Monitoring system, device and computer-implemented method for monitoring pressure sores

Abstract

The present invention provides a monitoring system, a device and a computer-implemented method for monitoring pressure sores. The monitoring system and the device are configured to: capture at least one image corresponding to a user; retrieve the medical history data of the user; and determine a pressure sore condition result corresponding to the at least one image and the medical history data according to a pressure sore prediction model.

Description

Technical Field

[0001] The present invention generally relates to a monitoring system, apparatus and computer-implemented method for monitoring a user, and more particularly to a monitoring system, apparatus and computer-implemented method for monitoring a user for pressure sores. Background Art

[0002] Pressure sores are injuries to the skin and tissues that result from pressure on the skin. Pressure sores can develop on the skin of the body, and usually develop on the skin that covers bony areas of the body.

[0003] In addition, people with medical conditions that can limit a person's ability to change posture or position are at high risk for pressure sores. For example, pressure sores often occur in patients who spend most of their time in bed or a wheelchair.

[0004] Some standard surgical procedures for treating pressure sores are applied to the patient. However, because pressure sores are difficult wounds to heal, and each patient may have a different medical condition, each patient with a pressure sore requires special care and treatment. Summary of the invention

[0005] Some embodiments of the present invention provide a computer-implemented method for monitoring pressure sores. The computer-implemented method for monitoring pressure sores includes: capturing at least one image corresponding to a user; retrieving medical history data of the user; and determining a pressure sore condition result corresponding to the at least one image and the medical history data according to a pressure sore prediction model.

[0006] Some embodiments of the present invention provide a computer monitoring system for monitoring pressure sores. The monitoring system includes an image capture unit, a processor, and a storage unit. The image capture unit captures at least one image corresponding to a user. The storage unit stores the user's medical record data, a pressure sore prediction model, and a program. When the program is executed, the program causes the processor to: retrieve the at least one image from the image capture unit; retrieve the medical record data from the storage unit; and determine a pressure sore condition result corresponding to the at least one image and the medical record data according to the pressure sore prediction model.

[0007] Some embodiments of the present invention provide a device for monitoring pressure sores. The device includes an image capture unit, a processor, a storage unit, and a network interface. The image capture unit captures at least one image corresponding to a user. The storage unit stores a pressure sore prediction model and a program. When the program is executed, the program causes the processor to: retrieve the at least one image from the image capture unit; retrieve medical record data from the network database via the network interface; and determine a pressure sore condition result corresponding to the at least one image and the medical record data according to the pressure sore prediction model.

[0008] The foregoing has been broadly summarized the features and technical advantages of the present invention so that the specific embodiments of the present invention below can be better understood. Additional features and advantages of the present invention will be described below and form the goal of the technical solution of the present invention. It should be understood by those skilled in the art that the disclosed concepts and specific embodiments can be easily used as the basis for modifying or designing other structures or processes for performing the same purpose of the present invention. It should also be recognized by those skilled in the art that such equivalent constructions do not depart from the spirit and scope of the present invention as stated in the appended claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] When read in conjunction with the accompanying drawings, aspects of the present invention are best understood from the following detailed description. It should be noted that, in accordance with standard practice in the industry, various components are not drawn to scale. In fact, the sizes of various components may be arbitrarily increased or reduced for clarity of discussion.

[0010] A more complete understanding of the present invention may be derived by referring to the detailed description and claims considered in conjunction with the accompanying drawings, wherein like reference numerals refer to like elements throughout the figures.

[0011] Figure 1 is a block diagram of a monitoring system according to some embodiments of the present invention.

[0012] Figure 2 is a block diagram of a monitoring system according to some embodiments of the present invention.

[0013] Figure 3 is a block diagram of a monitoring system according to some embodiments of the present invention.

[0014] Figure 4 is a block diagram of an apparatus according to some embodiments of the present invention.

[0015] Figure 5 is a flow chart of a computer-implemented method for monitoring pressure sores according to some embodiments of the present invention.

[0016] Figures 6A to 6C is a flow chart of a computer-implemented method for monitoring pressure sores according to some embodiments of the present invention.

[0017] Figure 7 is a flow chart of a computer-implemented method for monitoring pressure sores according to some embodiments of the present invention. DETAILED DESCRIPTION

[0018] Specific language is now used to describe the embodiments or examples of the present invention illustrated in the accompanying drawings. It should be understood that it is not intended to limit the scope of the present invention. Any changes or modifications to the described embodiments and any further applications of the principles described in the present invention will be considered as generally thought of by a person skilled in the art associated with the present invention. Component symbols may be repeated across embodiments, but this does not necessarily mean that the (several) features of one embodiment are applicable to another embodiment, even if they share the same component symbols.

[0019] It should be understood that although the terms first, second, third, etc. may be used herein to describe various elements, components, regions, layers, or sections, these elements, components, regions, layers, or sections are not limited by such terms. In fact, these terms are only used to distinguish elements, components, regions, layers, or sections from each other. Therefore, without departing from the teachings of the present invention, the first element, component, region, layer, or section discussed below may be referred to as the second element, component, region, layer, or section.

[0020] The terms used herein are only used to describe specific example embodiments and are not intended to limit the present invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should be further understood that, as used in the specification, the term "comprising" specifically refers to the presence of stated features, integers, steps, operations, elements or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components or groups thereof.

[0021] Pressure sores are difficult wounds to heal. Some standard operating procedures (SOPs) for treating pressure sores are developed and applied to patients with pressure sores. However, since each patient may have a different medical condition, each patient with a pressure sore requires a customized treatment to care for his / her pressure sore.

[0022] Figure 1 A block diagram illustrating a monitoring system 1 according to some embodiments of the present invention. The monitoring system 1 includes an image capture unit 11, a processor 13, and a storage unit 15. In some embodiments, the image capture unit 11, the processor 13, and the storage unit 15 are electrically coupled via a communication bus 17. The communication bus 17 may allow the image capture unit 11 to transmit image data to the processor 13, and may allow the processor 13 to execute a computer program PG stored in the storage unit 15.

[0023] The processor 13, the storage unit 15 and the computer program PG may be cooperatively configured to perform the functions of the computer program PG. In other words, when executed, the computer program PG may generate one or more interrupts (e.g., software interrupts) to cause the processor 13 and the storage unit 15 to perform the functions of the computer program PG for monitoring the user's pressure sores. The functions of the computer program PG will be further described below.

[0024] In some embodiments, when monitoring of a user's pressure sores is desired, the image capture unit 11 is positioned adjacent to the user for capturing at least one image 110 of the user. After capturing at least one image 110 of the user, the image capture unit 11 transmits the at least one image 110 to the processor 13.

[0025] In other words, the processor 13 fetches at least one image 110 from the image capturing unit 11. At the same time, the processor 13 fetches the medical record data AD stored in the storage unit 15 corresponding to the user.

[0026] Then, in some embodiments, the pressure sore prediction model BM stored in the storage unit 15 is a machine learning model for receiving image data and medical history data corresponding to the user and then predicting changes in the user's pressure sore condition. Therefore, the processor 13 determines a pressure sore condition result 130 corresponding to at least one image 110 and medical history data AD for the user according to the pressure sore prediction model BM.

[0027] In some embodiments, the pressure sore condition result 130 may include different levels for indicating different predictions of the user's pressure sore condition. For example, the pressure sore condition result 130 includes n levels designated as level "1" to level "n". When the pressure sore condition result 130 is determined as level "1", it means that the user's pressure sore condition will get better. When the pressure sore condition result 130 is determined as level "n", it means that the user's pressure sore condition will get worse.

[0028] Therefore, the caregiver who cares for the user may treat the user's pressure sore based on the pressure sore status result 130. For example, the frequency of wound dressing may depend on the pressure sore status result 130. When the pressure sore status result 130 is determined to be level "1", the frequency of wound dressing may be reduced. When the pressure sore status result 130 is determined to be level "n", the frequency of wound dressing should be increased.

[0029] In some embodiments, the pressure sore prediction model BM may include two stages for processing data and generating results. Specifically, in the first stage, images corresponding to the user are classified according to different movement types corresponding to the user. In the second stage, the pressure sore condition result of the user is determined based on the movement type and medical history data of the user.

[0030] Specifically, after the image capturing unit 11 captures at least one image 110 corresponding to the user, the processor 13 determines the movement type corresponding to the at least one image 110 according to the first stage of the pressure ulcer prediction model BM.

[0031] More specifically, the processor 13 takes out at least one image 110 and determines the movement type through the pressure sore prediction model BM. In other words, the processor 13 inputs at least one image 110 into the first stage of the pressure sore prediction model BM to derive the movement type as output data from the first stage of the pressure sore prediction model BM.

[0032] Next, the processor 13 determines the pressure sore condition result 130 corresponding to the movement type and the medical record data AD according to the second stage of the pressure sore prediction model BM. In other words, the processor 13 inputs the movement type and the medical record data AD into the second stage of the pressure sore prediction model BM to derive the pressure sore condition result 130 as output data from the second stage of the pressure sore prediction model BM.

[0033] In some embodiments, before determining the pressure sore condition result 130 , the movement frequency corresponding to the user may be introduced for a more accurate result. In detail, the processor 13 determines the movement frequency according to at least one image 110 corresponding to the user.

[0034] Next, according to the second stage of the pressure sore condition model BM, the processor 13 determines the pressure sore condition result 130 corresponding to the movement frequency, movement type and medical record data AD. In other words, the processor 13 inputs the movement frequency, movement type and medical record data AD into the second stage of the pressure sore prediction model BM to derive the pressure sore condition result 130 as output data from the second stage of the pressure sore prediction model BM.

[0035] Before being put into use, the machine learning model of the present invention may be first generated and trained. Specifically, in some embodiments, the first stage of the pressure sore prediction model BM is configured to convert an image corresponding to a user into information corresponding to the user's movement type, so that the first stage of the pressure sore prediction model BM is trained using a plurality of first data including images and movement types.

[0036] Specifically, each first data includes first training input data and first training output data. The first training input data includes at least one training image. The first training output data includes a movement type corresponding to the at least one training image.

[0037] Therefore, the plurality of first data are used to train the first stage of the pressure sore prediction model BM to identify the image as a moving type. In other words, the plurality of first data can be used to train the first stage of the pressure sore prediction model BM to convert the image into a moving type.

[0038] For example, the training image corresponding to the user may be marked as a training movement type, which includes turning over, feeding, changing the position of an intravenous drip, massaging, cleaning a wound, changing medical items, and the like.

[0039] Therefore, after training using training images and training movement types, the first stage of the pressure ulcer prediction model BM can classify some captured images corresponding to the user according to different movement types, including turning over, feeding, changing the position of intravenous drip, massage, wound cleaning, changing medical items, etc.

[0040] In summary, the first stage of the pressure sore prediction model BM is a trained machine learning scheme, data is input to the first stage of the pressure sore prediction model BM, and then, another data is output from the first stage of the pressure sore prediction model BM based on the input data. In some embodiments, at least one image corresponding to the user is given as input data of the first stage of the pressure sore prediction model BM, and the subsequent output data is a movement type.

[0041] It should be noted that in some embodiments, the first stage of the pressure ulcer prediction model BM can be established based on the You Only Look Once version 3 (YOLOv3) algorithm, the Single Shot multiBox Detection (SSD) algorithm, or the Region with Convolutional Neural Network (R-CNN) algorithm.

[0042] How to establish a machine learning scheme based on the YOLOv3 algorithm, SSD algorithm or R-CNN algorithm through images and movement types (i.e., the first stage of the pressure ulcer prediction model BM in some embodiments) should be understood by technicians in the field based on the above disclosure, and therefore will not be further described in this article.

[0043] In some embodiments, the second stage of the pressure sore prediction model BM is configured to convert the movement type, movement frequency and medical history data corresponding to the user into a pressure sore condition result, so that the second stage of the pressure sore prediction model BM is trained using multiple second data including movement type, movement frequency and medical history data.

[0044] Specifically, each second data includes second training input data and second training output data. The second training input data includes training movement type, training movement frequency and training medical history data. The second training output data includes training pressure sore condition results, and the training pressure sore condition results correspond to the training movement type, training movement frequency and training medical history data.

[0045] Therefore, the plurality of second data are used to train the second stage of the pressure sore prediction model BM to determine the pressure sore condition result according to the movement type, movement frequency and medical history data. In other words, the second stage of the pressure sore prediction model BM can be trained with the plurality of second data to convert the movement type, movement frequency and medical history data into the pressure sore condition result.

[0046] In some embodiments, each training pressure ulcer condition result may be judged by an experienced doctor based on the corresponding training movement type, training movement frequency, and medical history data.

[0047] In some embodiments, the medical record data is in text format and includes factors determined to be imageable pressure ulcer conditions. For example, the medical record data includes at least one of disease type information, prescription information, medical test information, blood pressure information, body temperature information, diagnosis information, medication information, laboratory test information, etc.

[0048] In summary, the second stage of the pressure sore prediction model BM is a trained machine learning scheme, data is input into the second stage of the pressure sore prediction model BM, and then, another data is output from the second stage of the pressure sore prediction model BM based on the input data. In some embodiments, the movement type, movement frequency, and medical history data corresponding to the user are given as input data of the second stage of the pressure sore prediction model BM, and the subsequent output data is the pressure sore condition result.

[0049] It should be noted that in some embodiments, the second stage of the pressure sore prediction model BM can be established according to a deep neural network (DNN) algorithm. How to establish a machine learning scheme (i.e., the second stage of the pressure sore prediction model BM in some embodiments) based on the DNN algorithm through movement type, movement frequency, medical history data, and pressure sore condition results should be understood by those skilled in the art based on the above disclosure, and therefore will not be further described herein.

[0050] In some embodiments, the pressure sore prediction model BM may be a self-learning machine learning model. Specifically, input data (e.g., images corresponding to the user) and output data (e.g., pressure sore condition results based on these images corresponding to the user) from the pressure sore prediction model BM may be used as training input data and training output data, respectively, for training the pressure sore prediction model BM.

[0051] Figure 2 A block diagram illustrating a monitoring system 1 according to some embodiments of the present invention. In some embodiments, the monitoring system 1 further comprises an indicator 18. After the processor 13 determines the pressure ulcer condition result 130, the processor 13 transmits a signal 132 to trigger the indicator 18 for informing the caregiver of the pressure ulcer condition result 130 so that the caregiver can apply the user's customized plan.

[0052] In some embodiments, the indicator 18 may include a display. The signal that triggers the display may include text information and the display shows the text information. In some embodiments, the indicator 18 may include a light that emits different colors of light to indicate different conditions. For example, red light means that the pressure sore condition will get worse, and green light means that the pressure sore condition will get better.

[0053] Figure 3 Block diagram illustrating monitoring system 1 according to some embodiments of the present invention. In some embodiments, monitoring system 1 further comprises device 19. After processor 13 determines pressure ulcer condition result 130, processor 13 transmits control signal 134 to control device 19 for adjusting the position of the user.

[0054] In some embodiments, the device 19 may include an electric mattress capable of adjusting the position of the user. The control signal 134 may be used to control the angle or inflation state of the electric mattress.

[0055] Figure 4 A block diagram illustrating a device 4 according to some embodiments of the present invention. In some embodiments, the device 4 includes an image capture unit 41, a processor 43, a storage unit 45, and a network interface 49. In some embodiments, the image capture unit 41, the processor 43, the storage unit 45, and the network interface 49 are electrically coupled by a communication bus 47. The communication bus 47 may allow the image capture unit 41 to transfer image data to the processor 43, and may allow the processor 43 to execute a computer program PG stored in the storage unit 45.

[0056] The processor 43, the storage unit 45 and the computer program PG may be configured cooperatively to execute the functions of the computer program. In other words, when executed, the computer program PG may generate one or more interrupts (e.g., software interrupts) to cause the processor 43 and the storage unit 45 to execute the functions of the computer program PG for monitoring the user's pressure sores. The functions of the computer program PG will be further described below.

[0057] In some embodiments, the device 4 may be positioned adjacent to the user, and the image capture unit 41 may capture at least one image 410 of the user. After capturing at least one image 410, the image capture unit 41 transmits the at least one image 410 to the processor. In other words, the processor 43 takes out at least one image 410 from the image capture unit 41.

[0058] In some embodiments, the processor 43 of the device 4 retrieves the medical record data AD from the network database NB via the network interface 49. Then, the processor 43 determines the pressure sore condition result 430 corresponding to the at least one image 410 and the medical record data AD according to the pressure sore prediction model BM stored in the storage unit 45.

[0059] It should be noted that in some embodiments, the apparatus 4 may be an image capture device having higher computing power, so that the image capture device can monitor the user to determine the pressure ulcer condition result 430 while retrieving the medical record data AD corresponding to the user from the network database NB.

[0060] Some embodiments of the present invention include a computer-implemented method for monitoring pressure ulcers, and wherein Figure 5 The flowchart of the method is shown in . The computer-implemented method of some embodiments is used in a monitoring system (such as the monitoring system 1 of the aforementioned embodiment). The detailed steps of the computer-implemented method are described below.

[0061] Step S501 is executed by the image capture unit of the monitoring system to capture at least one image corresponding to the user. Step S502 is executed by the processor of the monitoring system to retrieve the medical record data of the user. The medical record data is stored in the storage unit of the monitoring system.

[0062] In some embodiments, the pressure sore prediction model stored in the storage unit is a trained machine learning scheme: image data and medical history data corresponding to the user are input into the pressure sore prediction model, and then, the pressure sore condition result is output from the pressure sore prediction model based on the input data.

[0063] Therefore, the processor executes step S503 to determine a pressure sore condition result corresponding to the at least one image and the medical record data according to the pressure sore prediction model. The pressure sore condition result may include different levels for indicating different predictions of the pressure sore condition of the user.

[0064] Some embodiments of the present invention include a computer-implemented method for monitoring pressure ulcers, and wherein Figures 6A to 6C The flowchart of the method is shown in . The computer-implemented method of some embodiments is used in a monitoring system (such as the monitoring system 1 of the aforementioned embodiment). The detailed steps of the computer-implemented method are described below.

[0065] In some embodiments, before being put into use, a pressure sore prediction model may be first generated and trained. Step S601 is executed by a processor of the monitoring system to generate a pressure sore prediction model according to a machine learning scheme of training data having a plurality of first data and a plurality of second data.

[0066] In some embodiments, each first data includes first training input data and first training output data. The first training input data includes a plurality of training images. The first training output data includes a plurality of training movement types corresponding to the training images.

[0067] During the generation of the pressure sore prediction model, the training images and training movement types are used in the first stage of training the pressure sore prediction model to identify the images as movement types. In some embodiments, the movement types may include turning over, feeding, changing the position of an IV drip, massaging, wound cleaning, or changing medical items.

[0068] In some embodiments, each second data includes second training input data and second training output data. The second training input data includes a plurality of training movement types, a plurality of training movement frequencies, and a plurality of training medical history data. The second training output data includes a plurality of training pressure ulcer condition results.

[0069] During the generation of the pressure sore prediction model, the training movement type, training movement frequency, training medical record data and training pressure sore condition results are used in the second stage of training the pressure sore prediction model to determine the pressure sore condition results based on the movement type, movement frequency and medical record data.

[0070] After the pressure sore prediction model is generated, the processor executes step S602 to store the pressure sore prediction model in a storage unit of the monitoring system for later use.

[0071] When it is necessary to monitor the user, the image capture unit executes step S603 to capture at least one image corresponding to the user. The processor executes step S604 to determine the movement frequency according to the at least one image corresponding to the user.

[0072] The processor executes step S605 to retrieve the user's medical record data. The medical record data is stored in a storage unit of the monitoring system. In some embodiments, the medical record data is in text format and includes at least one of disease type information, prescription information, medical test information, blood pressure information, and body temperature information.

[0073] The processor executes step S606 to determine the movement type corresponding to the at least one image according to the pressure sore prediction model. The processor executes step S607 to determine the pressure sore condition result corresponding to the movement type, movement frequency and medical record data according to the pressure sore prediction model.

[0074] In some embodiments, the processor optionally performs step S608 to transmit a signal to trigger an indicator according to the pressure sore condition result. In some embodiments, the processor optionally performs step S609 to transmit a control signal to control the device according to the pressure sore condition result.

[0075] Some embodiments of the present invention include a computer-implemented method for monitoring pressure ulcers, and wherein Figure 7 The flowchart of the method is shown in . The computer-implemented method of some embodiments is used in a device (such as the device 4 of the above-mentioned embodiment). The detailed steps of the computer-implemented method are described below.

[0076] Step S701 is executed by the image capture unit of the device to capture at least one image corresponding to the user. Step S702 is executed by the processor of the device to retrieve the medical record data of the user via the network interface of the device. The medical record data is stored in the network database.

[0077] In some embodiments, the pressure sore prediction model stored in the storage unit is a trained machine learning scheme: image data and medical history data corresponding to the user are input into the pressure sore prediction model, and then, the pressure sore condition result is output from the pressure sore prediction model based on the input data.

[0078] Therefore, the processor executes step S703 to determine a pressure sore condition result corresponding to the at least one image and the medical record data according to the pressure sore prediction model. The pressure sore condition result may include different levels for indicating different predictions of the pressure sore condition of the user.

[0079] In particular, it should be understood that the processor mentioned in the above embodiments may be a central processing unit (CPU), other hardware circuit components capable of executing relevant instructions, or a combination of computing circuits known to those skilled in the art based on the above disclosure.

[0080] Furthermore, the storage unit mentioned in the above embodiments may include a memory (e.g., ROM, RAM, etc.) or a storage device (e.g., flash memory, HDD, SSD, etc.) for storing data. In addition, the communication bus mentioned in the above embodiments may include a communication interface (or I / O interface) for transmitting data between elements such as an image capture unit, a processor, a storage unit, an indicator, and a device, and may include an electrical bus interface, an optical bus interface, or even a wireless bus interface. However, this description is not intended to limit the hardware implementation embodiment of the present invention.

[0081] Although the present invention and its advantages have been described in detail, it should be understood that various changes, substitutions and modifications may be made herein without departing from the spirit and scope of the present invention as defined by the appended claims. For example, many of the processes discussed above may be implemented in different methodologies and replaced by other processes or combinations thereof.

[0082] Furthermore, the scope of the present application is not intended to be limited to the specific embodiments of the processes, machines, manufactures, compositions of matter, components, methods, and steps described in the specification. A person skilled in the art will readily understand from the disclosure of the present invention that currently existing or later developed processes, machines, manufactures, compositions of matter, components, methods, or steps that perform substantially the same functions or achieve substantially the same results as the corresponding embodiments described herein may be utilized in accordance with the present invention. Therefore, the appended claims intend to include such processes, machines, manufactures, compositions of matter, components, methods, or steps within their scope.

[0083] Explanation of symbols

[0084] 1: Monitoring system

[0085] 4: Installation

[0086] 11: Image capture unit

[0087] 13: Processor

[0088] 15: Storage Unit

[0089] 17: Communication bus

[0090] 18: Indicator

[0091] 19: Installation

[0092] 41: Image capture unit

[0093] 43: Processor

[0094] 45: Storage unit

[0095] 47: Communication bus

[0096] 49: Network Interface

[0097] 110: Image

[0098] 130: Pressure ulcer status results

[0099] 132: Signal

[0100] 134: Control signal

[0101] 410: Image

[0102] 430: Pressure ulcer status results

[0103] S501: Steps

[0104] S502: Step

[0105] S503: Step

[0106] S601: Steps

[0107] S602: Step

[0108] S603: Step

[0109] S604: Steps

[0110] S605: Step

[0111] S606: Steps

[0112] S607: Steps

[0113] S608: Steps

[0114] S609: Steps

[0115] S701: Steps

[0116] S702: Steps

[0117] S703: Steps

[0118] AD: Medical record data

[0119] BM:Pressure Ulcer Prediction Model

[0120] NB: Network database

[0121] PG: Computer Programs

Claims

1. A computer-implemented method for monitoring pressure ulcers, comprising: capturing a plurality of images corresponding to a user; determining a movement frequency based on the plurality of images corresponding to the user; Retrieving the medical record data of the user; and Determine the pressure sore condition results corresponding to the multiple images and the medical record data according to the pressure sore prediction model, Wherein determining the pressure sore condition result further comprises: determining a movement type corresponding to the plurality of images according to the pressure sore prediction model; and The pressure sore condition result corresponding to the movement type, the movement frequency and the medical record data is determined according to the pressure sore prediction model.

2. The computer-implemented method of claim 1 , further comprising: Generate the pressure sore prediction model according to a machine learning scheme of training data having a plurality of first data and a plurality of second data; in: Each first data includes first training input data and first training output data, wherein the first training input data includes a plurality of training images, and the first training output data includes a plurality of training movement types; Each second data includes second training input data and second training output data, wherein the second training input data includes the plurality of training movement types, the plurality of training movement frequencies and the plurality of training medical history data, and The second training output data includes a plurality of training pressure ulcer condition results.

3. The computer-implemented method of claim 1, wherein the movement type comprises turning over, feeding, changing the position of an intravenous drip, massaging, wound cleaning, or changing medical items. 4 . The computer-implemented method of claim 1 , wherein the medical record data has a text format and includes at least one of disease type information, prescription information, medical test information, blood pressure information, and body temperature information.

5. The computer-implemented method of claim 1 , further comprising: A signal is transmitted based on the pressure sore condition result to trigger an indicator.

6. The computer-implemented method of claim 1 , further comprising: A control signal is transmitted according to the pressure sore condition result to control the device.

7. A monitoring system for monitoring pressure sores, comprising: an image capturing unit for capturing a plurality of images corresponding to a user; processor; and A storage unit, which is used to store the user's medical record data, the pressure sore prediction model and a program, wherein when the program is executed, the processor: fetching the plurality of images from the image capture unit; determining a movement frequency corresponding to the plurality of images; Retrieving the medical record data from the storage unit; and Determine the pressure sore condition results corresponding to the multiple images and the medical record data according to the pressure sore prediction model, Wherein determining the pressure sore condition result further comprises: determining a movement type corresponding to the plurality of images according to the pressure sore prediction model; and The pressure sore condition result corresponding to the movement type, the movement frequency and the medical record data is determined according to the pressure sore prediction model.

8. The monitoring system of claim 7, wherein the program, when executed, further causes the processor to: Generate the pressure sore prediction model according to a machine learning scheme of training data having a plurality of first data and a plurality of second data; in: Each first data includes first training input data and first training output data, wherein the first training input data includes a plurality of training images, and the first training output data includes a plurality of training movement types; Each second data includes second training input data and second training output data, wherein the second training input data includes the plurality of training movement types, the plurality of training movement frequencies and the plurality of training medical history data, and The second training output data includes a plurality of training pressure ulcer condition results.

9. The monitoring system according to claim 7, wherein the movement type includes turning over, feeding, changing the position of an intravenous drip, massaging, wound cleaning, or changing medical items.

10. The monitoring system of claim 7, wherein the medical history data has a text format and includes at least one of disease type information, prescription information, medical test information, blood pressure information, and body temperature information.

11. The monitoring system of claim 7, further comprising an indicator, wherein the program, when executed, further causes the processor to transmit a signal to trigger the indicator according to the pressure ulcer condition result.

12. The monitoring system of claim 7, further comprising a device, wherein the program, when executed, further causes the processor to transmit a control signal to control the device according to the pressure sore condition result.

13. A monitoring device for monitoring pressure sores, comprising: A network interface, which is used to connect to a network database; an image capturing unit for capturing a plurality of images corresponding to a user; processor; and A storage unit, which is used to store a pressure sore prediction model and a program, wherein when the program is executed, the processor: fetching the plurality of images from the image capture unit; determining a movement frequency corresponding to the plurality of images; Retrieving medical record data from the network database via the network interface; and Determine the pressure sore condition results corresponding to the multiple images and the medical record data according to the pressure sore prediction model, Wherein determining the pressure sore condition result further comprises: determining a movement type corresponding to the plurality of images according to the pressure sore prediction model; and The pressure sore condition result corresponding to the movement type, the movement frequency and the medical record data is determined according to the pressure sore prediction model.

Images