Bedsore risk assessment system and method of the same

Abstract

A pressure ulcer risk assessment system according to the present invention comprises: a pressure data receiving unit that receives pressure data from a plurality of pressure sensors; a posture and body part estimation unit that detects keypoint coordinates, which are characteristic points of the body, and estimates a posture based on the pressure data, and outputs PROI (Pressure Region of Interest) coordinate data, which is a pressure region of interest, and estimates a body part using the keypoint coordinates and the posture; a body part risk assessment unit that evaluates the risk for each body part using the PROI coordinate data, statistical data, and learning data, and generates body part risk assessment data; and a body part risk assessment data output unit that classifies and outputs the body part risk assessment data into grades according to the risk of pressure ulcer occurrence.

Description

Technology Field

[0001] The present invention relates to a pressure ulcer risk assessment system and method, and more specifically, to a pressure ulcer risk assessment system and method that can estimate posture and body parts using pressure data, assess pressure ulcer risk for each estimated body part, and provide intuitive data using pressure ulcer risk assessment data. Background Technology

[0002] Body parts subjected to continuous and repetitive pressure may experience impaired blood circulation. In particular, pressure applied to bony prominences can lead to a lack of oxygen and nutrients supplied to capillaries, potentially causing skin damage due to ischemia in the skin, subcutaneous fat, and muscles. Elderly individuals, people with disabilities, and patients who have difficulty controlling their own bodies are at a high risk of developing pressure sores on skin areas in direct contact with the surface as they maintain the same posture for extended periods. To prevent such pressure sores, it is necessary to change positions at regular intervals to relieve pressure on the body and keep the area dry.

[0003] Traditionally, to prevent pressure ulcers, caregivers such as medical personnel and nursing assistants identified areas of pressure on the patient and performed pressure relief activities. However, since caregivers must manually change the patient's position to relieve pressure applied to the body, this requires significant time and cost. Furthermore, there are limitations to caregivers continuously and consistently assessing the risk of pressure ulcer development throughout the day.

[0004] Furthermore, since the pressure applied to the same body part varies depending on individual conditions such as body weight, height, gender, and age, the risk of developing pressure ulcers must be managed on a per-patient basis. However, it is difficult for caregivers to individually identify the pressure applied to each body part according to the patient's posture and assess the risk of developing pressure ulcers.

[0005] Therefore, in order to prevent pressure ulcers, it is necessary to develop a pressure ulcer risk assessment system that can identify the pressure applied to each body part according to the patient's posture, determine the risk level, and provide intuitive data for proper posture changes. Prior art literature

[65535] Republic of Korea Registered Patent Publication No. 10-2822986; Japanese Published Patent Publication No. 2013-517850 The problem to be solved

[0006] The technical problem that the present invention aims to solve is to provide a pressure ulcer risk assessment system and a method that can estimate posture and body parts using pressure data, assess pressure ulcer risk for each estimated body part, and provide intuitive data using pressure ulcer risk assessment data.

[0007] In addition, the technical problem that the present invention aims to solve is to a pressure ulcer risk assessment system and method capable of detecting keypoint coordinates, which are characteristic points of the body, by calculating the depth of pressure data.

[0008] In addition, the technical problem that the present invention aims to solve is to provide a pressure ulcer risk assessment system and method that can estimate body parts according to posture by activating keypoint coordinates required for each posture and extracting PROI coordinate data, which is a pressure region of interest, centered on the activated keypoint coordinates.

[0009] In addition, the technical problem that the present invention aims to solve is to provide a pressure ulcer risk assessment system and method that can evaluate the risk level for each body part based on extracted PROI coordinate data and output risk assessment data customized to the user by reflecting statistical data and learning data.

[0010] In addition, the technical problem that the present invention aims to solve is to provide intuitive auxiliary data by generating and providing 3D data based on pressure data and PROI coordinate data, and to provide a pressure ulcer risk assessment system and method. means of solving the problem

[0011] A pressure ulcer risk assessment system according to one embodiment of the present invention for achieving the aforementioned technical challenges may include: a pressure data receiving unit that receives pressure data from a plurality of pressure sensors; a posture and body part estimation unit that detects keypoint coordinates, which are characteristic points of the body, based on the pressure data and estimates the posture, and outputs PROI (Pressure Region of Interest) coordinate data, which is a pressure region of interest, and estimates the body part using the keypoint coordinates and the posture; a body part risk assessment unit that evaluates the risk level for each body part using the PROI coordinate data, statistical data, and learning data, and generates body part risk assessment data; and a body part risk assessment data output unit that classifies the body part risk assessment data into grades according to the risk of pressure ulcer occurrence and outputs them.

[0012] The posture and body part estimation unit may include a posture estimation unit that stores a posture estimation model that outputs the keypoint coordinates and the posture with the pressure data as input; and a body part estimation unit that stores a body part estimation model that outputs the PROI coordinate data with the keypoint coordinates and the posture as input.

[0013] The above posture estimation model can calculate the depth using the pressure intensity of the pressure data and detect the keypoint coordinates in 3D space using the calculated depth.

[0014] The above body part estimation model can activate the keypoint coordinates required for each posture, generate a clipping mask centered on the activated keypoint coordinates, and project the generated clipping mask onto the pressure data to extract the PROI coordinate data.

[0015] The above-mentioned risk assessment unit for each body part assigns cumulative weights over time and outputs the above-mentioned risk assessment data for each body part, and the cumulative weights may include at least one of cumulative time, the presence of bedsores, nutritional status, or skin condition.

[0016] It further includes a 3D data generation unit that generates and provides the pressure data as 3D data based on the PROI coordinate data, supplements the generated 3D data using pre-stored posture image data, and can display the risk assessment data for each part by reflecting it.

[0017] In addition, a method for assessing pressure ulcer risk according to an embodiment of the present invention for achieving the aforementioned technical challenge may include: a) receiving pressure data from a plurality of pressure sensors; b) inputting the pressure data into a posture estimation model to detect keypoint coordinates, which are characteristic points of the body, and estimating the posture; c) inputting the keypoint coordinates and the posture into a body part estimation model to output PROI coordinate data, which is a pressure region of interest, and estimating the body part; and d) evaluating the risk for each body part using the PROI coordinate data, statistical data, and learning data, generating risk assessment data for each part, and classifying and outputting the risk assessment data for each part into grades according to the risk of pressure ulcer occurrence.

[0018] Step b) above calculates the depth using the pressure intensity of the pressure data and can detect the keypoint coordinates in 3D space using the calculated depth.

[0019] The above step c) may include: a step of activating the keypoint coordinates required for each posture; a step of generating a clipping mask centered on the activated keypoint coordinates; and a step of projecting the generated clipping mask onto the pressure data to extract the PROI coordinate data.

[0020] Step d) above outputs the risk assessment data for each body part by assigning cumulative weights over time, and the cumulative weights may include at least one of cumulative time, presence of bedsores, nutritional status, or skin condition.

[0021] e) further includes the step of generating and providing the pressure data as 3D data based on the PROI coordinate data, wherein step e) can supplement the generated 3D data using pre-stored posture image data and display the risk assessment data for each part. Effects of the invention

[0022] According to an embodiment of the present invention, posture and body parts can be estimated using pressure data, and the risk of pressure ulcers can be evaluated for each body part according to the estimated posture. That is, since it is possible to evaluate the risk of pressure ulcers for each body part according to posture, the pressure applied to the subject's body can be accurately identified and relieved.

[0023] In addition, according to an embodiment of the present invention, keypoint coordinates, which are characteristic points of the body, can be detected based on pressure data. Furthermore, since the keypoint coordinates are pre-set for each posture, the subject's posture can be estimated more accurately.

[0024] In addition, according to an embodiment of the present invention, keypoint coordinates required for each posture can be activated, and a clipping mask can be generated centered on the activated keypoint coordinates. Furthermore, by projecting the clipping mask onto pressure data, PROI coordinate data representing the pressure region of interest can be obtained. Here, since the clipping mask is sized according to conditions such as the subject's height and weight, it is possible to identify a customized pressure region of interest and estimate the body part region by reflecting the subject's physical characteristics.

[0025] In addition, according to an embodiment of the present invention, personalized risk assessment data for each body part can be output by reflecting PROI coordinate data, training data, and statistical data.

[0026] In addition, according to an embodiment of the present invention, 3D data can be generated and provided based on pressure data and PROI coordinate data. At this time, inaccurate pressure data for parts of the body that are not in contact can be supplemented by using pre-stored posture image data. Furthermore, by reflecting risk assessment data for each body part in the generated 3D data, the operator can intuitively understand the situation through the 3D image and respond quickly.

[0027] The effects of the present invention are not limited to the effects described above, and should be understood to include all effects that can be inferred from the configuration of the invention described in the detailed description of the invention or the claims. Brief explanation of the drawing

[0028] FIG. 1 is a diagram illustrating the configuration of a pressure ulcer risk assessment system according to an embodiment of the present invention. FIG. 2 is a diagram illustrating the operation of a posture and body part estimation unit according to an embodiment of the present invention. FIG. 3 is a diagram illustrating the output of PROI coordinate data by the body part estimation unit according to an embodiment of the present invention. FIG. 4 is a diagram illustrating the output of risk assessment data by body part according to an embodiment of the present invention. FIG. 5 is a diagram illustrating the generation of 3D data from a 2D pressure image according to an embodiment of the present invention. FIG. 6 is a diagram illustrating a method for assessing pressure ulcer risk according to an embodiment of the present invention. Specific details for implementing the invention

[0029] The present invention will be described below with reference to the attached drawings. However, the present invention may be implemented in various different forms and is therefore not limited to the embodiments described herein. Furthermore, in order to clearly explain the present invention in the drawings, parts unrelated to the explanation have been omitted, and similar parts throughout the specification have been given similar reference numerals.

[0030] Throughout the specification, when it is stated that a part is "connected (connected, in contact, combined)" with another part, this includes not only cases where they are "directly connected," but also cases where they are "indirectly connected" with other members interposed between them. Furthermore, when it is stated that a part "includes" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but rather allows for the inclusion of additional components.

[0031] The terms used herein are used merely to describe specific embodiments and are not intended to limit the invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this specification, terms such as “comprising” or “having” are intended to indicate the presence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0032] In this specification, "module" includes a unit composed of hardware, software, or firmware, and may be used interchangeably with terms such as logic, logic block, component, or circuit. A module may be a component formed as a whole, or a minimum unit or part thereof that performs one or more functions. For example, a module may be composed of an application-specific integrated circuit (ASIC).

[0033] FIG. 1 is a diagram illustrating the configuration of a pressure ulcer risk assessment system according to an embodiment of the present invention.

[0034] As illustrated in FIG. 1, the pressure ulcer risk assessment system (100) may include a pressure data receiving unit (110), a posture and body part estimation unit (120), a risk assessment unit by part (130), a risk assessment data output unit by part (140), and a 3D data generation unit (150).

[0035] The pressure data receiving unit (110) can receive pressure data from a pressure sensor (not shown). Here, the pressure data receiving unit (110) can receive pressure data in preset time intervals. For example, the pressure data may be in the form of a 2D image.

[0036] Additionally, the pressure sensor may be configured as a modular unit in which multiple pressure sensors are divided. That is, the pressure sensor can measure the pressure applied to each pressure sensor and transmit the measured pressure value to the pressure data receiving unit (110).

[0037] The posture and body part estimation unit (120) can estimate posture and body parts based on pressure data. At this time, the posture and body part estimation unit (120) may store a posture estimation model and a body part estimation model.

[0038] For example, a pose estimation model can estimate a pose and detect keypoints using pressure data as input. A keypoint refers to a pre-set feature point of the body. The pose estimation model can calculate depth using the pressure intensity of the pressure data. Using the calculated depth, the pose estimation model can detect keypoint coordinates in 3D space and estimate and output the pose of a 2D pressure image.

[0039] Meanwhile, the body part estimation model can output PROI (Pressure Region of Interest) coordinate data using the keypoint coordinates and posture output from the posture estimation model as input. PROI stands for Pressure Region of Interest and refers to the area subject to pressure ulcer risk assessment. Here, the PROI can be configured differently depending on the posture. The body part estimation model can output PROI coordinate data that matches the posture using the keypoint coordinates and posture. For example, a clipping mask can be matched with the activated keypoint coordinates. As will be described later, the body part estimation model can extract PROI coordinate data by projecting the clipping mask onto a 2D pressure image centered on the activated keypoint coordinates. Additionally, the body part estimation model can estimate the body part based on the extracted PROI coordinate data.

[0040] The risk assessment unit (130) for each body part can assess the risk of each body part based on PROI coordinate data extracted from a body part estimation model, statistical data, and training data. Here, the risk assessment unit (130) for each body part can receive statistical data from the statistical data DB (10). For example, the statistical data may include demographic data such as gender, age, vital signs, and time stamps. Additionally, the risk assessment unit (130) for each body part can receive training data from the training data DB (20). The training data may include the risk of pressure ulcers according to the body part.

[0041] Additionally, the risk assessment unit (130) for each body part may assign cumulative weights over time. For example, the risk assessment unit (130) for each body part may assess the risk level for each body part by assigning weights to various parameters such as cumulative time, presence of bedsores, nutritional status, and skin condition.

[0042] The area-specific risk assessment data output unit (140) can receive and output area-specific risk assessment data from the area-specific risk assessment unit (130). For example, the area-specific risk assessment data output unit (140) can output area-specific risk assessment data over time. Here, the area-specific risk assessment data can be classified into grades 1 to 4, which indicate the stage requiring action by a rescuer due to the risk of pressure ulcers. The area-specific risk assessment data can be set such that as the grade number increases, active action by a rescuer is required.

[0043] The 3D data generation unit (150) can generate and provide 2D pressure images as 3D data. For example, the 3D data generation unit (150) can generate 2D pressure images as 3D data by receiving PROI coordinate data from the posture and body part estimation unit (120). Additionally, the 3D data generation unit (150) can receive risk assessment data by body part from the risk assessment unit by body part (130) and reflect the risk assessment data by body part in the 3D data. By doing so, the 3D data generation unit (150) can intuitively provide the risk level by body part of the estimated posture to the person taking action.

[0044] FIG. 2 is a drawing illustrating the operation of a posture and body part estimation unit according to an embodiment of the present invention.

[0045] Referring to FIG. 2, the posture and body part estimation unit may include a posture estimation unit (121) and a body part estimation unit (122). The posture estimation unit (121) may have a posture estimation model stored therein and may output keypoint coordinates and a posture by taking pressure data (P) as input. Here, a keypoint refers to a pre-set feature point of the body. For example, keypoints may include the head, elbow, shoulder, buttocks, and heel. Additionally, the pressure data (P) may be composed of a 2D pressure image. The posture estimation unit (121) may calculate the depth using the pressure intensity of the pressure data (P) to detect keypoint coordinates in 3D space and estimate the posture of the 2D pressure image.

[0046] Additionally, the body part estimation unit (122) can store a body part estimation model. The body part estimation unit (122) can output PROI (Pressure Region of Interest) coordinate data using keypoint coordinates and posture output from the posture estimation unit (121) as input.

[0047] First, the body part estimation unit (122) can activate keypoint (K) coordinates required for each posture. The body part estimation unit (122) can generate a clipping mask centered on the activated keypoint (A) coordinates. Here, the clipping mask is used by cutting out only the desired part of a specific image so that the first layer enters the area of ​​the second layer. The body part estimation unit (122) can project the generated clipping mask onto pressure data (P) to output PROI coordinate data. Based on the output PROI coordinate data, the body part estimation unit (122) can estimate the body part according to the posture from the pressure data (P). At this time, the body part estimation unit (122) can adjust the size of the clipping mask (M) according to the subject's height and weight, etc. That is, by generating a personalized clipping mask (M), the body part estimation unit (122) can identify the pressure area of ​​interest that reflects the subject's physical characteristics.

[0048] FIG. 3 is a diagram illustrating a body part estimation unit according to an embodiment of the present invention outputting PROI coordinate data.

[0049] Referring to FIG. 3, the body part estimator can generate a clipping mask (M) centered on the activated keypoint (A) coordinates. The body part estimator can activate the required keypoint coordinates for each posture. For example, when estimated as lying on one's back, the keypoint coordinates of the head (0), elbows (3, 6), shoulders (2, 5), buttocks (8, 11), and heels (10, 13) may be activated. The body part estimator can generate and match a clipping mask (M) only to the activated keypoint (A).

[0050] In addition, the body part estimation unit can project the generated clipping mask (M) onto the pressure data (P) to output PROI coordinate data. As described above, since the clipping mask is used to cut out only the desired parts, the body part estimation unit can use the clipping mask to extract the pressure region of interest according to the posture from the pressure data (P). Here, the PROI coordinate data may include coordinate data corresponding to the clipping mask region. Thus, the body part estimation unit can estimate the body part according to the posture based on the PROI coordinate data.

[0051] FIG. 4 is a diagram illustrating the output of risk assessment data by part according to an embodiment of the present invention.

[0052] As illustrated in FIG. 4, the risk assessment unit (130) by body part can receive PROI coordinate data, statistical data, and training data. The statistical data may include demographic data such as gender, age, vital signs, and time stamps stored in the statistical data DB. Additionally, the training data may include pressure ulcer risk levels according to body parts stored in the training data DB. Thus, the risk assessment unit (130) by body part can perform a risk assessment according to the subject's conditions.

[0053] The risk assessment unit (130) for each body part can assess the risk level for each body part based on received PROI coordinate data, statistical data, and training data. The risk assessment unit (130) for each body part can use the received PROI coordinate data at specific time intervals. Referring to FIG. 4, the PROI coordinate data may include information about body parts such as the head, hips, left elbow, right elbow, and ankle. That is, the risk assessment unit (130) for each body part can assess the risk level corresponding to the PROI coordinate data over time.

[0054] Additionally, the risk assessment unit (130) for each body part can assign cumulative weights over time. For example, the risk assessment unit (130) for each body part can assign weights to various parameters such as cumulative time, presence of bedsores, nutritional status, and skin condition. By doing so, the risk assessment unit (130) for each body part can perform a risk assessment for each body part more accurately by assigning weights to each body part.

[0055] The area-specific risk assessment data output unit (140) can receive and output area-specific risk assessment data from the area-specific risk assessment unit (130). For example, the area-specific risk assessment data output unit (140) can classify the area-specific risk assessment data into grades 1 to 4, which are the stages requiring action by a person taking measures according to the risk of pressure ulcers. Additionally, the area-specific risk assessment data output unit (140) can provide area-specific risk assessment data over time in a table format. As illustrated in FIG. 4, the output area-specific risk assessment data may display the risk assessment grade over time for each body part.

[0056] Here, Grade 1 indicates a state where the risk of developing pressure ulcers is very high and immediate action by the caregiver is required; Grade 2 indicates a state where the risk of developing pressure ulcers is high and continuous monitoring by the caregiver is required; Grade 3 indicates a state where the risk of developing pressure ulcers is moderate and caution by the caregiver is required; and Grade 4 indicates a state where the risk of developing pressure ulcers is low and relatively safe. In addition, the risk assessment data by body part can be set so that as the grade number increases, active action by the caregiver is required. Furthermore, the risk assessment data output unit (140) by body part can assign a color to each grade so that the caregiver can intuitively understand the risk assessment data.

[0057] FIG. 5 is a drawing illustrating the generation of a 2D pressure image into 3D data according to an embodiment of the present invention.

[0058] The 3D data generation unit (150) can generate and provide input pressure data as 3D data. For example, the 3D data generation unit (150) can generate 2D pressure images as 3D data based on PROI coordinate data containing body part information according to posture. Here, the 3D data generation unit (150) can receive PROI coordinate data corresponding to the 2D image from the posture and part estimation unit. The 3D data generation unit (150) can generate 2D pressure images as 3D data more accurately by reflecting the identified body part information according to posture.

[0059] Additionally, the 3D data generation unit (150) can supplement the generated 3D data using pre-stored posture image data. Since the 2D pressure image is information about the pressure applied to the pressure sensor by the subject's body, parts that are not in contact with the body do not have pressure data measured and are therefore not accurately displayed in the 2D pressure image. Accordingly, the 3D data generation unit (150) can generate 3D data more accurately by reflecting posture image data corresponding to the PROI coordinate data.

[0060] Additionally, the 3D data generation unit (150) can receive risk assessment data by part and reflect the risk assessment data by part in the 3D data. By doing so, the 3D data generation unit (150) can intuitively provide the risk level by part of the estimated posture to the person taking action.

[0061] FIG. 6 is a diagram illustrating a method for evaluating the risk of bedsores according to an embodiment of the present invention.

[0062] In step (S110), the pressure ulcer risk assessment system can receive pressure data from a pressure sensor. The pressure ulcer risk assessment system can receive pressure data in the form of a 2D image at preset time intervals.

[0063] In step (S120), the pressure ulcer risk assessment system can detect keypoint coordinates and estimate the posture by using received pressure data as input. For example, the pressure ulcer risk assessment system can detect keypoint coordinates in 3D space by calculating the depth using the pressure intensity of the pressure data and estimate the posture of the 2D pressure image. Here, a keypoint refers to a pre-set feature point of the body. For example, the keypoint may include feature points of the body such as the head, elbows, shoulders, buttocks, and heels. Thus, the pressure ulcer risk assessment system can estimate the posture by extracting keypoints.

[0064] In step (S130), the pressure ulcer risk assessment system can output PROI (Pressure Region of Interest) coordinate data using keypoint coordinates and posture as inputs. PROI is a pressure region of interest, which refers to the area subject to pressure ulcer risk assessment. Here, PROI can be configured differently for each posture.

[0065] In addition, the pressure ulcer risk assessment system can activate keypoint coordinates required for each position. The keypoint coordinates required for each position can be pre-set. Furthermore, the pressure ulcer risk assessment system can generate and match a clipping mask centered on the activated keypoints. Here, a clipping mask is a method in which the first layer enters the area of ​​the second layer to crop and use only the desired part of a specific image. Therefore, the pressure ulcer risk assessment system can extract a 2D pressure image corresponding to the clipping mask as a pressure region of interest. Additionally, the pressure ulcer risk assessment system can output coordinate data corresponding to the extracted 2D pressure image as PROI coordinate data. At this time, the pressure ulcer risk assessment system can estimate body parts according to the position using the outputted PROI coordinate data.

[0066] In step (S140), the pressure ulcer risk assessment system can assess the risk level for each body part based on extracted PROI coordinate data, statistical data, and training data. Here, the pressure ulcer risk assessment system can receive statistical data from a statistical data DB. Additionally, the pressure ulcer risk assessment system can receive training data from a training data DB. The statistical data DB and the training data DB can be stored in the pressure ulcer risk assessment system. Additionally, the statistical data DB and the training data DB can be stored externally and provide data upon request from the pressure ulcer risk assessment system. For example, the statistical data may include demographic data such as gender, age, vital signs, and timestamps. Additionally, the training data may include the pressure ulcer risk level according to body part.

[0067] In addition, the pressure ulcer risk assessment system can assign cumulative weights over time. For example, the pressure ulcer risk assessment system can assess risk by site by assigning weights to various parameters such as cumulative time, presence of pressure ulcers, nutritional status, and skin condition. Furthermore, the pressure ulcer risk assessment system can output risk assessment data for each site over time. Here, the risk assessment data by site can be classified into grades 1 through 4, indicating the level at which intervention by a rescuer is required due to the risk of pressure ulcer development. The risk assessment data by site can be configured so that higher grade numbers indicate a greater need for active intervention by a rescuer.

[0068] In step (S150), the pressure ulcer risk assessment system can generate and provide 3D data based on pressure data and PROI coordinate data. The pressure ulcer risk assessment system can generate 3D data from 2D pressure images based on PROI coordinate data that includes body part information according to posture. The pressure ulcer risk assessment system can generate 3D data from 2D pressure images more accurately by reflecting the identified body part information according to posture.

[0069] delete

[0070] delete

[0071] delete

[0072] delete

[0073] delete

[0074] The foregoing description of the present invention is for illustrative purposes only, and those skilled in the art will understand that other specific forms can be easily modified without altering the technical spirit or essential features of the present invention. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. For example, each component described as a single unit may be implemented in a distributed manner, and components described as distributed may likewise be implemented in a combined form.

[0075] The scope of the present invention is defined by the claims set forth below, and all modifications or variations derived from the meaning and scope of the claims and equivalent concepts thereof should be interpreted as being included within the scope of the present invention. Explanation of the symbols

[0076] 100: Pressure Ulcer Risk Assessment System 110: Pressure data receiver 120: Estimated posture and body part 130 : Risk Assessment by Body Part 140 : Section-specific risk assessment data output section 150 : 3D Data Generation Unit

Claims

Claim 1 A system for evaluating the risk of pressure ulcers by reflecting the physical characteristics of a subject, comprising: a pressure data receiving unit that receives pressure data, which is information regarding pressure applied to the subject's body from multiple pressure sensors; a posture and body part estimation unit that detects keypoint coordinates, which are preset feature points of the body, based on the pressure data, estimates the posture, outputs PROI (Pressure Region of Interest) coordinate data, which is a pressure region of interest, and estimates the body part using the keypoint coordinates and the posture; a body part risk evaluation unit that evaluates the risk for each body part and generates body part risk evaluation data using learning data including PROI coordinate data for the head, buttocks, left elbow, right elbow, and ankle received at specific time intervals, statistical data including demographic data, and previously stored pressure ulcer risk according to the body part; and a body part risk evaluation data output unit that classifies and outputs the body part risk evaluation data into grades according to the risk of pressure ulcer occurrence; wherein the posture and body part estimation unit includes a posture estimation unit that stores a posture estimation model that takes the pressure data as input and outputs the keypoint coordinates and the posture. A pressure ulcer risk assessment system comprising: a body part estimation unit storing a body part estimation model that outputs PROI coordinate data using the key point coordinates and the posture as inputs; wherein the body part estimation unit activates key point coordinates required for each posture, generates a clipping mask centered on the activated key point coordinates, adjusts the size of the clipping mask according to the subject's body condition, and projects the clipping mask onto the pressure data to extract the PROI coordinate data. Claim 2 delete Claim 3 A pressure ulcer risk assessment system according to claim 1, wherein the posture estimation model calculates depth using the pressure intensity of the pressure data and detects the keypoint coordinates in 3D space using the calculated depth. Claim 4 delete Claim 5 A pressure ulcer risk assessment system according to claim 1, wherein the part-specific risk assessment unit assigns a cumulative weight over time to each body part, and the cumulative weight includes cumulative time, presence of pressure ulcer, nutritional status, and skin condition, and the part-specific risk assessment data output unit provides a risk assessment grade over time for each body part in the form of a table. Claim 6 A pressure ulcer risk assessment system further comprising: a 3D data generation unit according to claim 1, which generates pressure data into 3D data based on PROI coordinate data including body part information according to posture, supplements the 3D data using pre-stored posture image data corresponding to the PROI coordinate data, receives risk assessment data by part, and reflects and displays the risk assessment data by part in the 3D data. Claim 7 A method for evaluating the risk of pressure ulcers by reflecting the physical characteristics of a subject, comprising: a) receiving pressure data, which is information regarding pressure applied from the subject's body from a plurality of pressure sensors; b) inputting the pressure data into a posture estimation model to detect keypoint coordinates, which are preset feature points of the body, and estimating the posture; c) inputting the keypoint coordinates and the posture into a body part estimation model to output PROI coordinate data, which is a pressure region of interest, and estimating the body part; and d) a step of evaluating the risk for each body part and generating risk assessment data for each body part using statistical data including PROI coordinate data for the head, buttocks, left elbow, right elbow, and ankle received at specific time intervals, demographic data, and learning data including a previously stored risk of pressure ulcer according to the body part, and classifying the risk assessment data for each body part into grades according to the risk of pressure ulcer occurrence and outputting them; wherein the step of estimating the body part includes: a step of activating keypoint coordinates required for each posture; a step of generating a clipping mask centered on the activated keypoint coordinates; a step of adjusting the size of the clipping mask according to the subject's body condition; and a step of projecting the clipping mask onto the pressure data to extract the PROI coordinate data; a method for evaluating the risk of pressure ulcers. Claim 8 A method for assessing pressure ulcer risk according to claim 7, wherein step b) calculates a depth using the pressure intensity of the pressure data and detects the keypoint coordinates in 3D space using the calculated depth. Claim 9 delete Claim 10 A method for assessing the risk of a pressure ulcer according to claim 7, wherein step d) outputs risk assessment data for each body part by assigning cumulative weights according to time to each body part, wherein the cumulative weights include at least one of cumulative time, presence of a pressure ulcer, nutritional status, and skin condition, and the step of outputting risk assessment data for each body part is a step of providing a risk assessment grade according to time for each body part in the form of a table. Claim 11 A method for assessing pressure ulcer risk according to claim 7, further comprising: e) a step of generating and providing pressure data as 3D data based on PROI coordinate data including body part information according to posture; f) a step of supplementing the 3D data using pre-stored posture image data corresponding to the PROI coordinate data; and g) a step of receiving the risk assessment data by body part and reflecting the risk assessment data by body part in the 3D data to display it.

Images