WO2017142488A1

WO2017142488A1 - System for the prediction and prevention of patient falls

Info

Publication number: WO2017142488A1
Application number: PCT/SG2017/050077
Authority: WO
Inventors: Kar Kit Bernard LOKE
Original assignee: Loke Kar Kit Bernard
Priority date: 2016-02-17
Filing date: 2017-02-17
Publication date: 2017-08-24

Abstract

A primary embodiment of the invention is a system for prediction and prevention of falls comprising: (a) a plurality of wearable motion detection devices, each wearable motion detection device attachable to at least one of a lower limb and a trunk of a patient and each wearable motion detection device including a band 2 and an electronics module 3; (b) at least one local gateway 8 configured to connect wirelessly to each wearable motion detection device and to receive each data log from wearable motion detection device; (c) at least one server 9, each server 9 including an alarm management module 10, a web service and a database 12; and (d) at least one workstation, each workstation including a user interface configured to receive alert events from the web service of the at least one server 9. The electronics module 3 comprises: (i) a plurality of micro-electro-mechanical sensors 4; (ii) a wireless transceiver 5; (iii) a non-volatile memory 6; and (iv) a microprocessor 7 configured to assemble at least one data log, wherein each data log comprises a patient ID and a plurality of time-stamped gesture indications. Each server 9 is configured to connect to the local gateway 8 and receive each data log from the plurality of wearable motion detection devices. The alarm management module 10 of each server 9 is configured to process each data log to determine an alert event from the plurality of time-stamped gesture indications and the patient ID in the data log.

Description

SYSTEM FOR THE PREDICTION AND PREVENTION OF PATIENT FALLS

TECHNICAL FIELD

The present disclosure relates to a monitoring of the movements of patients at risk for a fall. More particularly, the present disclosure relates to a system for prediction and prevention of patient falls. BACKGROUND

Elderly people are at a high risk for falling, which can lead to hip fractures or other injuries. The risk of fall is increased because the elderly have have poor eye sight, coordination, balance and/or physical strength. When the elderly fall, they tend to have a higher frequency of fractures due to osteoporosis and weaker bones. Breaking a hip can lead to a significant loss of independence, physical fitness and a decline in health. Falls and subsequent injuries can also lead to a greater probability of ending up in long- term care. A large proportion of fall deaths are due to complications following a hip fracture.

The extent of the danger presented by falls can be found in "Survey on Fall Detection and Fall Prevention Using Wearable and External Sensors," Yueng Santiago Delahoz and Miguel Angel Labrador, Sensors, 14:19806-19842, October 2014, doi: 10.3390/sl41019806. This publication discusses how falling represents a great threat as people get older, and states that mechanisms to detect and prevent falls are critical to improve people's lives. The publication surveys the state of the art in fall detection and fall prevention and discusses the different types of sensors used in both approaches. A similar study, relating to fall monitoring, is "Automatic Fall Monitoring: A Review," Natthapon Pannurat, Surapa Thiemjarus and Ekawit Nantajeewarawat, Sensors, 14(7): 12900- 12936, July 2014, doi: 10.3390/s 140712900. This study reviews the existing fall detection systems and some of the key research challenges faced by the research community in this field. The study discusses wearable and ambient devices.

One computerized solution for detecting and preventing falls is disclosed in PCT Publication No. WO/2016/003365. This publication discloses a wearable input device for communication with a computer. The wearable input device comprises a body formed for wearing on a hand of a user, a plurality of sensors for detecting gestures made by the user with at least a portion of the hand, and a microprocessor communicatively coupled to the plurality of sensors. The microprocessor is further for transmitting input signals to the computer in response to the detection of the gestures. Each gesture is associated with one of the input signals, and each of the input signals is further associated with one from a plurality of functions performable on the computer. The microprocessor is configured for automatically deactivating the transmission of at least some of the input signals to the computer in response to the microprocessor receiving a deactivation signal. Systems for prediction and prevention of falls therefore is a significant value for hospitals taking care of elderly patients or other patients at risk of a fall.

SUMMARY

A primary embodiment of the invention is a system for prediction and prevention of falls comprising: (a) a plurality of wearable motion detection devices, each wearable motion detection device attachable to at least one of a lower limb and a trunk of a patient and each wearable motion detection device including a band and an electronics module; (b) at least one local gateway configured to connect wirelessly to each wearable motion detection device and to receive each data log from the wearable motion detection device; (c) at least one server, each server including an alarm management module, a web service and a database; and (d) at least one workstation, each workstation including a user interface configured to receive alert events from the web service of the at least one server. The electronics module comprises: (i) a plurality of micro-electro-mechanical sensors; (ii) a wireless transceiver; (iii) a nonvolatile memory; and (iv) a microprocessor configured to assemble at least one data log, wherein each data log comprises a patient ID and a plurality of time-stamped gesture indications. Each server is configured to connect to the local gateway and receive each data log from the plurality of wearable motion detection devices. The alarm management module of each server is configured to process each data log to determine an alert event from the plurality of time- stamped gesture indications and the patient ID in the data log.

In an optional design of the primary embodiment, the at least one alert event received by the workstation can optionally includes the patient ID, a description of the alert event and a severity level. Optionally the server determines a longer horizon fall prediction for each patient ID by evaluation of a history period of the indication of gesture of the patient ID including at least one of a gait analysis, a rest period analysis, a daily routine analysis and a medication regime and the history period is at least one day and can includes weeks or months. The longer the history period typically the more accurate the longer horizon fall prediction. In a second embodiment of the invention, the design additionally includes, beyond the elements of the primary embodiment, an alarm triggered by each alert event with a severity level exceeding a sensitivity threshold of the patient ID. In an optional design of the second embodiment, the sensitivity threshold for each patient ID server can optionally be configurable by at least one of: (a) a patient mobility level assessment; (b) an activity level assessment; (c) a prevailing medical condition; (d) an end-user data input to the workstation; and (e) an event calendar that tracks at least one of a visitation schedule, a feeding schedule and a staff shift schedule. Optionally, an audio cue by the workstation can be sounded by the alarm. Optionally, the system can further comprises a speaker proximate to the patient capable of playing an audio cue sounded by the alarm. Optionally, the system further comprises an alarm management system alerted by the alarm. Optionally, the alarm can be a short horizon fall prediction, wherein the short horizon fall prediction: (a) provides at least 5 seconds, preferably at least 20 seconds, advance notice of a potential fall hazard; and (b) is determined by a recent abnormal movement of the patient. Optionally, the a confirmation prompt can be provided on the workstation for an end-user of the workstation to respond to each alarm with an alarm confirmation and each alarm confirmation can be stored in a confirmation log on the server.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the present disclosure are described herein with reference to the drawings in which:

FIG. 1A is a wearable motion detection device of an embodiment of the invention with a band and an electronics module. FIG. IB is the electronics module of an embodiment of the invention.

FIG. 2 is a diagram of a system for an embodiment of the invention.

FIG. 3 is a flowchart representing the analysis and use of motion information from the wearable motion detection device by the system in an embodiment of the invention. DETAILED DESCRIPTION

In the following detailed description, reference is made to the accompanying drawings, which form a part hereof. The illustrative embodiments described in the detailed description, drawings and claims are not meant to be limiting. Other embodiments can be utilized, and other changes can be made, without departing from the spirit or scope of the subject matter presented herein. Unless specified otherwise, the terms "comprising," "comprise," "including" and "include" used herein, and grammatical variants thereof, are intended to represent "open" or "inclusive" language such that they include recited elements but also permit inclusion of additional, un-recited elements. The term "attachable," "connect," "connected," or "connecting" used herein, and grammatical variants thereof, are intended to represent a linking of two items either by electronic communication or physically together, directly or indirectly.

Components in FIGs. 1A-2 are numbered as listed in Table 1 below.

Description of Components FIG. 1A is a wearable motion detection device 1 of an embodiment of the invention with an electronics module 3 mounted on a band 2. The band 2 is attachable to a lower limb or a trunk of a patient. Details of the electronics module 3 are shown in FIG. IB. The electronics module 3 includes: a micro-electro-mechanical sensors 4, a wireless transceiver 5, a non- volatile memory 6 and a microprocessor 7.

FIG. 2 is a diagram of a system 14 for an embodiment of the invention. From top to bottom in FIG. 2, the system 14 includes a plurality of wearable motion detection devices 1 connected wirelessly to a local gateway 8. The local gateway 8 is connected through a network to a server 9. The server 9 includes an alarm management module 10, a web service 11 and a database 12. The server 9 (with the web service 11) is attached through a network to the workstation 13. The workstation 13 has a user interface. The server 9 can be remote, as in a cloud server, or in the same building as the workstation 9.

FIG. 3 is a flowchart 3-00 representing the analysis and use of motion information from the wearable motion detection device 1 by the system 14 in an embodiment of the invention. In the first step of this flowchart 3-01, within wearable motion detection device 1, there is a detection of patient movement, creation of a data log of time- stamped gesture indications and transmission of a data log to local gateway 8 wirelessly. In the second step of this flowchart 3-02, the data log is received by the local gateway, the data logs are consolidated and the consolidated data logs are transmitted to the server 9. In the third step of this flowchart 3-03, the data logs are received from the local gateway 8, the data logs are processed to determine alert events and the web service 11 is populated for access by the workstation 13. In the fourth step of this flowchart 3-04, the the web service 11 is accessed from the workstation 13 to access alert events, audio cue are sounded as needed, sensitivity thresholds are set for patients by an end-user of the workstation 13 and the end-user can view patient history logs on the workstation 13. The information on the server 9 is accessible through the web service 11 by the end-user at the workstation 13. A primary embodiment of the invention is a system for prediction and prevention of falls comprising: (a) a plurality of wearable motion detection devices, each wearable motion detection device attachable to at least one of a lower limb and a trunk of a patient and each wearable motion detection device including a band 2 and an electronics module 3; (b) at least one local gateway 8 configured to connect wirelessly to each wearable motion detection device and to receive each data log from wearable motion detection device; (c) at least one server 9, each server 9 including an alarm management module 10, a web service and a database 12; and (d) at least one workstation, each workstation including a user interface configured to receive alert events from the web service of the at least one server 9. The electronics module 3 comprises: (i) a plurality of micro-electro-mechanical sensors 4; (ii) a wireless transceiver 5; (iii) a non-volatile memory 6; and (iv) a microprocessor 7 configured to assemble at least one data log, wherein each data log comprises a patient ID and a plurality of time-stamped gesture indications. Each server 9 is configured to connect to the local gateway 8 and receive each data log from the plurality of wearable motion detection devices. The alarm management module 10 of each server 9 is configured to process each data log to determine an alert event from the plurality of time-stamped gesture indications and the patient ID in the data log. The wearable motion detection device 1 can be worn anywhere on the lower limb or trunk of a patient. For instance, the wearable motion detection device 1 can be worn on the ankle, above the knee, on the waist or the chest of a patient. Each patient can have one or more of the wearable motion detection device 1 attached on their body. The local gateway 8 of the system 14 can be wirelessly connected to one or more wearable motion detection devices 1 attached to a single patient or multiple patients. For instance, a hospital can use the wearable motion detection device 1 on multiple patients in the hospital for monitoring the motion of each patient.

The micro-electro-mechanical sensors 4 can include at least one solid state device and/or microscopic devices with moving parts. The micro-electro-mechanical sensors 4 can include accelerometers and/or gyroscopes. The wireless transceiver 5 can be a component designed for wireless information exchange under a standardized frequency and protocol. One example is the wireless standard for transmission of data over short distances designated by the trademark Bluetooth®, which has low power consumption. Other examples of wireless transmission options are the wireless standards designated by the trademark WiFi™ or otherwise compliant with 802.11 standards of Institute of Electrical and Electronics Engineers.

The non-volatile memory 6 can be NAND flash, NOR flash, EEPROM, MRAM, resistive RAM or other non-volatile solid state memory. The web service 11 is run on the server 9 and formats available data in a format optimized for connection with other web-connected devices such as the workstation 13. The web service 11 typically provides an object-oriented web-based interface to a database 12 residing on the server 9. The web service 11 can consolidate information from several servers 9 so that the information is more easily viewed by an end-user on a workstation 13. The database 12 and the web service 11 can therefore be run on one server or multiple servers. The server 9 can be remote, as in a cloud server, or in the same building as the workstation 9. The workstation 13 can be a standard tablet, smart phone, laptop or desktop computer connected to a local network or the internet. The workstation 13 can also be a dumb terminal or other dedicated device connected to a local network or the internet.

The time-stamped gesture indications can identify simple movements such as, but not limited to a patient's: walking, gait, standing, sitting, laying down, turning over, standing up from a sitting position, turning, sitting down, transitional movement, restlessness, getting out of bed, falling, waking up, repetitive motion or stationary behavior.

In an optional design of the primary embodiment, the at least one alert event received by the workstation can optionally includes the patient ID, a description of the alert event and a severity level. Optionally the server 9 determines a longer horizon fall prediction for each patient ID by evaluation of a history period of the indication of gesture of the patient ID including at least one of a gait analysis, a rest period analysis, a daily routine analysis and a medication regime and the history period is at least one day and can includes weeks or months. The longer the history period typically the more accurate the longer horizon fall prediction. The patient ID is used by the database 12 to identify a patient. The patient ID can be cross-referenced to: (a) patient mobility level assessment; (b) an activity level assessment; (c) a prevailing medical condition; (d) an end-user data input to the workstation; and (e) an event calendar that tracks at least one of a visitation schedule, a feeding schedule and a staff shift schedule.

In a second embodiment of the invention the design additionally includes, beyond the elements of the primary embodiment, an alarm triggered by each alert event with a severity level exceeding a sensitivity threshold of the patient ID. In an optional design of the second embodiment, the sensitivity threshold for each patient ID server 9 can optionally be configurable by at least one of: (a) a patient mobility level assessment; (b) an activity level assessment; (c) a prevailing medical condition; (d) an end-user data input to the workstation; and (e) an event calendar that tracks at least one of a visitation schedule, a feeding schedule and a staff shift schedule.

The sensitivity threshold is adjustable because some patients are more active than others. What may be cause for alarm from a normally inactive patient (such as sitting up in the bed) may be normal activity for another patient. Thus the sensitivity level can be adjusted in the database 12 record for the patient.

Optionally, an audio cue by the workstation can be sounded by the alarm. Optionally, the system can further comprises a speaker proximate to the patient capable of playing an audio cue sounded by the alarm. Optionally, the system further comprises an alarm management system alerted by the alarm. Optionally, the alarm can be a short horizon fall prediction, wherein the short horizon fall prediction: (a) provides at least 5 seconds, preferably at least 20 seconds, advance notice of a potential fall hazard; and (b) is determined by a recent abnormal movement of the patient. Optionally, the a confirmation prompt can be provided on the workstation for an end-user of the workstation to respond to each alarm with an alarm confirmation and each alarm confirmation can be stored in a confirmation log on the server 9. The audio cue can be a speaker on the workstation 13 or a speaker in the hospital room of the patient. The audio cue could also be a tone played by a tablet or cell phone.

While various aspects and embodiments have been disclosed herein, it will be apparent that various other modifications and adaptations of the invention will be apparent to the person skilled in the art after reading the foregoing disclosure without departing from the spirit and scope of the invention and it is intended that all such modifications and adaptations come within the scope of the appended claims. The various aspects and embodiments disclosed herein are for purposes of illustration and are not intended to be limiting, with the true scope and spirit of the invention being indicated by the appended claims.

REFERENCES

1 . "Survey on Fall Detection and Fall Prevention Using Wearable and External Sensors," Yueng Santiago Delahoz and Miguel Angel Labrador, Sensors, 14:19806-19842, October 2014, doi:10.3390/s141019806.

2. "Automatic Fall Monitoring: A Review," Natthapon Pannurat, Surapa Thiemjarus and Ekawit Nantajeewarawat, Sensors, 14(7):12900-12936, July 2014, doi:10.3390/s140712900.

3. PCT Publication No. WO/2016/003365

Claims

We claim:

\. A system for prevention and prediction of falls comprising:

(a) a plurality of wearable motion detection devices, each wearable motion detection device attachable to at least one of a lower limb and a trunk of a patient and each wearable motion detection device including a band 2 and an electronics module 3 comprising:

(i) a plurality of micro-electro-mechanical sensors 4;

(ii) a wireless transceiver 5;

(iii) a non- volatile memory 6;

(iv) a microprocessor 7 configured to assemble at least one data log, wherein each data log comprises a patient ID and a plurality of time-stamped gesture indications;

(b) at least one local gateway 8 configured to connect wirelessly to each wearable motion detection device and to receive each data log from wearable motion detection device;

(c) at least one server 9, each server 9 including an alarm management module 10, a web service and a database 12, wherein:

(i) each server 9 is configured to connect to the local gateway 8 and receive each data log from the plurality of wearable motion detection devices;

(ii) the alarm management module 10 of each server 9 is configured to process each data log to determine an alert event from the plurality of time- stamped gesture indications and the patient ID in the data log;

(d) at least one workstation, each workstation including a user interface configured to receive alert events from the web service of the at least one server 9.

2. The system of claim 1, wherein the at least one alert event received by the workstation includes:

(a) the patient ID;

(b) a description of the alert event; and

(c) a severity level.

3. The system of claim 1, wherein an alarm is triggered by each alert event with a severity level exceeding a sensitivity threshold of the patient ID.

4. The system of claim 3, wherein the sensitivity threshold for each patient ID server 9 is configurable by at least one of:

(a) a patient mobility level assessment;

(b) an activity level assessment;

(c) a prevailing medical condition;

(d) an end-user data input to the workstation; and

(e) an event calendar that tracks at least one of a visitation schedule, a feeding schedule and a staff shift schedule.

5. The system of claim 3, wherein an audio cue by the workstation is sounded by the alarm.

6. The system of claim 3, the system further comprising a speaker proximate to the patient capable of playing an audio cue sounded by the alarm.

7. The system of claim 3, the system further comprising an alarm management system alerted by the alarm.

8. The system of claim 3, wherein the alarm is a short horizon fall prediction, wherein the short horizon fall prediction:

(a) provides at least 5 seconds advance notice of a potential fall hazard; and

(b) is determined by a recent abnormal movement of the patient.

9. The system of claim 3, wherein:

(a) a confirmation prompt is provided on the workstation for an end-user of the workstation to respond to each alarm with an alarm confirmation; and

(b) each alarm confirmation is stored in a confirmation log on the server 9.

10. The system of claim 1, wherein:

(a) the server 9 determines a longer horizon fall prediction for each patient ID by evaluation of a history period of the indication of gesture of the patient ID including at least one of a gait analysis, a rest period analysis, a daily routine analysis and a medication regime; and

(b) the history period is at least one day.