CPR intelligent training management method and system using computer vision

Abstract

The present invention provides a method and system for intelligent CPR training management using computer vision, comprising: acquiring a video image of a CPR training session; identifying a trainee's head from the video image; identifying the trainee's body posture based on the height of the head above the ground; and identifying the trainee's movement information based on the head's motion information. Specifically for CPR training scenarios, the present invention uses the head as the primary recognition target, enabling more accurate analysis of the trainee's body posture. It can promptly detect incorrect postures based on the trainee's head position and avoid misjudgments when the trainee switches between correct postures, significantly improving the level of intelligence.

Description

Technical Field

[0001] The present invention relates to the field of CPR intelligent training, and in particular to a CPR intelligent training management method and system using computer vision. Background Art

[0002] CPR (Cardio Pulmonary Resuscitation) refers to cardiopulmonary resuscitation, including artificial respiration and external cardiac massage. Whether caused by cardiac or non-cardiac causes, sudden cessation of effective cardiac pumping and circulation leads to severe ischemia, hypoxia, and metabolic disorders in tissues and cells throughout the body. Without prompt treatment, life can be lost instantly. However, with timely and effective resuscitation measures, life can be saved and recovery is possible. Therefore, CPR courses are extremely important for medical personnel to learn first aid knowledge.

[0003] Patent document CN102713992B discloses a computer-implemented method for managing rescue training, comprising: registering multiple individuals as potential rescue trainees via a central computer server system; receiving information indicating the rate and depth of compressions of multiple of the potential rescue trainees from a computing device remote from the central server system; and generating comparative data reflecting CPR chest compressions performed by a first rescue trainee relative to other rescue trainees. The method also includes providing the comparative data via a network for review by one or more of the rescue trainees. However, patent document CN102713992B fails to utilize video information.

[0004] Patent document CN113592788A discloses a CPR compression depth measurement method and system based on machine vision. The method includes the following steps: scanning a QR code on a CPR feedback pad to obtain the position information of the CPR feedback pad, which is a passive component; based on the position information, obtaining an inclination angle β with the feedback pad, capturing a static image of the scale lines on the CPR feedback pad, and calculating a pixel coefficient k based on the static image of the scale lines; obtaining dynamic compression video data, processing the dynamic compression video data, and calculating the compression depth based on the dynamic compression video data, the inclination angle β, and the pixel coefficient k. The present invention realizes the function of measuring compression depth without a cable. The CPR feedback pad is a passive component that assists compression and will not cause damage to the components due to compression, thereby interrupting the CPR operation. However, patent document CN113592788A cannot identify and judge the actions of the rescuer.

[0005] Patent document CN105469679A discloses a Kinect-based cardiopulmonary resuscitation auxiliary training system and method. The cardiopulmonary resuscitation auxiliary training system can display the straightness and verticality of the operator's arm in real time, and save the skeletal data and video information of the entire operation. The data collected by the Kinect sensor can be integrated with the cardiopulmonary resuscitation operation information collected by the traditional cardiopulmonary resuscitation training system to obtain comprehensive evaluation results including operation posture, pressing intensity, pressing speed, pressing position, etc., which fully reflect the accuracy of the operator's overall operation. However, the recognition of bones such as arms in patent document CN105469679A is easily blocked by clothing and other parts of the body, and the recognition accuracy needs to be improved. Summary of the Invention

[0006] In view of the defects in the prior art, the purpose of the present invention is to provide a CPR intelligent training management method and system using computer vision.

[0007] According to the present invention, a CPR intelligent training management method using computer vision is provided, comprising:

[0008] Step S1: Acquire a video image of CPR training;

[0009] Step S2: Recognizing the trainee's head from the video image;

[0010] Step S3: Identify the trainee's body posture based on the height of the head from the ground;

[0011] Step S4: According to the head movement information, the trainee's movement information is identified.

[0012] Preferably, step S3 includes:

[0013] Step S301: Recognize the head of the trainee in the video image;

[0014] Step S302: Identify the ground in the video image;

[0015] Step S303: obtaining the height of the trainee's head from the ground;

[0016] Step S304: obtaining the movement amplitude of the trainee's head;

[0017] Step S305: If the movement amplitude of the trainee's head is smaller than the first movement amplitude, and the height of the trainee's head from the ground is greater than the first height, it is considered that the trainee's body posture is external cardiac massage;

[0018] If the movement amplitude of the trainee's head is greater than the first movement amplitude, and the height of the trainee's head from the ground is less than the first height, then the trainee's body posture is considered to be artificial respiration;

[0019] In other cases, it is considered that the trainee's body posture is incorrect.

[0020] Preferably, the step S4 includes:

[0021] For external cardiac massage, the body posture is based on the degree of vertical movement of the trainee's head to obtain the compression depth; or

[0022] As for the body posture for artificial respiration, the frequency of artificial respiration is obtained according to the swing frequency of the trainee's head.

[0023] Preferably, a reminder is given to trainees whose body posture is incorrect.

[0024] Preferably, if it is identified that the duration of the trainee's incorrect body posture exceeds a first time threshold, a prompt is given; otherwise, it is considered that the trainee switches between different body postures.

[0025] According to the present invention, a CPR intelligent training management system using computer vision is provided, comprising:

[0026] Module M1: Acquire video images of CPR training;

[0027] Module M2: Recognizing the trainee's head from the video image;

[0028] Module M3: Identify the trainee’s body posture based on the height of the head from the ground;

[0029] Module M4: Identify the trainee's movement information based on the head movement information.

[0030] Preferably, the module M3 includes:

[0031] Module M301: Recognize the trainee's head in the video image;

[0032] Module M302: Identify the ground in the video image;

[0033] Module M303: Obtain the height of the trainee's head from the ground;

[0034] Module M304: obtaining the range of motion of the trainee's head;

[0035] Module M305: If the movement amplitude of the trainee's head is less than the first movement amplitude, and the height of the trainee's head from the ground is greater than the first height, it is determined that the trainee's body posture is external cardiac massage;

[0036] If the movement amplitude of the trainee's head is greater than the first movement amplitude, and the height of the trainee's head from the ground is less than the first height, then the trainee's body posture is considered to be artificial respiration;

[0037] In other cases, it is considered that the trainee's body posture is incorrect.

[0038] Preferably, the module M4 includes:

[0039] For external cardiac massage, the body posture is based on the degree of vertical movement of the trainee's head to obtain the compression depth; or

[0040] As for the body posture for artificial respiration, the frequency of artificial respiration is obtained according to the swing frequency of the trainee's head.

[0041] Preferably, a reminder is given to trainees whose body posture is incorrect.

[0042] Preferably, if it is identified that the duration of the trainee's incorrect body posture exceeds a first time threshold, a prompt is given; otherwise, it is considered that the trainee switches between different body postures.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] 1. The present invention is aimed at CPR training scenarios. Using the head as the main recognition target can more accurately analyze the trainee's body posture.

[0045] 2. The present invention can promptly detect incorrect body postures based on the position of the trainee's head.

[0046] 3. The present invention avoids misjudging incorrect body postures when the trainee switches between different correct body postures, thereby greatly improving the level of intelligence. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Other features, objects and advantages of the present invention will become more apparent upon reading the detailed description of non-limiting embodiments with reference to the following drawings:

[0048] Figure 1 It is a schematic diagram of the steps of the present invention. DETAILED DESCRIPTION

[0049] The present invention will be described in detail below with reference to specific embodiments. The following examples will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those skilled in the art, several changes and improvements can be made without departing from the scope of the present invention. These all fall within the scope of protection of the present invention.

[0050] According to a CPR intelligent training management method using computer vision provided by the present invention, the present invention is aimed at CPR training scenarios, and uses the head as the main identification target to more accurately analyze the trainee's body posture, including: step S1: obtaining a video image of CPR training; for example, obtaining a video screen of CPR training as a video image through a camera and / or a depth camera. In the CPR training site, markers can be set, or calibration can be performed in advance to obtain the spatial position relationship between the camera equipment and the CPR training site in advance, so that the ground can also be identified; step S2: identifying the trainee's head from the video image; specifically, the trainee's head can be identified through trained neural network face recognition.

[0051] The present invention is described in more detail below. Step S3: Identify the trainee's body posture based on the height of the head from the ground; specifically, step S3 includes: step S301: identify the trainee's head in the video image; step S302: identify the ground in the video image; step S303: obtain the height of the trainee's head from the ground; step S304: obtain the movement amplitude of the trainee's head; step S305: if the movement amplitude of the trainee's head is less than the first movement amplitude, and the height of the trainee's head from the ground is greater than the first height, then it is considered that the trainee's body posture is external cardiac massage; wherein, the first movement amplitude can be set in advance, and the first height can be set in advance. If the movement amplitude of the trainee's head is greater than the first movement amplitude, and the height of the trainee's head from the ground is less than the first height, then it is considered that the trainee's body posture is artificial respiration; in other cases, it is considered that the trainee's body posture is incorrect. In a further preferred embodiment, specifically, a prompt is provided for a trainee with an incorrect body posture. More specifically, if it is determined that the duration of the trainee's incorrect body posture exceeds a first time threshold, a prompt is provided; otherwise, it is assumed that the trainee is switching between different body postures. The first time threshold is pre-set to match the time it takes to switch from an external cardiac massage posture to an artificial respiration posture, and vice versa. In this way, the present invention can promptly detect incorrect body postures based on the position of the trainee's head. Furthermore, the present invention avoids misjudging incorrect body postures when a trainee switches between different correct body postures, significantly improving the level of intelligence.

[0052] Furthermore, step S4: based on the head movement information, the trainee's motion information is identified. Specifically, step S4 includes: for external cardiac massage, the compression depth is determined based on the amplitude of the trainee's head in a direction perpendicular to the ground; or for artificial respiration, the frequency of artificial respiration is determined based on the swing frequency of the trainee's head. During external cardiac massage, the amplitude of the head's up and down movement is proportional to the compression depth. Therefore, the present invention does not require a piezoelectric sensor to convert the compression depth by proportionally converting the head's movement amplitude.

[0053] According to the present invention, a CPR intelligent training management system using computer vision is provided, comprising:

[0054] Module M1: Acquire video images of CPR training;

[0055] Module M2: Recognizing the trainee's head from the video image;

[0056] Module M3: Identifies the trainee's body posture based on the height of the trainee's head from the ground. Module M3 includes: Module M301: Identifies the trainee's head in the video image; Module M302: Identifies the ground in the video image; Module M303: Obtains the height of the trainee's head from the ground; Module M304: Obtains the amplitude of movement of the trainee's head; Module M305: If the amplitude of movement of the trainee's head is less than a first amplitude and the height of the trainee's head from the ground is greater than the first height, the trainee's body posture is considered to be external cardiac massage; if the amplitude of movement of the trainee's head is greater than the first amplitude and the height of the trainee's head from the ground is less than the first height, the trainee's body posture is considered to be artificial respiration; otherwise, the trainee's body posture is considered incorrect. A prompt is provided for trainees with incorrect body postures. If the duration of the trainee's incorrect body posture exceeds a first time threshold, a prompt is provided; otherwise, the trainee is considered to be switching between different body postures.

[0057] Module M4: Identifies the trainee's motion information based on the head movement information. This module includes: for external cardiac massage, determining the compression depth based on the trainee's head's vertical amplitude relative to the ground; or for artificial respiration, determining the frequency of artificial respiration based on the trainee's head's head movement frequency.

[0058] Those skilled in the art will appreciate that, in addition to implementing the system, device, and various modules provided by the present invention in purely computer-readable program code, it is entirely possible to implement the same program in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, embedded microcontrollers, and the like by logically programming the method steps. Therefore, the system, device, and various modules provided by the present invention can be considered a hardware component, and the modules included therein for implementing various programs can also be considered structures within the hardware component; the modules for implementing various functions can also be considered both software programs for implementing the method and structures within the hardware component.

[0059] The above describes specific embodiments of the present invention. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art may make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. The embodiments of this application and the features in the embodiments may be combined with each other in any manner unless there is a conflict.

Claims

1. A CPR intelligent training management method using computer vision, characterized in that: include: Step S1: Acquire a video image of CPR training; Step S2: Recognizing the trainee's head from the video image; Step S3: Identify the trainee's body posture based on the height of the head from the ground; Step S4: Identify the trainee's movement information based on the head movement information; The step S3 comprises: Step S301: Recognize the head of the trainee in the video image; Step S302: Identify the ground in the video image; Step S303: obtaining the height of the trainee's head from the ground; Step S304: obtaining the movement amplitude of the trainee's head; Step S305: If the movement amplitude of the trainee's head is smaller than the first movement amplitude, and the height of the trainee's head from the ground is greater than the first height, it is considered that the trainee's body posture is external cardiac massage; If the movement amplitude of the trainee's head is greater than the first movement amplitude, and the height of the trainee's head from the ground is less than the first height, then the trainee's body posture is considered to be artificial respiration; In other cases, it is considered that the trainee's body posture is incorrect.

2. The CPR intelligent training management method using computer vision according to claim 1, characterized in that: The step S4 comprises: For external cardiac massage, the body posture is based on the degree of vertical movement of the trainee's head to obtain the compression depth; or As for the body posture for artificial respiration, the frequency of artificial respiration is obtained according to the swing frequency of the trainee's head.

3. The CPR intelligent training management method using computer vision according to claim 1, characterized in that: Provide reminders to trainees with incorrect body posture.

4. The CPR intelligent training management method using computer vision according to claim 3, characterized in that: If it is identified that the duration of the trainee's incorrect body posture exceeds the first time threshold, a prompt is given; otherwise, it is considered that the trainee is switching between different body postures.

5. A CPR intelligent training management system using computer vision, characterized in that: include: Module M1: Acquire video images of CPR training; Module M2: Recognizing the trainee's head from the video image; Module M3: Identify the trainee’s body posture based on the height of the head from the ground; Module M4: Identify the trainee’s movement information based on the head movement information; The module M3 includes: Module M301: Recognize the trainee's head in the video image; Module M302: Identify the ground in the video image; Module M303: Obtain the height of the trainee's head from the ground; Module M304: obtaining the range of motion of the trainee's head; Module M305: If the movement amplitude of the trainee's head is less than the first movement amplitude, and the height of the trainee's head from the ground is greater than the first height, it is determined that the trainee's body posture is external cardiac massage; If the movement amplitude of the trainee's head is greater than the first movement amplitude, and the height of the trainee's head from the ground is less than the first height, then the trainee's body posture is considered to be artificial respiration; In other cases, it is considered that the trainee's body posture is incorrect.

6. The CPR intelligent training management system using computer vision according to claim 5, characterized in that: The module M4 includes: For external cardiac massage, the body posture is based on the degree of vertical movement of the trainee's head to obtain the compression depth; or As for the body posture for artificial respiration, the frequency of artificial respiration is obtained according to the swing frequency of the trainee's head.

7. The CPR intelligent training management system using computer vision according to claim 6, characterized in that: Provide reminders to trainees with incorrect body posture.

8. The CPR intelligent training management system using computer vision according to claim 7, characterized in that: If it is identified that the duration of the trainee's incorrect body posture exceeds the first time threshold, a prompt is given; otherwise, it is considered that the trainee is switching between different body postures.

Images