

In a further alternative embodiment, at least two of the following units of the comparison unit 4, the presentation unit 6 or the evaluation unit 10 use the same electronic data processing module 24.

In another alternative embodiment, the bone 50, which represents the spine, will be flexible. In this standardization, the shoulders and hips are no longer necessarily parallel to each other.

In a further alternative embodiment, the normalization of the measured pressure pattern 3 is carried out in such a way, the lengths of the individual bones 50 are normalized individually, the normalization being specified by error calculation and error correction.

List of Reference Numerals and Symbols 1 patient 32 visualization unit 2 key unit 33 data management unit 3 print pattern 34 data storage unit 4 comparison unit 35 data transfer unit 5 database print pattern 36 evaluation program module 6 presentation unit 37 print pattern data line 7 avatar 38 electronic 8 bed data processing module to 9 control unit print pattern matching 10 pattern evaluation unit 39 comparison program module 11 Patient data 40 Basic position module 12 Mattress 41 Basic position 13 Sensor 42 Printed data base 14 Fiber optic pair 43 Reduced database 15 Fiber-optic collective 44 Comparison module 16 Input sub-string 45 Database entries 17 Output sub-string 46 Skeleton data 18 Processor unit 47 Sample data 19 Power supply 48 test subject 20 external interface 49 joints 21 light source 50 bones 22 light detection unit 51 skeleton data line 23 memory unit 52 electronic 24 electronic data processing module for

Data processing module for visualization

Pattern analysis 53 Presentation program module 25 Input interface for 54 area

Data sets 55 Head area 26 Housing 56 Shoulder area 27 Processor unit 57 Hip area 28 External access interface 58 Leg area 29 Monitor interface 59 Severity points 30 Interface for bidirectional 60 Isobar

Interactive System 61 Difference 31 Interface for 62 Interpolated External Data Exchange Database Print Pattern

Claims (29)

claims A method of automatically detecting a home position (41) of a person lying on a mattress (12) provided with pressure sensors (13), also called a patient (1), wherein a print pattern (3) matches that in the mattress (12), in FIG a support, or in bed (8), which may also be designed as an operating table or support, integrated pressure sensors (13) is detected, severity points (59) in the print pattern (3) are determined, and a basic position (41) of the patient (1 ) is determined by means of predetermined rules on the basis of the gravity points (59), wherein the pressure values of the gravity points (59) and / or the course of the pressure values in the region of the gravity points (59) are taken into account with the rules. 2. The method according to claim 1, characterized in that each gravity point (59) either a certain point on the mattress (12) and / or a local maximum of the measured pressure, preferably in a certain area of the mattress (12). 3. The method according to any one of claims 1 or 2, characterized in that the mattress (12) in areas (54), such as a head portion (55), a shoulder region (56), a hip region (57) and a left and right leg area (58), wherein measured pressure values of the different areas (54) are considered differently in the rules and / or the arrangement and the relative intensities of the gravity points (59) the orientation of the patient (1) on the mattress (12) is considered and at least some severity points (59) are assigned to a body part, in particular a head, a shoulder, a hip and at least one leg severity point are determined. 4. The method according to claim 3, characterized in that the basic position (41) is recognized as a side position when one of the two leg portions (58) is unloaded, or if only a single leg portion (58) is detected, whereby the orientation of the side layers also can be determined. 5. The method according to any one of claims 3 or 4, characterized in that the basic position (41) is recognized as a side position when the pressure value of the shoulder severity point or the pressure value of the hip severity point exceeds a predetermined value. 6. The method according to any one of claims 3 to 5, characterized in that the basic position (41) is recognized as a side position when the slope of a pressure curve in the print pattern (3) in the region of the shoulder severity point and / or in the region of the hip severity point exceeds a predetermined value. 7. The method according to any one of claims 3 to 6, characterized in that the base position (41) is recognized as a side position when the area which is trapped by a particular isobar in the shoulder region, falls below a predetermined value. 8. The method according to any one of claims 3 to 7, characterized in that the basic position (41) is detected as a prone position, when a pressure in both leg areas (58) is measured, the intensity of the pressure in the gravity point (59) of the head / shoulder area (55/56) is above a predetermined value and / or the intensity of the pressure at the point of gravity (59) of the hip region (57) is below a predetermined value. 9. The method according to any one of claims 3 to 8, characterized in that the base position (41) is detected as a supine position when a pressure in both leg areas (58) is measured, the intensity of the pressure in the point of gravity (59) of the head / shoulder area (55/56) is below a predetermined value and / or the intensity of the pressure at the point of gravity of the hip region (57) is above a predetermined value. 10. The method according to any one of claims 3 to 9, characterized in that the basic position (41) is detected as a supine position when the positions of the leg points of gravity are below a predetermined limit. 11. The method according to any one of claims 1 to 10, characterized in that the basic position (41) may also consist proportionally from back, belly or side position. 12. A method for generating a graphic representation of a lying on a mattress (12) patient (1) based on a pressure pattern (3) which is measured in the mattress (12) integrated pressure sensors (13), wherein a plurality of database printing pattern (5 ) are stored in a database (42) along with respective skeletal data (46), the skeletal data (46) being a simplified human skeleton located in the space above the mattress (12) in an array corresponding to the respective database Printing pattern (5), wherein the measured printing pattern (3) is compared with the database printing patterns (5) from a printing pattern database (42), and the similarity of the measured printing pattern (3) with the respective database printing pattern (5) is determined, one or more database print patterns (5) are selected with the respective best similarity selected to the one or more database print patterns (5) associated skeleton data (46) t, and the selected skeleton data (46) are displayed on a display device (6). A method according to claim 12, characterized in that prior to comparing the measured print pattern (3) with the database print patterns (5), the database print patterns (5) are evaluated according to predetermined criteria, e.g. the particular particular dimensions of skeletal parts, certain proportions, or patient data (11) associated with skeletal data, e.g. Size of the patient, sex, weight, body fat percentage, dimensions of various limbs, skeletal type or basic position (41) are selected, the base position (41), as explained in the claims 1 to 12 are measured, and the measured pressure pattern (3 ) is then compared only with the selected database print patterns (5) forming a subset of all database print patterns (5). 14. The method according to claim 13, characterized in that the data type containing the parameters of the patient data (11) can be subsequently extended and adapted. 15. The method according to any one of claims 12 to 14, characterized in that by transformation, such as scaling, mirroring, rotation and / or translation, the measured pressure pattern (3) is normalized. 16. The method according to any one of claims 12 to 15, characterized in that in an assignment of multiple database print patterns (5) to the measured print pattern (3), the corresponding skeleton data (46) are interpolated to form a skeleton. A method according to claim 16, characterized in that, in interpolating the plurality of skeleton data, these are weighted according to the evaluation of the correspondence with the measured printing pattern. 18. The method according to any one of claims 12 to 17, characterized in that the interpolated printing pattern (62) is recorded with the interpolated skeleton file and the patient data (11) as a new data point in the database (42). 19. The method according to any one of claims 12 to 18, characterized in that certain patient data (11), such as size and / or weight, automatically calculated from the print pattern (3) and output. 20. The method according to any one of claims 12 to 19, characterized in that the skeleton data for forming a virtual body, also called avatar (7), are covered with a virtual skin and this avatar is the graphic representation of lying on the mattress patient (1) forms. 21. The method according to any one of claims 12 to 20, characterized in that the avatar (7) in addition to the skeletal data and patient data (11) of the patient (1), such as the body fat fraction, includes. 22. The method according to any one of claims 12 to 21, characterized in that the avatar (7) is projected onto a screen on a two-dimensional image for display. 23. The method according to any one of claims 12 to 22, characterized in that a plurality of temporally successive print pattern (3) are measured and to each print pattern (3) generates a graphic representation (7) of the patient on the mattress (12) (1) becomes. 24. Method according to claim 21, characterized in that the skeleton assigned to it and / or its associated avatar (7) are stored for each measured print pattern (3) and are stored on the basis of this series of stored print patterns (3) and stored skeleton data (46 ) or series of stored avatars (7) a sequence of movements is generated. 25. Method according to claim 22, characterized in that avatars (7) are interpolated between the individual calculated avatars (7) in order to display the sequence of movements as a film. 26. The method according to any one of claims 23 to 25, characterized in that in the determination of the skeleton, the movement sequence and typical movement patterns are taken into account. 27. The method according to any one of claims 12 to 26, characterized in that relevant medical details, such as pressure ulcers, discs, organs and the like are drawn in the illustrated three-dimensional body and stored as an element in the skeletal data (46). 28. The method according to any one of claims 12 to 27, characterized in that the mattress (12) or parts of the mattress (12) is tilted, said tilting by sensors (13) on the mattress (12) directly or by analysis of the printing pattern (3) measured indirectly and displayed accordingly in the presentation of the patient (1). 29. The method according to any one of claims 12 to 28, characterized in that the acute and temporally integrated pressure loads through the sensors (13) in the mattress (12), in a support, or in the bed (8), which also serves as an operating table or wearer can be trained, is measured and assigned to individual body parts, so that the pressure load monitoring can be maintained even after a reburial. 30. The method according to any one of claims 12 to 29, characterized in that the mattress and / or the bed at previously set events, such as the excessive pressure load of a particular body part, counteract mechanically to minimize the pressure load.