Fall early warning method and device and storage medium

Abstract

The invention provides a fall early warning method and device and a storage medium. The method comprises the following steps: receiving a first height and a first time measured by a wearable device; receiving a second height and a second time measured after the height of the wearable device changes; calculating a height difference between the second height and the first height and a time difference between the second time and the first time; and judging whether the height difference is greater than a preset value and the time difference is smaller than the preset value, and if the previous two conditions are simultaneously met, triggering an alarm. The device equipment and the storage medium are used for storing a program product for running the fall early warning method. The system is used for monitoring the elderly patients and can give an alarm in time in the early stage of falling down, so that intervention or rescue can be performed in time, and various health threats caused by falling down of the elderly patients are reduced.

Description

Technical Field

[0001] The present invention belongs to the field of medical care technology, and specifically relates to a fall warning method, device, storage medium and program product. Background Art

[0002] With the acceleration of the aging of society, the number of elderly people is also increasing. Falls are one of the main causes of disability among the elderly. In order to protect the health of the elderly, medical staff are generally used to care for the elderly, so that they can be treated and rescued in time after they suffer accidental injuries such as falls and sudden illnesses.

[0003] Some fall prevention devices currently exist, but they primarily rely on passive sensing for prevention, triggering alarms by detecting the patient's movements. These devices are unable to provide accurate fall warning information to medical staff at the nursing station. For example, patent document CN117137748A provides a clinical fall-sensing mattress, a fall warning system, and a fall warning method. These devices monitor the pressure distribution data of the patient's body and head on the bed, as well as the corresponding pressure distribution changes, and transmit these data to the nurse station via a communication data line. The nurse station then calculates the rate of change of the body, head, or combined pressure distribution (P1, P2, or P3) in real time, and determines whether P1, P2, or P3 has reached a corresponding threshold. If so, a corresponding alarm signal is issued, which is then responded to by the alarm module. This signal is then shared with the nursing terminal PDA, which issues a corresponding risk alert, alerting nursing staff to provide care. This allows for accurate fall warnings based on different distributions, avoiding false alarms from patients and wasting nursing resources, and enabling medical staff to respond promptly and effectively, ensuring optimal care arrangements.

[0004] Such devices are primarily used to prevent patients from leaving their beds to prevent falls. However, in actual medical care, bed mobility is an important measure to promote patient recovery, and not allowing patients to leave their beds significantly reduces the versatility and applicability of such devices. Summary of the Invention

[0005] The main purpose of the present invention is to provide a fall warning method, device, equipment, storage medium and program product for providing fall warning to elderly patients leaving bed, thereby solving or partially solving the problems identified in the background art. To this end, the present invention provides the following technical solutions:

[0006] In a first aspect, the present invention provides a fall warning method, characterized by comprising:

[0007] receiving a first height and a first time measured by the wearable device;

[0008] Receiving a second height and a second time measured after the wearable device changes height;

[0009] Calculate the height difference between the second height and the first height, and the time difference between the second time and the first time;

[0010] Determine whether the height difference is greater than a preset value and whether the time difference is less than a preset value. If the first two conditions are met at the same time, an alarm is triggered.

[0011] In some embodiments, the wearable device includes smart glasses, wireless headphones, smart bracelets, smart watches and smart clothes.

[0012] In some embodiments, the alarm further includes: sending the location information of the wearable device. In at least one embodiment, the location information is obtained by a method including satellite navigation positioning, communication base station positioning, RFID positioning, WiFi positioning, and Bluetooth positioning.

[0013] In some embodiments, the method for obtaining the preset value includes: continuously crawling videos of elderly people falling on the Internet to obtain a first video set with an increasing sample size; continuously crawling videos of elderly people walking normally on the Internet to obtain a second video set with an increasing sample size; based on the first video set, counting the falling speed of the elderly during the fall process; based on the second video set, counting the maximum height change of the body part corresponding to the wearable device, and using the value of the maximum height change as the preset value of the height difference; calculating the quotient of the maximum height change and the falling speed to obtain the fall time, and using the fall time as the preset value of the time difference.

[0014] In at least one embodiment, the statistical method for the fall speed includes: using the first video set as a training sample and the elderly as the video cutting target, and training and video segmenting the first video set using a video segmentation model; using the picture before the fall in the video as the initial frame, and the picture after the fall as the ending frame, counting the picture changes of the ending frame relative to the initial frame, obtaining the height change of the elderly and the total duration of the fall, and counting the quotient of the height change and the total duration to obtain the fall speed.

[0015] In at least one embodiment, the statistical method of the maximum height change includes: using the second video set as a training sample, the elderly as the video cutting target, and the body parts corresponding to the wearable device as marking points, and training and video segmenting the second video set using a video segmentation model; using the highest position of the marking point in the video as the first position, and the lowest position in the video as the second position, and counting the difference between the first position and the second position to obtain the numerical value of the maximum height change.

[0016] In a second aspect, the present invention provides a fall warning device, comprising a memory and a processor, wherein the memory stores computer program instructions, and the computer program instructions are read by the processor and execute the above method when executed.

[0017] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program executes the above method when executed by a computer.

[0018] In a fourth aspect, the present invention provides a computer program product, which executes the above method when run by a computer.

[0019] Compared with the existing technology, the present invention can at least achieve the following beneficial effects: by constantly monitoring the height change value and height change speed measured by the wearable device on the elderly patient, the present invention can issue an alarm at the moment when the elderly patient has more violent body shaking, so that nearby medical staff can be informed of the situation in time, so that they can rush to the patient's side in the initial stage of the patient's fall or immediately after the fall, and intervene in the fall or provide immediate treatment, reducing the harm caused by the fall to the elderly patient. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] One or more embodiments of the present invention will now be described, by way of example only, with reference to the accompanying drawings, in which:

[0021] Figure 1 A flowchart of a fall warning method provided by an embodiment of the present invention;

[0022] Figure 2 A flowchart of a method for obtaining a preset value in a fall warning method;

[0023] Figure 3 This is a structural block diagram of the fall warning device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The present invention will be described in detail below with reference to the exemplary embodiments shown in the accompanying drawings. However, it should be understood that the present invention can be implemented in a variety of different forms and should not be construed as being limited to the embodiments set forth herein. These embodiments are provided herein to make the disclosure of the present invention more complete and to fully convey the concepts of the present invention to those skilled in the art.

[0025] First, refer to Figure 1 The present invention provides a fall warning method, comprising the following steps:

[0026] S1. Receive a first height and a first time measured by a wearable device;

[0027] S2, receiving a second height and a second time measured after the wearable device changes height;

[0028] S3, calculating the height difference between the second height and the first height, and the time difference between the second time and the first time;

[0029] S4. Determine whether the height difference is greater than a preset value and whether the time difference is less than a preset value. If the first two conditions are met at the same time, trigger an alarm.

[0030] In step S1, the wearable device is preferably a device worn on the upper body of the elderly patient so that it can accurately measure the fall at the early stage. For example, the wearable device can be smart glasses, wireless headphones, smart bracelets, smart watches or smart clothes. These listed wearable devices are all common devices in this field, and their specific structures and operating principles are not described in detail herein. However, it should be emphasized that in order to be able to measure the height, a corresponding sensor unit should be provided inside the wearable device, such as a vertical height sensor.

[0031] In step S4, the method for obtaining the preset value includes the following steps:

[0032] S41. Continuously crawl videos of elderly people falling on the Internet to obtain a first video set with an increasing sample size.

[0033] S42, continuously crawling videos of elderly people walking normally on the Internet to obtain a second video set with an increasing sample size;

[0034] S43. Counting the falling speed of the elderly person during the falling process based on the first video set;

[0035] S44. Counting the maximum height change of the body part corresponding to the wearable device based on the second video set, and using the value of the maximum height change as the preset value of the height difference;

[0036] S45. Calculate the quotient of the maximum height change and the falling speed to obtain the falling time, and use the falling time as the preset value of the time difference.

[0037] Web crawling for relevant information is a common web crawler technology that simulates human browsing behavior, automatically accesses web pages, extracts data and stores it according to preset rules. When crawling information about related videos, you can first set relevant keywords. After obtaining the relevant web pages, the relevant video materials contained therein will be downloaded and stored.

[0038] The statistical method for calculating the falling speed includes: using the first video set as a training sample and the elderly as the video cutting target, and training and segmenting the first video set using a video segmentation model; using the picture before the fall in the video as the initial frame and the picture after the fall as the ending frame, counting the picture changes of the ending frame relative to the initial frame, obtaining the height change of the elderly and the total duration of the fall, and calculating the quotient of the height change and the total duration to obtain the falling speed.

[0039] The statistical method for the maximum height change includes: using the second video set as a training sample, the elderly as the video cutting target, and the body parts corresponding to the wearable device as marking points, and training and segmenting the second video set using a video segmentation model; taking the highest position of the marking point in the video as the first position, and the lowest position in the video as the second position, and counting the difference between the first position and the second position to obtain the numerical value of the maximum height change.

[0040] The present invention forms a method of continuously increasing the video sample inventory by continuously expanding the video set through different crawled videos, which can also be more accurate when conducting model training. Moreover, in the actual operation process, it uses a method of conducting data statistics while training, rather than the conventional method of first conducting a large number of sample training and then using them for individual quantity detection. This can make the final training and statistical results more inclined to ideal results.

[0041] In a specific embodiment, the large-scale video segmentation model used can be the UniVS model (see https: / / arxiv.org / abs / 2402.18115). This model consists of three main modules: an image encoder, a hint encoder, and a unified video mask decoder. The image encoder converts the input RGB image into picture tokens, while the hint encoder converts the raw visual / textual hints into hint embeddings. The unified video mask decoder explicitly decodes the mask of any entity or hint-guided object in the video. The training process consists of three stages: image-level training, video-level training, and long-video fine-tuning. In the first stage, UniVS is trained on multiple image segmentation datasets, pre-trained using image-level annotations to obtain good visual representations. In the second stage, a short video clip consisting of a few frames is fed into the pre-trained model and fine-tuned on a video segmentation dataset to perceive changes in object motion trajectory over short time periods. In the third stage, the unified video mask decoder is further fine-tuned using long video sequences consisting of multiple frames, encouraging it to learn more discriminative features and trajectory information over longer time frames.

[0042] In step S4, the alarm further includes: sending the location information of the wearable device. The location information is obtained by methods including satellite navigation positioning, communication base station positioning, RFID positioning, WiFi positioning, and Bluetooth positioning. The location information is mainly used to provide remote medical personnel with patient location information.

[0043] Finally, the present invention also provides a fall warning device, a computer-readable storage medium, and a computer program product.

[0044] The fall warning device includes a memory and a processor. The memory stores a computer program that, when read and executed by the processor, executes the method described above. The fall warning device described in the embodiments of the present invention is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, servers, mainframe computers, and other suitable computers. Computer devices may also represent various forms of mobile devices, such as personal digital assistants (PDAs), smartphones, wearable devices, and other similar computing devices. Furthermore, the components, their connections and relationships, and their functions shown herein are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein. In addition to the memory and processor, the computer device may also include a display, etc. These components are commonly found in the art, and their types and models are conventionally selected, and will not be further described in detail in this invention. It should be noted that, for the present invention, the fall warning device preferably operates in two simultaneous modes: one is deployed in a nurse's station to remotely alert medical staff at the nurse's station; the other is a mobile device carried by a nurse to promptly alert the medical staff closest to the patient.

[0045] The computer-readable storage medium stores a computer program, which executes the above method when executed by a computer. The computer-readable medium described in the embodiment of the present invention may be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. The computer-readable storage medium may be, for example, (but not limited to) an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or device. In the present invention, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wire, optical cable, RF, or any suitable combination thereof.

[0046] The computer program product executes the above method when it is run by a computer. The computer program product described in the embodiment of the present invention is stored in a specific readable storage medium and can be run by a computer device. The computer program product should not be limited to running on a certain operating system. These operating systems include: Unix-like systems (such as Linux), Microsoft Windows systems, Apple macOS, Chrome OS, Android, etc. In addition, the programming language used by the computer program product should not be limited to a specific programming language. These programming languages ​​include: C language, C++, Java, Python, JavaScript, Go language, etc.

[0047] It should be understood that all the above embodiments are illustrative rather than restrictive, and various modifications or variations made by those skilled in the art to the specific embodiments described above under the concept of the present invention should be within the scope of protection of the present invention.

Claims

1. A fall warning method, characterized in that: include: receiving a first height and a first time measured by the wearable device; Receiving a second height and a second time measured after the wearable device changes height; Calculate the height difference between the second height and the first height, and the time difference between the second time and the first time; Determine whether the height difference is greater than a preset value and whether the time difference is less than a preset value. If the first two conditions are met at the same time, an alarm is triggered.

2. The fall warning method according to claim 1, characterized in that: The wearable devices include smart glasses, wireless headphones, smart bracelets, smart watches and smart clothes.

3. The fall warning method according to claim 1, characterized in that: The alarm also includes: Sending location information of the wearable device.

4. The fall warning method according to claim 4, characterized in that: The method for obtaining the location information includes satellite navigation positioning, communication base station positioning, RFID positioning, WiFi positioning and Bluetooth positioning.

5. The fall warning method according to claim 1, characterized in that: The method for obtaining the preset value includes: Continuously crawl videos of elderly people falling on the Internet to obtain the first video set with an increasing sample size; Continuously crawl videos of elderly people walking normally on the Internet to obtain a second video set with increasing sample size; According to the first video set, counting the falling speed of the elderly during the falling process; Counting the maximum height change of the body part corresponding to the wearable device based on the second video set, and using the value of the maximum height change as the preset value of the height difference; The quotient of the maximum height change and the falling speed is calculated to obtain the falling time, and the falling time is used as the preset value of the time difference.

6. The fall warning method according to claim 5, characterized in that: The statistical method of the falling speed includes: Using the first video set as training samples and the elderly as video segmentation targets, the first video set is trained and segmented using a video segmentation model; The picture before the fall in the video is taken as the initial frame, and the picture after the fall is taken as the ending frame. The picture changes of the ending frame relative to the initial frame are counted to obtain the elderly person's height change and the total duration of the fall. The quotient of the height change and the total duration is calculated to obtain the fall speed.

7. The fall warning method according to claim 5, characterized in that: Statistical methods for maximum height change include: Using the second video set as training samples, the elderly as video segmentation targets, and the body parts corresponding to the wearable device as markers, the second video set is trained and segmented using a video segmentation model; The highest position of the marking point in the video is taken as the first position, and the lowest position in the video is taken as the second position. The difference between the first position and the second position is counted to obtain the value of the maximum height change.

8. A fall warning device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the computer program is read and executed by the processor, the method according to any one of claims 1 to 7 is executed.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed on a computer, the method according to any one of claims 1 to 7 is executed.

10. A computer program product, characterized in that: When the computer program product is executed by a computer, the method according to any one of claims 1 to 7 is executed.

Images