A computer-implemented method for user fall assessment implementing a trained machine learning model with an action plan

A computer-implemented method using a trained machine learning model for fall assessment continuously processes sensor and positioning data to calculate FRP, addressing reliability issues in existing methods by providing adaptive and personalized preventive measures for enhanced user safety.

JP2024543801APending Publication Date: 2024-11-26AUTOLIV DEV AB
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
JP2024524659
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2021-10-29
Filing Date
2022-10-28
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

Existing fall detection and risk assessment methods lack reliability and fail to provide timely, accurate, and adaptive responses, especially in dynamic environments and for various activities such as walking, biking, or riding vehicles, and often require intrusive or privacy-compromising solutions.

Method used

A computer-implemented method using a trained machine learning model for user fall assessment that continuously acquires and processes data from sensors and positioning systems to calculate Fall Risk Probability (FRP), updating an action plan based on real-time and historical information, including environmental conditions, user health, and geographic location, and automatically adjusts preventive measures.

Benefits of technology

Provides accurate, real-time fall detection and risk assessment, enabling proactive preventive measures tailored to individual and environmental conditions, enhancing user safety by minimizing fall risks through adaptive and personalized action plans.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024543801000001_ABST
    Figure 2024543801000001_ABST
Patent Text Reader

Abstract

The present disclosure relates to a computer-implemented method for user fall assessment for a fall assessment environment. The computer-implemented method includes the steps of implementing a training machine learning model (S100) including an action plan, continuously acquiring data (S200), continuously processing the data, fall assessment for a user (1) (S300), and validating the action plan over time (S400). The validation is based on a fall risk probability, a continuous calculation of a FRP (S410), and monitoring whether and how the FRP changes (S420). If a set condition for updating the action plan is met in the monitoring, the method includes a step of updating the action plan (S500). The fall assessment includes fall detection based on continuous data and information, and the continuously acquired data includes data detected by sensors (2, 3, 6, 7) worn by the user (1) and positioning data of the user (1). The calculation of the FRP also includes the positioning data.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present disclosure relates to an improved computer-implemented method for user fall assessment, a computer program with computer readable instructions for applying the computer-implemented method, a computer program with computer readable instructions for training machine learning models, a computer program with computer readable instructions for a "supervisory machine learning module", a computer readable medium with a computer program, and a control unit arrangement, device and system for user fall assessment. [Background technology]

[0002] Falls are one of the leading causes of human injuries and injury-related deaths. Fall detection and fall prediction are performed with machine learning techniques with increasing utility. Machine learning techniques are used today, for example, to distinguish between fall and non-fall activities. Broadly speaking, these techniques may include techniques involving vision-based configurations that can classify falls in real time based on live feed video. Vision-based configurations may be less practical in some cases, for example in external environments, and may involve potential privacy issues. Furthermore, there are machine learning techniques that focus on more pervasive solutions based on wearable device sensors, which are non-intrusive and pervasive. In this field, see for example "A Cascade-Classifier Approach for Fall Detection, Putra et al., MOBIHEALTH 14-16 October 2015, London, UK, January 2015, DOI:10.4108 / eai.14-10-2015.2261619," which proposes a cascade-classifier approach for this purpose. Other examples include M. Musci, D. D. Martini, N. Blago, T. Facchinetti, and M. Piastra, "Fall Detection using Recurrent Neural Networks," p. 7, and F. J. Gonzalez-Canete and E. Casilari, "A Feasibility Study of the Use of Smartwatches in Wearable Fall Detection Systems," Sensors, Vol. 21, No. 6, p. 2254, March 2021, doi:10.3390 / s21062254.

[0003] WO 2018127506(A1) in the field of fall risk assessment for elderly people relates to an apparatus and method for triggering a fall risk alert for a person. Further, in WO 2018127506(A1), the time to trigger the fall risk alert may be determined using person-derived input parameters, and / or the person's environment, and / or further person-derived input parameters. WO 2018127506(A1) further discloses an integrated solution for adaptive multifactorial fall risk assessment, detection and prevention in elderly people.

[0004] A fall may occur, for example, to a person walking in a domestic environment or outside the home. In this context, a fall also includes falls occurring while riding a bicycle, scooter, motorcycle, etc., due to, for example, skidding, slipping, loss of control of the vehicle, collision with another vehicle or object, etc. Falls involving a moving vehicle are often more abrupt and require more immediate attention than falls occurring, for example, to a moving or slipping pedestrian. Summary of the Invention [Problem to be solved by the invention]

[0005] However, there is still room for improvement in this field, especially with regard to requirements in the area of ​​reliable methods for fall risk assessment.

[0006] This is accomplished by a computer-implemented method for a fall assessment environment and for user fall assessment, the computer-implemented method including the steps of: implementing a training machine learning model with an action plan, continuously acquiring data, continuously processing the data, fall assessment for the user, and validation of the action plan over time, the validation being based on a fall risk probability, continuous calculation of the FRP, and monitoring whether and how the FRP changes.

[0007] If a set condition for updating the action plan is met in the monitoring, the action plan is updated, and the fall assessment includes fall detection based on the continuous data and the continuous information. The continuously acquired data includes data detected by a sensor worn by the user and positioning data about the user, and the calculation of the FRP also includes the positioning data.

[0008] This means that the positioning data can be used to determine where a fall or near-fall occurred, indoors, outdoors, on stairs, on a slope, etc., and also, given the appropriate time / date stamps, whether the fall or near-fall occurred in a generally cold and icy environment, or in a generally warm environment. In the case of a motorcyclist, the positioning data can be used to determine on what type of road the fall or near-fall occurred, and in what environment the fall occurred, taking into account, for example, the road surface type, road conditions, and / or the amount of traffic on the road, all based on the positioning.

[0009] According to some aspects, the calculation of the FRP also includes environmental conditions, including weather conditions, at the user's determined location.

[0010] This means that real-time environmental conditions, for example in the form of weather data, can be used in the calculation of the FRP.

[0011] The computer-implemented methods described herein include fall detection based on continuous data and information, where the fall detection is detected using 3D sensor measurements and using a training ML model trained using training data collected from users, such as trainers or other wearers, e.g., motorcycle and bicycle riders. Once a fall is detected using a computer program, it may be recorded and used in the input to the calculation of the FRP.

[0012] The computer-implemented method described herein further includes continuous calculation of the FRP based on the continuous data FRI and the continuous information, the FRI comprising at least one of "skid", "slip", "loss of control", "collision and / or contact with another object or vehicle", "near-fall", "tripping while walking", and / or "slip", and optionally other suitable key performance indicators (KPIs) for the FRI.

[0013] The continuous calculation of the FRP is based on continuous data FRI and on real-time information, such as data collected from sensors worn by the user or other sensor types, as well as on continuous information that references historical information about the geographical environment, e.g. obtained from positioning data.

[0014] This means that the FRP is calculated using the FRI and real-time information, which provides an accurate measure of the FRP.

[0015] According to some aspects, the continuing information refers to information regarding the user age, physical and health profile. GPS (Global Positioning System) positioning may optionally be included.

[0016] This means that the continuous calculation of FRP is based on continuous data, continuous information and FRI, where FRI includes at least one of "skidding", "slip", "loss of control", "collision and / or contact with another object or vehicle", "near fall", "stumbling while walking", and / or "slip", and continuous information refers to real-time information such as data collected from sensors worn by the user, as well as historical information about the geographical environment obtained from positioning data.

[0017] According to some aspects, the computer-implemented method includes guidance according to an action plan, including activation and preventative measures, that correlates with and / or reflects a fall assessment for the user, which may include fall detection and calculation of a FRP for that particular user.

[0018] This means that preventative measures are possible, including countermeasures to avoid falls in the first place, before they occur.

[0019] According to some aspects, the computer-implemented method includes management, eg, continuity management, of the action plan.

[0020] The continuous calculation of the FRP further includes calculating FRPt1 at a first time t1 and calculating FRPt2 at a second time t2 after the first time t1. The step of monitoring whether and how the FRP changes includes calculating a difference between FRPt1 and FRPt2 (Δ FRP ), determining Δ FRP If is less than or equal to the set FRP threshold, the action plan is valid. Δ FRP If is greater than the set FRP threshold, update the action plan.

[0021] If a significant value of Δ correlates with a significant value of abnormality score, for example from the 6-minute walk test (6MWT), it validates that the FRP has indeed changed significantly and that the action plan needs to be updated at this point.

[0022] According to some aspects, the computer-implemented method includes managing, for example, a continuum of, the action plan. FRP If FRP is greater than the set FRP threshold, the validation is also based on using functional state data from the user acquired at the first time t1 and the second time t2 to identify any abnormality scores of the functional state from the functional state data, and if the identified abnormality scores are less than or equal to the set abnormality threshold, the action plan is valid. If the identified abnormality scores are greater than the set abnormality threshold, the increased FRP is correlated with the identified abnormality scores of the functional state and the action plan is updated.

[0023] According to some aspects, a computer-implemented method according to the disclosure described herein is disclosed, wherein the continuously acquired data includes data from fall detection.

[0024] According to some embodiments, the data is obtained from one or more sensor devices worn during the user's normal daily physical activities.

[0025] Fall detection based on continuous data and information constitutes one of the key inputs to manually created and / or generated action plans, including guidance, which includes activation and preventive measures.

[0026] According to some aspects, a computer-implemented method according to the present disclosure as described herein is disclosed, wherein a "supervisory machine learning module" enables said supervision of an action plan during said implementation of a training machine learning model, and wherein the supervision of the action plan is continuous supervision.

[0027] The computer-implemented method for user fall assessment includes said management of an action plan, where the management of the action plan is continuous management, e.g., continuous management, whereby the action plan can be updated as needed, typically periodically at a particular frequency.

[0028] According to some aspects, the management of the action plan is continuous and automatic. A computer-implemented method for user fall assessment includes said management of the action plan, where the management of the action plan is continuous and automatic, e.g., continuous and automatic, whereby the action plan can be updated as needed, typically periodically with a particular frequency.

[0029] According to some aspects, the management of the action plan is enabled by a “management machine learning module.” The present disclosure also relates to computer programs, control units, devices, systems, and parts of clothing associated with the above advantages.

[0030] It is therefore understood that the computer-implemented methods, computer programs, and computer-readable media according to the present disclosure, all as described herein, may be realized in hardware, such as the control unit arrangements and devices, all as described herein, and in the systems described herein. The hardware, such as the control unit arrangements and devices as described herein, and the systems as described herein, are then configured to execute the computer-implemented methods and computer programs, thereby obtaining the same advantages and benefits as described herein for the computer-implemented methods. [Brief description of the drawings]

[0031] The present disclosure will now be described in more detail with reference to the accompanying drawings. [Figure 1] FIG. 1 shows a schematic diagram of a user walking while wearing a sensor. [Diagram 2] FIG. 2 shows a schematic representation of a user riding a motorcycle and wearing a sensor. [Diagram 3] FIG. 3 illustrates generally a schematic diagram of the “Fall Risk Probability (FRP) and Action Plan Calculation” aspect, i.e., the calculation of FRP, and the generation of an action plan, both according to the present disclosure. [Figure 4] FIG. 4 illustrates a schematic diagram of the “Verification of FRP using 6MWT” and “Update or not update action plan” aspects according to the present disclosure. [Diagram 5] FIG. 5 shows a schematic representation of the control unit. [Figure 6] FIG. 6 illustrates an exemplary computer program product. [Figure 7] FIG. 7 shows a flow chart of a method according to the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0032] Aspects of the present disclosure are more fully described below with reference to the accompanying drawings. However, the different configurations, devices, systems, computer programs, and methods disclosed herein may be embodied in many different forms and should not be construed as being limited to the aspects set forth herein. Like numbers in the drawings refer to like elements throughout.

[0033] The terms used herein are for the purpose of describing aspects of the present disclosure only and are not intended to be limiting of the present invention. As used herein, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.

[0034] 1, there is a user 1 wearing at least one sensor device 2, 3, here one wrist sensor device 2 and one waist sensor device 3. In this case, the user 1 is walking.

[0035] 2, there is a user 1 wearing at least one sensor device, here one wrist sensor device 6 and one vest sensor device 7, which are provided in protective clothing such as a protective vest 10. In this case, the user 1 is riding a motorcycle 8.

[0036] The present disclosure is applicable to users walking or moving on or in any type of vehicle, such as a bicycle, a motorcycle 8, a car, etc.

[0037] The acquired data includes data acquired from sensors 2, 3, 6, 7 according to the present disclosure and may include data acquired from other sources. The acquired data from sensors 2, 3, 6, 7 is typically collected, i.e. acquired, from sensors included in, e.g. on-board sensors, e.g. motion sensors, e.g. 3d sensors such as accelerometers for measuring acceleration in three dimensions, as well as gyroscopes for measuring rotational speed in three dimensions from a person 1, i.e. a user, according to the present disclosure wearing such a wearable device. The wearable devices 2, 3, 6, 7 may suitably be edge devices.

[0038] The acquired data may include data acquired from sources other than sensors. The acquired data may refer to current and historical data. This may include geographic positioning of the user from a GPS system as well as weather information.

[0039] According to some embodiments, the user 1 wears a means 11 for providing and communicating the position of the user 1. Such means 11 may comprise any type of suitable positioning system, for example GPS or GNSS (Global Navigation Satellite System), and any type of wireless communication system. The means may be comprised in a vest 11 or in any suitable type of clothing or wearable device.

[0040] The data collected from sensors 2, 3, 6, 7 is used as input by, for example, a fall detection computer program, i.e., fall detection as described herein in accordance with the present disclosure, which is included in a machine learning model as described herein in accordance with the present disclosure, and, for example, a FRI detection computer program, i.e., FRI as described herein in accordance with the present disclosure, which is included in a "machine learning model" and "FRP machine learning module calculations" both as described herein in accordance with the present disclosure.

[0041] According to the present disclosure, and also with reference to Figure 7, there is a computer-implemented method for user fall assessment for a fall assessment environment. The computer-implemented method includes implementing a training machine learning model that includes an action plan (S100), continuously acquiring data (S200), and continuously processing the data.

[0042] The method further comprises a fall assessment for user 1 (S300) and validation of the action plan over time (S400), which is based on a continuous calculation of a fall risk probability FRP (S410) and on monitoring if and how the FRP changes (S420).

[0043] If a set condition for updating the action plan is met in the monitoring, the method includes a step S500 of updating the action plan. The fall assessment includes fall detection based on continuous data and continuous information, where the continuously acquired data includes data detected by sensors 2, 3, 6, 7 worn by the user 1 and positioning data of the user 1, and the calculation of the FRP also includes the positioning data.

[0044] This means that positioning data, particularly geo-positioning data, can be used to determine where a fall or near-fall occurred, indoors, outdoors, on stairs, on a slope, etc., and, given the appropriate time / date stamp, whether the fall or near-fall occurred in a generally cold and icy environment, or in a generally warm environment. In the case of the motorcyclist 1, the positioning data can be used to determine on what type of road the fall or near-fall occurred, and in what environment the fall occurred.

[0045] The present disclosure is applicable to users walking or moving on or in any type of vehicle, such as a bicycle, a motorcycle 8, a car, etc.

[0046] According to some embodiments, the calculation of the FRP also includes environmental conditions, including weather conditions, at the determined location of user 1.

[0047] This means that real-time environmental conditions, for example in the form of weather data, can be used in the calculation of the FRP.

[0048] According to some embodiments, the continuous calculation of the FRP is based on continuous data, continuous information, and fall risk indicators (FRI), which include at least one of "skidding", "slip", "loss of control", "collision and / or contact with another object or vehicle", "near-fall", "stumbling while walking", and / or "slip". Continuous information refers to real-time information, such as data collected from sensors 2, 3, 6, 7 worn by the user 1, as well as historical information about the geographic environment obtained from positioning data.

[0049] This means that the FRP is calculated using the FRI and real-time information, which provides an accurate measure of the FRP. The FRP calculation also takes into account non-real-time (offline) information, such as from historical / past information about the user and the current environment.

[0050] According to some embodiments, the continuously acquired data includes data from fall detection. According to some embodiments, the data is acquired from one or more sensor devices 2, 3, 6, 7 worn by the user 1 during normal daily physical activities.

[0051] According to some embodiments, the computer-implemented methods described herein include fall detection based on continuous data and continuous information, where the fall detection is detected using 3D sensor measurements and using a training ML model trained using training data collected from a user 1, such as a trainer or other wearer, such as a motorcycle and bicycle rider. Once a fall is detected using the computer program, it may be recorded and used in the input to the calculation of the FRP.

[0052] According to some aspects, as described herein, the action plan may be based on past actual falls and FRP, i.e., FRP of an individual person / user and FRP associated with geographic location. The fall risk probability (FRP) itself is calculated using fall risk indicators (FRIs) such as "skid", "near fall", "trip while walking", "slip" in a certain time period (and possibly other similar KPIs). The geo-positioned FRP (calculated by sharing geographic location and frequency of falls). It should be noted that the FRP impacts the creation / generation of the action plan. As an example, an increased FRP for a user may require the use of a more stringent actuation plan, such as a higher specification version of a portable airbag with more inflation volume as well as a higher inflation rate when triggered.

[0053] The acquired data, according to some aspects of the present disclosure, may include data acquired from sources other than sensors, such as historical data of the user's age, health profile, weather forecasts, user positioning information from a GPS (Global Positioning System), etc.

[0054] This means that the continuous calculation of FRP is based on continuous data, continuous information and FRI, where FRI includes at least one of "skidding", "slip", "loss of control", "collision and / or contact with another object or vehicle", "near fall", "stumbling while walking", and / or "slip", and continuous information refers to real-time information such as data collected from sensors worn by the user, as well as historical information about the geographical environment obtained from positioning data.

[0055] The fall assessment includes fall detection and calculation of the FRP, or fall detection and calculation of the FRP and the FRP associated with a geographic location.

[0056] The fall detection history data is one of the inputs for creating / generating a suitable action plan for the user.

[0057] In aspects of the present disclosure, as described herein, further inputs into the calculation of the appropriate action plan are calculated FRPs, such as FRPs, e.g., FRPs for humans, and geographic location-based FRPs. For example, FRPs can be calculated for geographic regions based on geographic locations.

[0058] As described herein, a computer-implemented method for user fall assessment includes the steps of continuously acquiring data, continuously processing the data, continuously generating information based on the acquired and processed data, and continuously communicating the information.

[0059] The computer-implemented methods described herein include continuously acquiring data, including data acquired from sensors and possibly other sources, and continuously processing the data, which allows for the calculation of the FRP and FRI, the FRP itself being calculated using the FRI.

[0060] The data obtained from the sensor is typically obtained from a sensor, e.g., a motion sensor, a 3D sensor, an accelerometer, a gyroscope and / or a camera, including, for example, a motion sensor, a 3D sensor, an accelerometer and / or a gyroscope. The sensor may be, for example, a wearable sensor, including a motion sensor, a 3D sensor, an accelerometer and / or a gyroscope. The sensor data may be obtained from a sensor, e.g., a wearable sensor, including a motion sensor, a 3D sensor, an accelerometer and / or a gyroscope. The wearable device may be equipped with a 3D sensor that is an accelerometer and used to measure acceleration in three dimensions, and the wearable device may be equipped with a 3D sensor that is a gyroscope and used to measure rotational velocity in three dimensions from a person wearing such a wearable device, i.e. a user according to the present disclosure. The wearable device may suitably be an edge device.

[0061] The computer-implemented method described herein involves continuous processing of data and FRIs are calculated over time using a training ML algorithm, i.e., a training ML computer program, trained on the training data to detect Fall Risk Indicators (FRIs) such as "skids," "near falls," "trips while walking," "slips" (and other KPIs) in specific time periods. These FRIs form the basis of input into the calculation of the FRP.

[0062] Further, the computer-implemented method includes calculation of information and data based on the acquired and processed data, such as detection and recording of falls, continuous creation of the FRI, and the FRP, which is ultimately used for creating / generating action plans for activation and preventive measures.

[0063] The computer-implemented method for user fall assessment described herein includes a fall assessment for a user, the fall assessment including fall detection based on continuous data and continuous information, and continuous calculation of a Fall Risk Index (FRI), and a FRP based on the continuous data, the continuous information.

[0064] According to some aspects, the computer-implemented method further includes guidance according to the action plan, the guidance including actuation and preventative measures, that correlates with and / or reflects a fall assessment for the user.

[0065] According to some embodiments, guidance includes activation and preventative measures, where activation refers to actions taken and / or performed according to a computer-implemented method for user fall assessment as described herein and according to an action plan that includes actions taken and / or performed by a suitable product to minimize any injuries after a fall incident has occurred (and when a so-called detent point has been reached).

[0066] 1 and 2, the actuation may include, for example, mechanical actuation of the portable airbags 4, 5, 9 (e.g., after detection of a fall event). The safety features of the portable airbags 4, 5, 9 and the characteristics of the airbags 4, 5, 9 such as volume, inflation capacity, inflation rate, etc. are based on action plans and guidance from the action plans. Guidance includes actuation and preventative measures and correlates to and / or reflects a fall assessment for the user, which includes fall detection and calculation of FRP for that particular user.

[0067] In aspects of the present disclosure, as described herein, guidance includes activation and preventative measures and correlates to and / or reflects a fall assessment for the user, which includes fall detection for that particular user and calculation of the FRP in conjunction with GPS geo-positioning.

[0068] Inputs into the manual creation of an action plan and / or generation of an action plan include inputs from a fall assessment including fall detection and calculation of FRP, and may also include inputs from a fall assessment including fall detection and calculation of FRP, for example in conjunction with GPS geographic positioning for a user currently using GPS.

[0069] Moreover, preventive measures are countermeasures to avoid a fall accident in the first place, before it occurs, including but not limited to actual warnings to the user(s) when a position is detected, such as visual, audible, and / or tactile warnings, such as a position that is in a dangerous position, such as close to a fall (but not yet reaching the detent point).

[0070] Additionally, preventative measures may also include other physical preventative measures, such as increased protective clothing and / or gear for the user(s) and / or may further include increased monitoring of the user. Additionally, these preventative measures may also be based on fall assessment, which may include fall detection and calculation of an FRP for a particular user, which may include calculation of the FRP in conjunction with GPS geographic positioning, as described herein, for example, in conjunction with geographic positioning of a user currently utilizing GPS.

[0071] According to some embodiments, the preventative measures include a warning signal and the protective measures include the inflation of one or more protective airbags 4, 5, 9 as shown in Figures 1 and 2. As shown in Figure 2, the sensors 6, 7, control unit 700, and / or protective airbag 9 may all be included in protective clothing such as a protective vest 10 for a person or user 1 riding a motorcycle 8 or the like.

[0072] Further inputs into the generation of new and / or updated action plans include inputs from fall assessment, i.e., inputs from detection of a fall for a particular user and calculation of the FRP, and optionally, inputs from calculation of the FRP in conjunction with geographic positioning of the user currently utilizing GPS.

[0073] In aspects of the present disclosure, as described herein, the action plan may be based on past actual falls and FRP, i.e., FRP of an individual person / user and FRP associated with geographic location. The fall risk probability (FRP) itself is calculated using fall risk indicators (FRIs) such as "skid", "near-fall", "trip while walking", "slip" in a particular time period (and possibly other similar KPIs). The geo-positioned FRP (calculated by sharing the geographic location, fall frequency, and information of the user's falls or near-fall accidents attached to that geographic location or area). It is noted that the FRP impacts the creation / generation of the action plan. As an example, an increased FRP for a user may require the use of a more stringent actuation plan, such as a higher specification version of a portable airbag with more inflation volume as well as a higher inflation rate when triggered.

[0074] Reference is now made to FIGS. 3 and 4 for the following description of examples according to some aspects of the present disclosure.

[0075] FIG. 3 illustrates how to compute the FRP, i.e., the computation of the FRP as described herein in accordance with certain aspects of the present disclosure, and the generation of an action plan, i.e., here including the generation of an action plan that reflects the training of a machine learning model as described herein in accordance with the present disclosure.

[0076] The generation, such as the generation of the action plan of the present disclosure, includes the calculation of plans for actuation and preventative measures, i.e., the generation of the action plan of the present disclosure as described herein. Data is streamed and collected, i.e., acquired, from sensors 2, 3, 6, 7, such as 3D sensors mounted on the wearable, and location information is acquired, such as from a GPS. See "Data Streaming from Wearable Device + GPS" in Figure 3.

[0077] The acquired data is processed and then fed to a trained fall detection algorithm 300, i.e., a trained fall detection computer program. The fall detection algorithm, i.e., the fall detection computer program, and fall detection according to the present disclosure, is trained using training data collected from a user(s), such as a trainer and other wearers, during a fall training session (see "Fall Detected or Not"). In 310 of FIG. 3, "Y" means that a fall was detected and "N" means that a fall was not detected (see FIG. 3).

[0078] The predictions and associated prediction probabilities from the fall detection algorithm may be stored and updated in an offline database, which forms one of the input information signals to the action plan calculation block, i.e., to the generation of the action plan of the present disclosure.

[0079] In parallel, there is another algorithm, i.e., another computer program, FRI detection algorithm 330, i.e., FRI detection computer program, which is the basis for the continuous calculation of the FRP of the present disclosure as described herein, also receives 3D sensor data, predicts and records / updates different FRIs as they occur in a database. See "FRI predicted or not". In 340 of FIG. 3, "Y" means predicted FRI and "N" means no predicted FRI (see FIG. 3). It should be noted that depending on the actual implementation, both the fall detection and FRI detection functions, i.e., the fall detection according to the present disclosure, and the FRI, which is the basis for the calculation of the FRP, may be provided by a single algorithm, i.e., a single computer program, or they may be implemented separately. However, this is just an implementation detail. The predicted FRI, both historical and more recent information from the database (see "FRI predicted or not" and "Y" in FIG. 3), along with historical information about user age, health profile, medical profile and environmental information such as weather (see "Historical Information (User Age, Health, Medical Profile), Environment" in FIG. 3), are provided to block 320, i.e., "Calculate FRP Machine Learning Module" 320 as described herein in accordance with the present disclosure, which calculates the FRP for both individual user(s) (see "Calculate / Update 'Human FRP'" 350 in FIG. 3) and, optionally, calculates a geographic positioning or region (see "Optionally Calculate / Update 'Geographical Location FRP'" 360 in FIG. 3).

[0080] Both types of FRP calculations take into account information from the detected FRI (current and past recorded in the database) and then calculate the FRP for the user and potentially for the geographic location. Finally, based on information about the detected fall and the calculated FRP, an action plan is generated, i.e., action plan generation, according to the present disclosure and as described herein, is customized for each user for preventive and actuation actions (see "Plan Update", "Calculated Plan for Actuation and Prevention Measures" 370, and "Plan for Prevention Measures / Actuation" in FIG. 3).

[0081] According to some embodiments, FIG. 4 shows one possible validation approach for the calculated (user) FRP, i.e., "validation is based on continuous calculation of FRP" as described herein (from "Verification of Action Plan Over Time" of the present disclosure). The calculated (user) FRP is shown above in FIG. 3. However, this is only an example, and a similar approach can be used to validate the change in FRP using other standard medical tests other than the 6-minute walk test (6MWT). The 6-minute walk test (6MWT) is a submaximal exercise test used to evaluate aerobic capacity and endurance as disclosed herein. The distance covered over a 6-minute period is used as an outcome to compare the change in performance capacity.

[0082] 4 illustrates that the FRP is calculated at different time intervals, i.e., T1 and then t2 (see "Calculate FRP T1" 400 and "Calculate FRP T2" 410 in FIG. 4), i.e., the calculation of FRPt1 at a first time t1 and the calculation of FRPt2 at a second time t2, as described herein. The FRP is calculated from data obtained from sensors, as described herein, see "Real-time data streaming from wearable time T1" and "Real-time data streaming from wearable time T2" ​​in FIG.

[0083] Then, the Δ difference of FRP was calculated (see “Δ:abs(FRP T2 -FRP T1 see “Calculate Δ FRP ), which is then compared to a predefined threshold 1 (see FIG. 4, “Δ>Thrsh1” 430, “Y” (yes) and “N” (no)), i.e., as described herein, in accordance with the present disclosure, Δ FRP is compared with the set FRP threshold.

[0084] According to some further aspects, for the same time periods t1 and t2, results from the 6MWT 440, 450 can also be calculated, i.e., based on also using functional status data from the user obtained at a first time t1 and a second time t2, in accordance with the present disclosure as described herein, an abnormality score 460 can be compared to identify (see FIG. 4 (see also "Abnormality Score > Thrsh 2" 470, "Y" (Yes) and "N" (No) in FIG. 4) (change in score between two different time instances), i.e., any abnormality score of functional status can be identified, in accordance with the present disclosure as described herein.

[0085] Furthermore, if both Δ are greater than their respective thresholds, i.e., Δ FRP If both the FRP and the "identified anomaly score" are greater than the set FRP threshold and the set anomaly threshold, respectively, a check 480 is made to ensure that both changes are positively correlated, as described herein. Then, if it is verified that the FRP has changed sufficiently, the action plan needs to be updated, otherwise the action plan does not need to be updated, i.e., included in the management of the action plan, according to the present disclosure and as described herein (see also "Δ" which has a high positive correlation with the "anomaly score"). In FIG. 2, "480", "Y" (Yes), "Verified (Update action plan)", "N" (No), "Not verified (Do not update action plan)".

[0086] Note that although this illustrates FRP validation for a user, the same concept can optionally be readily adopted to validate FRP for geographic locations or regions at different time intervals.

[0087] According to some aspects, a computer-implemented method includes managing an action plan, the successive calculation of the FRP further including calculating a first FRP, FRPt1, at a first time t1 and calculating a second FRP, FRPt2, at a second time t2 after the first time t1. The step of monitoring whether and how the FRP changes includes calculating a difference Δ between FRPt1 and FRPt2. FRP determining Δ FRP If is less than or equal to the set FRP threshold, the action plan is valid and Δ FRP If is greater than the set FRP threshold, update the action plan.

[0088] According to some aspects, a computer-implemented method according to the disclosure as described herein is disclosed, wherein a "supervisory machine learning module" enables supervision, e.g., continuous supervision, of an action plan during said implementation of a training machine learning model, the continuous computation of the FRP further comprising computing an FRPt1 at a first time t1 and computing an FRPt2 at a second time t2 after said first time t1.

[0089] The step of monitoring whether and how the FRP changes may include monitoring the difference between FRPt1 and FRPt2 (Δ FRP ), determining Δ FRP If is less than or equal to the set FRP threshold, the action plan is valid and Δ FRP If FRP is greater than the set FRP threshold, update the action plan (i.e., this means that the FRP has changed more than the set FRP threshold, and the action plan needs to be updated).

[0090] Furthermore, Δ FRPIt should be noted that if the action plan needs to be updated after the value is calculated, a further validation can be performed by calculating an abnormality score based on the results of a 6-minute walk test (6MWT) at the first time t1 and the second time t2. If a significant value of Δ correlates with a significant value of the abnormality score from the 6MWT, it is validated that the FRP has indeed changed significantly and the action plan needs to be updated at this point.

[0091] According to some aspects, the computer-implemented method includes managing an action plan. FRP If the FRP is greater than the set FRP threshold, the validation is also based on using functional state data from the user acquired at the first time t1 and the second time t2 to identify any abnormality scores of the functional state from the functional state data, and if the identified abnormality scores are less than or equal to the set abnormality threshold, the action plan is valid. If the identified abnormality scores are greater than the set abnormality threshold, the increased FRP is correlated with the identified abnormality scores of the functional state and the action plan is updated.

[0092] According to some aspects, a computer-implemented method according to the present disclosure as described herein is disclosed, wherein a "supervisory machine learning module" enables supervision, e.g., continuous supervision, of an action plan during said implementation of a training machine learning model. FRP If the FRP is greater than the set FRP threshold, the validation is also based on using functional state data from the user acquired at the first time t1 and the second time t2 to identify any abnormality scores of the functional state from the functional state data, and if the identified abnormality scores are less than or equal to the set abnormality threshold, the action plan is valid. If the identified abnormality scores are greater than the set abnormality threshold, the increased FRP is correlated with the identified abnormality scores of the functional state and the action plan is updated.

[0093] According to some aspects, a computer-implemented method according to the present disclosure is disclosed, where the continuously acquired data includes positioning data, e.g., geographic positioning data, as described herein. The calculation of the FRP also includes positioning data, e.g., geographic positioning data. The FRP may also be calculated for geographic positioning using the FRI, as described herein, with an additional tag indicating the geographic location of where the FRI occurred (e.g., using GPS). This, in turn, allows for the calculation of the FRP for a geographic location or geographic region.

[0094] Furthermore, it may be required to have a mechanism for sharing relevant information / data among entities in a collaborative manner so that such information can be developed in an efficient manner. One way to do this is that each of the entities may "opt-in" to share their FRI relevant information / data to a central entity, which then calculates the FRP associated with each "geographic location" / "geographic area" and then shares this information to all entities in the collaborative system, i.e., all entities that have "opted-in".

[0095] According to some aspects, the management of the action plan is a continuous management, and the computer-implemented method for user fall assessment includes said management of the action plan, and the management of the action plan is a continuous management.

[0096] For all embodiments disclosed herein, it should be understood that Figure 6 discloses a general representation of a computer program 810 including computer readable instructions 820 on a computer readable medium 830. The computer program may be considered, according to some aspects, a computer program product.

[0097] The present disclosure also relates to a computer program 810 comprising computer readable instructions 820 for applying the computer-implemented methods described herein, and / or a computer readable medium 830 comprising said computer program 810.

[0098] The present disclosure also relates to a computer program 810 including computer readable instructions 820 for training machine learning models as described herein, and / or a computer readable medium 830 including the computer program 810.

[0099] The present disclosure also relates to a computer program 810 including computer readable instructions 820 for a "supervised machine learning module" as described herein, and / or a computer readable medium 830 including said computer program 810.

[0100] 5, the present disclosure also relates to a control unit arrangement 700 for user fall assessment, adapted to at least control the implementation of a training machine learning model as described herein, the guidance following an action plan as described herein, and / or the activation of the management of the action plan. The control unit arrangement 700 is hereinafter referred to as the control unit 700.

[0101] FIG. 5 shows diagrammatically the components of a control unit 700 according to one embodiment with respect to some functional units.

[0102] The processing circuitry 710 may be provided using one or more combinations of suitable central processing units (CPUs), multiprocessors, microcontrollers, digital signal processors (DSPs), dedicated hardware accelerators, etc. capable of executing software instructions stored, for example, in a computer program product in the form of a storage medium 730. The processing circuitry 710 may further be provided as at least one application specific integrated circuit (ASIC) or field programmable gate array (FPGA).

[0103] In particular, the processing circuitry 710 is configured to cause the control unit 700 to perform a set of operations or steps, which have been described above in connection with various radar transceivers and methods. For example, the storage medium 730 may store a set of operations, and the processing circuitry 710 may be configured to retrieve the set of operations from the storage medium 730 and cause the control unit 700 to perform the set of operations. The set of operations may be provided as a set of executable instructions. Thus, the processing circuitry 710 is configured to perform the methods and operations disclosed herein.

[0104] The storage medium 730 may also comprise persistent storage, which may be, for example, any one or combination of magnetic memory, optical memory, solid-state memory, or even remotely mounted memory.

[0105] The control unit 700 may further comprise a communication interface 720 for communicating with at least one other unit. The communication interface 720 may thus comprise one or more transmitters and receivers with analog and digital components and a suitable number of ports for wired or wireless communication.

[0106] The processing circuitry 710 is adapted to control the general operation of the control unit 700, for example, by sending data and control signals to an external unit and storage medium 730, by receiving data and reports from an external unit, and by retrieving data and instructions from the storage medium 730. Other components and associated functions of the control unit 700 have been omitted so as not to obscure the concepts presented herein.

[0107] The present disclosure also relates to a device, e.g., a wearable device, for enabling implementation of a training machine learning model, as described herein, for communication of information, for user fall assessment, for guidance according to an action plan, and for enabling management of the action plan, guidance including activation and preventive measures, and fall assessment including fall detection and calculation of FRP based on continuously acquired data and FRI, FRI including FRIs such as "skid", "slip", "loss of control", "collision and / or contact with another object or vehicle", "near fall", "stumbling while walking", and / or "slip".

[0108] The device includes a trained machine learning model trained on a user fall assessment and including an action plan, a sensor for detecting data, a means for continuously acquiring data, a means for processing data 700, a means for generating information based on the acquired and processed data, and a means for communicating information. The device comprises a computer program 810 including computer readable instructions 820 according to above, and / or a computer readable medium 830 according to above.

[0109] According to some embodiments, the device is at least partially in the form of a vest 10 or similar garment. The device may also be comprised of a control unit arrangement 700 and / or a bracelet or the like. In Fig. 1, a user 1 is shown listening to a sensor 2 contained in a bracelet. In general, the present disclosure relates to a piece 15 of garment 10 comprising a control unit arrangement 700 as described herein, one or more sensors 2, 3, 6, 7 for detecting data as described herein, and at least one airbag 4, 5, 9.

[0110] According to some aspects, a computer program 810 is used to apply a computer-implemented method as described herein, and includes a computer-readable medium containing the computer program, for training machine learning models as described herein, and / or for a "supervised machine learning module" as described herein.

[0111] The present disclosure also relates to a system for enabling implementation of training machine learning models, the training machine learning models as described herein, for communication of information, for user fall assessment, for guidance according to an action plan, and for enabling management of the action plan, the action plan as described herein, the guidance including activation and preventative measures. Fall assessment includes fall detection and calculation of an FRP based on continuously acquired data and an FRI, the FRI including FRIs such as "skid", "slip", "loss of control", "collision and / or contact with another object or vehicle", "near fall", "stumbling while walking", and / or "slip".

[0112] The system includes a trained machine learning model trained on a user fall assessment and including an action plan, a sensor for detecting data, a means for continuously acquiring data, a means for processing the data, a means for generating information based on the acquired and processed data, and a means for communicating the information. The system includes a computer program(s) 810 including computer readable instructions 820 for applying the computer-implemented methods described herein, and / or a computer readable medium 830 as described herein.

[0113] According to some aspects, a computer program 810 is used to train machine learning models as described herein and / or to apply a "supervised machine learning module" as described herein, and a computer-readable medium comprises the computer program.

[0114] According to some aspects, a "supervisory machine learning module" enables said supervision of an action plan during said implementation of a training machine learning model, wherein the supervision of the action plan is continuous supervision, and a computer-implemented method for user fall assessment includes said supervision of an action plan, wherein the supervision of the action plan is continuous supervision.

[0115] According to some aspects, the fall assessment environment includes a training machine learning model that is trained on a user fall assessment, a user, a sensor that detects data from the fall assessment environment, means for continuously acquiring data, means for processing the data, means for generating information based on the acquired and processed data, means for communicating the information, and means for providing guidance based on the ongoing data and based on the information according to an action plan.

[0116] According to some embodiments, the trained machine learning model, during its implementation in the computer-implemented method for user fall assessment, continuously enables continuous acquisition of data and continuous processing of data, continuous production and continuous communication of information, fall assessment of the user including fall detection and continuous calculation of FRP, and guidance to follow an action plan.

[0117] In aspects of the present disclosure, as described herein, generating the action plan includes manual creation of the action plan that specifies guidance that correlates with a fall assessment for the user, and / or generating the action plan includes generating an action plan that reflects training of a machine learning model, where the guidance reflects a fall assessment for the user. Alternatively, creating / updating the action plan may be automatic.

[0118] Further, as described herein, in accordance with aspects of the computer-implemented method for user fall assessment, the fall assessment includes fall detection and calculation of a fall risk probability (FRP).

[0119] The continuous calculation of the FRP may be, for example, an automatic continuous calculation of the FRP.

[0120] According to some embodiments, FRI over time is detected and recorded for each user using a trained ML model. This may form the basis for the calculation of the FRP. Note that for the calculation of the FRP from the FRI, the individual FRIs may also be weighted, such as giving a higher weight to FRIs that are "skids" or "trips while walking" compared to "slips". A "skid" may be due to road and environmental conditions, and a "trips while walking" may indicate other medical issues. A "skid" is due to road and environmental conditions of the motorcycle 8.

[0121] According to some further aspects, the FRI may include FRIs such as "skid," "slip," "loss of control," "collision and / or contact with another object or vehicle," "near fall," "trip while walking," and / or "slip," and may optionally include other suitable key performance indicators (KPIs) for the FRI.

[0122] Additionally, it should be noted that the FRIs may not be limited to those listed above and may optionally include other KPIs as well depending on the particular use case and scenario.

[0123] The continuously acquired data and the continuously processed data, according to some aspects, comprise data detected by the sensors, FRI, and data related to the FRI, and may optionally also include other data, for example from weather forecasts from the Internet.

[0124] Further, according to some aspects, there is a fall assessment environment, comprising a trained machine learning model trained on a user fall assessment, a user, a sensor for detecting data from the fall assessment environment, means for continuously acquiring data, means for processing the data, means for generating information based on the acquired data and the processed data, means for communicating the information, and means for guiding according to an action plan based on the ongoing data and information.

[0125] The training machine learning model is trained on user fall assessment, where the fall assessment includes fall detection, and where the calculation of the FRP includes detection of the user's FRI. That is, the training may be considered to include several types of ML models and ML model training, according to some aspects. For example, a model is trained for fall detection, and a second model is trained for calculation of the FRP including detection of the user's FRI.

[0126] According to some aspects, the sensors 2, 3, 6, 7 for detecting data include sensors for detecting data about the user, e.g., from sensors on or at the user, and optionally further sensors for detecting data in or about the environment. Further, the sensors include, e.g., a fall sensor, a motion sensor, and a location sensor, e.g., a Global Positioning System (GPS), and optionally further sensors in the environment. In further embodiments, the sensors include, e.g., a fall sensor and a motion sensor.

[0127] The means for continuously acquiring data may be any means suitable for continuously acquiring data, according to some aspects. The acquired data includes "data detected by the sensor", FRI, and data related to the FRI.

[0128] Further, according to some aspects, the acquired data includes "sensor detected data", FRI and data related to the FRI, and optionally other data, for example from a weather forecast from the Internet.

[0129] The means for generating information based on the obtained data and the processed data may, according to some aspects, comprise any means suitable for generating information.

[0130] The means for communicating information may, according to some aspects, comprise any means suitable for communicating information.

[0131] Furthermore, the means for guiding based on ongoing data and information according to an action plan may, according to some aspects, include any means suitable for guidance.

[0132] According to some aspects, the training machine learning model, during its implementation in a computer-implemented method for user fall assessment as described herein, comprises: Continuous acquisition of data and continuous processing of data, The continuous creation and communication of information; Continuously enabling fall assessment for a user, the fall assessment including fall detection and continuous calculation of FRP; Includes guidance based on an action plan.

[0133] According to some embodiments, the computer-implemented methods for user fall assessment described herein further include the training machine learning model including a "computation of a FRP machine learning module," which enables continuous computation of the FRP during said implementation of the training machine learning model.

[0134] According to some aspects, the trained machine learning model further includes a “supervisory machine learning module” that enables management, e.g., continuous management, of the action plan during said implementation of the trained machine learning model.

[0135] For example, management may include validation of the action plan over time, where validation is based on continuous calculation of the FRP, monitoring whether and how the FRP changes, and updating the action plan if set conditions for updating the action plan are met upon monitoring.

[0136] According to some embodiments, the computer-implemented method for user fall assessment further includes managing, e.g., continuously managing, the action plan. For example, the managing includes verifying the action plan over time. The verifying is based on continuously calculating the FRP, monitoring whether and how the FRP changes, and updating the action plan if set conditions for updating the action plan are met in the monitoring.

[0137] According to some embodiments, the "Compute FRP Machine Learning Module" enables continuous computation of the FRP during said implementation of the training machine learning model.

[0138] According to some embodiments, a "management machine learning module" enables management, e.g., continuous management, of the action plan during said implementation of the training machine learning model, where the management includes validation of the action plan over time, where the validation is based on continuous calculation of the FRP, monitoring whether and how the FRP changes, and updating the action plan if set conditions for updating the action plan are met upon monitoring.

[0139] Validation and subsequent calculation of the FRP may, according to some aspects, be further based on positioning data.

[0140] The validation and subsequent calculation of the FRP may be further based on positioning data and / or fall detection data, according to some embodiments.

[0141] According to some aspects, the computer-implemented method for user fall assessment described herein provides a customized action plan including activation and preventative measures, where the action plan is based on each user's needs, where the needs are based on the user's current and past circumstances of the detected fall and the fall risk probability using the detected fall risk indicator. This has the effect of enabling improved action plans, including enhanced protection for users who are more vulnerable to falls, based on the estimations of the computer-implemented method for user fall assessment as described herein. The estimations of the computer-implemented method for user fall assessment may also be updated over time.

[0142] Furthermore, the improved action plan, according to some aspects, also allows for tailoring of protection plans for users who are less prone to falls based on the current estimation of the computer-implemented method for user fall assessment. In essence, the action plan for each user can be optimized and updated over time according to changing needs rather than "one solution fits all". Furthermore, it should be noted that the FRP may also be associated with the user as well as the geographical positioning, and thus the geographical positioning may also affect the action plan generated for each user. Furthermore, the action plan may also be dynamically updated over time with a certain frequency. From another perspective, the proposed computer-implemented method for user fall assessment, as described herein, allows for improved use and allocation of limited physical resources (e.g., protection equipment and / or fall and / or hip protection equipment) and human resources (close observation and monitoring of individuals) in a particular facility, such as an adult care facility, based on the needs of each needy user, e.g., the needs of the person / patient.

[0143] According to some embodiments, management, eg, continuous management, of the action plan includes validation of the action plan over time, the validation being based on continuous calculation of the FRP.

[0144] According to some embodiments, the management, eg, the continuous management, includes a verification of the action plan over time, the verification being based on a continuous calculation of the FRP.

[0145] The management, eg, continuous management, includes validation of the action plan over time, and the validation and continuous calculation of the FRP may be further based on positioning data, according to some aspects.

[0146] According to some aspects, the management, eg, continuous management, includes validation of the action plan over time, and the validation and continuous calculation of the FRP may be further based on positioning data and / or fall detection data.

[0147] According to some aspects, the computer-implemented method for user fall assessment as described herein further includes monitoring whether and how the FRP changes, and updating the action plan if set conditions for updating the action plan are met in the monitoring.

[0148] According to some aspects, the present disclosure relates to a computer-implemented method for user fall assessment for a fall assessment environment, the computer-implemented method including implementing a training machine learning model that includes an action plan and guiding according to the action plan.

[0149] According to some aspects, the guidance includes activation and preventative measures and correlates to and / or reflects a fall assessment for the user, the fall assessment including fall detection and calculation of a fall risk probability (FRP). The computer-implemented method includes continuously acquiring data and continuously processing the data, the computer-implemented method includes continuously generating information based on the acquired and processed data, continuously communicating the information, and a fall assessment for the user. The fall assessment includes fall detection based on the continuous data and the continuous information, and It includes continuous calculation of FRP based on continuous data, Fall Risk Index (FRI) and continuous information.

[0150] According to some aspects, the FRI includes FRIs such as "skid," "slip," "loss of control," "collision and / or contact with another object or vehicle," "near-fall," "stumbling while walking," and / or "slip." The fall assessment environment includes a training machine learning model trained on a user fall assessment, a user, a sensor that detects data from the fall assessment environment, a means for continuously acquiring data, a means for processing the data, a means for generating information based on the acquired and processed data, a means for communicating the information, and a means for guiding based on the ongoing data and information according to an action plan.

[0151] According to some embodiments, the trained machine learning model, during its implementation in the computer-implemented method for user fall assessment, continuously enables continuous acquisition of data and continuous processing of data, continuous production and continuous communication of information, fall assessment of the user including fall detection and continuous calculation of FRP, and guidance to follow an action plan.

[0152] According to some embodiments, the training machine learning model includes a "computation of a FRP machine learning module" that enables continuous computation of the FRP during said implementation of the training machine learning model, and the training machine learning model further includes a "supervisory machine learning module" that enables management of action plans during said implementation of the training machine learning model.

[0153] According to some embodiments, management includes verification of the action plan over time, where the verification is based on continuous calculation of the FRP, monitoring whether and how the FRP changes, and updating the action plan if set conditions for updating the action plan are met in the monitoring.

[0154] According to some embodiments, the computer-implemented method for user fall assessment further includes managing the action plan, where the managing includes verifying the action plan over time, where the verifying is based on continuous calculation of the FRP, monitoring whether and how the FRP changes, and updating the action plan if set conditions for updating the action plan are met in the monitoring.

Claims

1. 1. A computer-implemented method for user fall assessment for a fall assessment environment, comprising: Implementing a training machine learning model that includes an action plan (S100); A step of continuously acquiring data (S200); and a step of continuously processing the data. A fall assessment (S300) for a user (1); A time-dependent verification (S400) of the action plan, Continuous calculation of the fall risk probability FRP (S410); Verification (S400), based on a step of monitoring (S420) whether and how the FRP changes; When a set condition for updating the action plan is met during the monitoring, updating the action plan (S500); and wherein the fall assessment includes fall detection based on continuous data and continuous information; 11. A computer-implemented method, comprising: a) providing a plurality of sensors (2, 3, 6, 7) mounted on a user (1) for detecting a location of the user (1) and a step of calculating the FRP based on the location of the user (1);

2. The computer-implemented method of claim 1 , wherein the calculation of the FRP also includes environmental conditions, including weather conditions, at the determined location of the user (1).

3. 3. The computer-implemented method of claim 1 or 2, wherein the continuous calculation of the FRP is based on the continuous data, continuous information, and a Fall Risk Index (FRI), the FRI including at least one of "skidding", "slip", "loss of control", "collision and / or contact with another object or vehicle", "near fall", "stumbling while walking", and / or "slip", the continuous information referring to real-time information such as data collected from sensors (2, 3, 6, 7) worn by the user (1) as well as historical information regarding the geographical environment obtained from the positioning data.

4. The computer-implemented method of claim 3 , wherein the continuing information refers to information regarding a user's age, physical and health profile.

5. 5. The computer-implemented method of claim 1, further comprising guiding the user according to the action plan, the guidance including activation and preventative measures, and correlating with and / or reflecting a fall assessment for the user.

6. the computer-implemented method includes managing the action plan; The continuous calculation of the FRP is performed at a first time t 1 The first FRP in 1 and calculating the first time t 1 A second time t 2 The second FRP in 2 and further comprising: The step of monitoring whether and how the FRP changes comprises: 1 and FRPt 2 The difference Δ between FRP determining Δ FRP is equal to or less than the set FRP threshold, the action plan is valid; Δ FRP The computer-implemented method of claim 1 , further comprising updating the action plan if FRP is greater than a set FRP threshold.

7. the computer-implemented method includes managing the action plan; In the monitoring, Δ FRP is greater than the set FRP threshold, the verification also includes the first time t 1 and the second time t 2 based on using the functional status data from the user (1) acquired in, identifying any abnormality score of the functional status from the functional status data, and if the identified abnormality score is equal to or less than a set abnormality threshold, the action plan is valid; 7. The computer-implemented method of claim 6, further comprising: correlating the increased FRP with the identified abnormality score of the functional state and updating the action plan if the identified abnormality score is greater than the established abnormality threshold.

8. The computer-implemented method of any preceding claim, wherein the continuously acquired data includes data from the fall detection.

9. 9. The computer-implemented method of claim 1, wherein the management of the action plan is continuous and automatic, and the computer-implemented method for user fall assessment comprises managing the action plan, and wherein the management of the action plan is continuous.

10. A computer program (810) comprising computer readable instructions (820) for applying the computer implemented method according to any one of claims 1 to 9 and / or a computer readable medium (830) comprising said computer program (810).

11. A computer program (810) comprising computer readable instructions (820) for implementing a training machine learning model according to any one of claims 1 to 9, and / or a computer readable medium (830) comprising said computer program (810).

12. A control unit arrangement (700) for user fall assessment adapted to control at least the implementation of a trained machine learning model according to any one of claims 1 to 9, the guidance according to an action plan according to any one of claims 1 to 9, and / or the activation of the management of said action plan.

13. A device, such as a wearable device, for a user fall assessment, for communication of information, for enabling the implementation of a trained machine learning model according to any one of claims 1 to 9, and for guidance according to an action plan according to any one of claims 1 to 9 and for enabling the management of an action plan according to any one of claims 1 to 9, the guidance includes operational and preventative measures; the fall assessment includes fall detection and calculation of a FRP based on continuously acquired data and FRI, the FRI including FRIs such as "skid", "slip", "loss of control", "collision and / or contact with another object or vehicle", "near fall", "tripping while walking", and / or "slip"; The device, the trained machine learning model is trained on the user fall assessment and includes an action plan; A sensor (2, 3, 6, 87) for detecting data; A means (700) for continuously acquiring data; A means (700) for processing data; means (700) for generating said information based on the acquired and processed data; means (11) for communicating said information, A device, said device comprising a computer program (810) according to one or more of claims 10 and 11 and / or a computer readable medium (830) according to one or more of claims 10 and 11.

14. A system for user fall assessment enabling the implementation of a trained machine learning model according to any one of claims 1 to 9 for communication of information and enabling guidance and management according to an action plan according to any one of claims 1 to 9, the guidance includes operational and preventative measures; the fall assessment includes fall detection and calculation of a FRP based on continuously acquired data and FRI, the FRI including FRIs such as "skid", "slip", "loss of control", "collision and / or contact with another object or vehicle", "near fall", "tripping while walking", and / or "slip"; The system further comprises: a trained machine learning model trained on the user fall assessment and including an action plan; Sensors (2, 3, 6, 7) for detecting data; A means (700) for continuously acquiring data; A means (700) for processing data; means (700) for generating said information based on the acquired and processed data; means (11) for communicating said information, The system comprises a computer program (810) according to one or more of claims 10 and 11 and / or a computer readable medium (830) according to one or more of claims 10 and 11.

15. A piece of clothing (10) comprising a control unit arrangement (700) according to claim 12, one or more sensors (2, 3, 6, 7) for detecting data, and at least one airbag (4, 5, 9).