Pressure injury early warning method and device and computer equipment

Abstract

The application relates to the technical field of medical treatment and discloses a pressure injury early warning method and device and computer equipment, the method comprising the following steps: for any medical object, obtaining a plurality of air chamber units required by the medical object for surgery; determining a plurality of groups of pressure early warning data and a plurality of groups of temperature early warning data of the medical object based on a threshold value determination model, wherein the threshold value determination model is obtained based on the relationship between pressure data, temperature data of the required air chamber units of a plurality of sample medical objects and whether pressure injury occurs; and based on the plurality of groups of pressure early warning data and the plurality of groups of temperature early warning data, adjusting the plurality of air chamber units to early warn the pressure injury. The application solves the problem that the medical object is not checked in time by the medical staff, leading to the occurrence of pressure injury, realizes timely early warning of the pressure injury, and guarantees the health of the medical object.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical treatment, in particular to a pressure injury early warning method and device and a computer device. BACKGROUND

[0002] Intraoperative pressure injury, also known as surgery-related pressure sores, accounts for a large proportion of surgical adverse events, and has a significant negative impact on medical objects and medical systems. For example, pressure injury can affect the health of medical objects, increase the treatment cost of medical objects and medical systems, and prolong the hospitalization time of medical objects. Therefore, how to timely early warn pressure injury during surgery is a problem to be solved.

[0003] Currently, the inspection massage of the pressure part is usually performed at regular intervals, the surgical position is slightly changed, or a soft pad is added to the pressure part. The problem of the above scheme is that in the actual surgical process, the medical care object is distracted due to long-time work or huge work pressure, and may forget to regularly check the skin of the pressure part of the medical object or adjust the position, so that the pressure injury cannot be timely early warned. SUMMARY

[0004] Therefore, the present application provides a pressure injury early warning method to solve the problem that the medical care object forgets to regularly check the pressure part of the medical object, thereby causing pressure injury.

[0005] In a first aspect, the present application provides a pressure injury early warning method, which comprises:

[0006] For any medical object, a plurality of gas chamber units required by the medical object for surgery are obtained;

[0007] Based on a threshold determination model, a plurality of sets of pressure early warning data and a plurality of sets of temperature early warning data of the medical object are determined, and the threshold determination model is trained based on the relationship between the pressure data, the temperature data of the required gas chamber units of a plurality of sample medical objects, and whether the pressure injury occurs;

[0008] Based on the plurality of sets of pressure early warning data and the plurality of sets of temperature early warning data, the plurality of gas chamber units are adjusted to early warn the pressure injury.

[0009] The pressure injury early warning method provided by the present application adjusts the gas chamber units used by the medical object based on the pressure early warning data and the temperature early warning data of the medical object, solves the problem that the manual inspection of the medical care object on the medical object may not be timely, thereby causing pressure injury, realizes timely early warning of the pressure injury, protects the health of the medical object, and avoids adjustment of the overall air cushion device, reduces interference with the surgical process.

[0010] In an optional implementation, each set of pressure warning data comprises a pressure warning threshold and a pressure duration threshold, and each set of temperature warning data comprises a temperature warning threshold and a temperature duration threshold; based on the multiple sets of pressure warning data and the multiple sets of temperature warning data, the multiple air chamber units are adjusted to warn of pressure injury, comprising:

[0011] For any set of pressure warning data, a first duration of the air chamber unit under the pressure warning threshold corresponding to the pressure warning data is determined;

[0012] For any set of temperature warning data, a second duration of the air chamber unit under the temperature warning threshold corresponding to the temperature warning data is determined;

[0013] In a case where the first duration is not less than the pressure duration threshold, a deflation operation is performed on the air chamber unit to warn of pressure injury;

[0014] In a case where the second duration is not less than the temperature duration threshold, a cooling operation is performed on the air chamber unit to warn of pressure injury.

[0015] The pressure injury warning method provided by the embodiments of the present application can perform deflation or cooling operation on the air chamber unit based on the duration, effectively warn of pressure injury, and protect the health of the medical object, because the medical object remaining still under the pressure warning threshold or the temperature warning threshold will cause the occurrence of pressure injury.

[0016] In an optional implementation, the training process of the threshold determination model comprises:

[0017] Obtain body position data, preoperative diagnosis information and sample medical results of multiple sample medical objects, and the sample medical results indicate whether the sample medical objects have pressure injury;

[0018] For any sample medical object, obtain sample pressure data and sample temperature data of multiple air chamber units used by the sample medical object during the operation;

[0019] Discretize the sample pressure data of the air chamber unit to obtain multiple sets of sample pressure warning data, and the sample pressure warning data comprises a sample pressure warning threshold and a sample pressure duration threshold;

[0020] Discretize the sample temperature data of the multiple air chamber units to obtain multiple sets of sample temperature warning data, and the sample temperature warning data comprises a sample temperature warning threshold and a sample temperature duration threshold;

[0021] The body position data of a sample medical object, preoperative diagnosis information and sample medical results are taken as inputs, and multiple sets of sample pressure warning data and multiple sets of sample temperature warning data are taken as labels to train the threshold determination model.

[0022] The pressure injury warning method provided by the embodiment of the application trains the threshold determination model based on the body position data of a sample medical object, preoperative diagnosis information, sample pressure warning data and sample temperature warning data, so that the trained threshold determination model can determine pressure warning data and temperature warning data of different medical objects in different body positions, effectively avoiding the occurrence of pressure injury.

[0023] In an optional implementation, the method further includes:

[0024] Multiple frames of panoramic pictures within a preset range of the medical object and air chamber pressure data of the medical object are acquired, and the air chamber pressure data represents pressure data of an air chamber unit used by the medical object during a surgery process;

[0025] The first frame of panoramic pictures is recognized based on an object detection model, and a circulating nurse object, a surgical bed and an operation computer are determined respectively, the object detection model being trained based on multiple sample panoramic pictures and being used to identify objects and objects in a picture;

[0026] The surgery process of the medical object is monitored based on the circulating nurse object, the surgical bed and the air chamber pressure data of the medical object.

[0027] The pressure injury warning method provided by the embodiment of the application is helpful to avoid pressure injury due to medical examination of the circulating nurse object on the medical object, so that the panoramic pictures in the surgery process are recognized based on the object detection model to determine the circulating nurse object, thereby the circulating nurse object and the surgery process can be monitored to avoid the occurrence of pressure injury.

[0028] In an optional implementation, the training process of the object detection model includes:

[0029] Multiple sample panoramic pictures are acquired;

[0030] The multiple sample panoramic pictures are recognized based on the object detection model to obtain sample labeled positions of multiple sample objects, a sample surgical bed and a sample operation computer respectively;

[0031] The object detection model is trained based on differences between the sample labeled positions of the multiple sample objects, the sample surgical bed and the sample operation computer and actual positions.

[0032] The pressure injury early warning method provided by the embodiment of the application can obtain sample object, sample operating table and sample operation computer sample annotation positions through the recognition of the sample panoramic picture by the object detection model, so that the object detection model can be trained based on the difference between the sample annotation positions and actual positions, the object detection model can accurately recognize all objects and objects in the picture, and the accuracy of monitoring the circulating nurse object and the operation process is ensured.

[0033] In an optional implementation, the first frame panoramic picture is recognized based on the object detection model, and the circulating nurse object, the operating table and the operation computer are determined respectively, including:

[0034] The first frame panoramic picture is recognized based on the object detection model, and the plurality of first objects, the operating table and the operation computer are determined;

[0035] The first center coordinate of the operation computer is determined;

[0036] The second center coordinates corresponding to the plurality of first objects are determined respectively;

[0037] For any first object, the distance between the second center coordinate of the first object and the first center coordinate is determined;

[0038] The first object corresponding to the smallest distance among the distances corresponding to the plurality of first objects is determined as the circulating nurse object.

[0039] The pressure injury early warning method provided by the embodiment of the application ensures the accuracy of determining the circulating nurse object because there are a plurality of first objects in the operation process, and the circulating nurse object is determined through the distance between the first object and the operation computer.

[0040] In an optional implementation, after the first frame panoramic picture is recognized based on the object detection model, and the circulating nurse object, the operating table and the operation computer are determined respectively, the method further includes:

[0041] The next frame panoramic picture of the first frame panoramic picture is recognized based on the object detection model, and a plurality of second objects are determined;

[0042] The feature of the circulating nurse object is extracted, and a first feature vector of the circulating nurse object is obtained;

[0043] The features of the plurality of second objects are extracted respectively, and second feature vectors corresponding to the plurality of second objects are obtained;

[0044] For any second feature vector, the similarity value of the second feature vector and the first feature vector is determined based on the cosine similarity algorithm;

[0045] The second object corresponding to the second feature vector with the largest similarity value is determined as a new circulating nurse object.

[0046] The pressure injury early warning method provided by the embodiment of the application can ensure the accuracy of determining the circulating nurse object by taking the second object most similar to the feature vector of the circulating nurse object in the next frame panoramic picture as the new circulating nurse object, because the position of the circulating nurse object during the operation process is subject to movement.

[0047] In an optional implementation, the operation process of the medical object is monitored based on the air chamber pressure data of the circulating nurse object, the operating table and the medical object, and the method comprises the following steps of:

[0048] Obtaining the surrounding coordinates of the operating table, wherein the surrounding coordinates comprise the upper-left corner coordinate, the upper-right corner coordinate, the lower-left corner coordinate and the lower-right corner coordinate of the operating table;

[0049] If the coordinates of the circulating nurse object appear in a first preset range of the surrounding coordinates within a preset time period, it is determined that the circulating nurse object appears in a second preset range of the operating table;

[0050] In the case that the air chamber pressure data of the medical object fluctuates, obtaining the fluctuation value of the air chamber pressure data;

[0051] In the case that the circulating nurse object appears in the second preset range of the operating table and the fluctuation value is greater than a preset fluctuation threshold, it is determined that the circulating nurse object performs medical examination on the medical object, thereby monitoring the operation process of the medical object.

[0052] The pressure injury early warning method provided by the embodiment of the application can effectively avoid the occurrence of pressure injury and protect the health of the medical object by determining that the circulating nurse object appears in the surrounding range of the operating table of the medical object and that the circulating nurse object has examined the pressure part of the medical object through the pressure fluctuation of the air chamber unit of the medical object, and monitoring the medical examination of the circulating nurse object based on the two aspects.

[0053] In a second aspect, the application provides a pressure injury early warning device, which comprises:

[0054] The obtaining module is configured to obtain, for any medical object, a plurality of air chamber units required by the medical object for operation;

[0055] The determining module is configured to determine a plurality of sets of pressure early warning data and a plurality of sets of temperature early warning data of the medical object based on a threshold value determination model, wherein the threshold value determination model is obtained based on the relationship between the pressure data, the temperature data of the required air chamber units of a plurality of sample medical objects and whether pressure injury occurs.

[0056] The early warning module is configured to adjust the plurality of air chamber units based on the plurality of sets of pressure early warning data and the plurality of sets of temperature early warning data to early warn the pressure injury.

[0057] In a third aspect, the present application provides a computer device, comprising a memory and a processor, the memory and the processor being communicatively connected with each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the pressure injury early warning method according to the first aspect or any one of the corresponding embodiments.

[0058] In a fourth aspect, the present application provides a computer readable storage medium, the computer readable storage medium storing computer instructions, and the computer instructions being configured to cause a computer to perform the pressure injury early warning method according to the first aspect or any one of the corresponding embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0059] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed in the specific embodiments or the prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0060] Figure 1 is a flowchart of the pressure injury early warning method according to an embodiment of the present application;

[0061] Figure 2 is a flowchart of another pressure injury early warning method according to an embodiment of the present application;

[0062] Figure 3 is a schematic diagram of training a threshold determination model according to an embodiment of the present application;

[0063] Figure 4 is a schematic diagram of applying a threshold determination model according to an embodiment of the present application;

[0064] Figure 5 is a flowchart of adjusting air chamber units according to an embodiment of the present application;

[0065] Figure 6 is a schematic diagram of sample labeling positions according to an embodiment of the present application;

[0066] Figure 7 is a schematic diagram of object or object recognition by an object detection model according to an embodiment of the present application;

[0067] Figure 8 is a flowchart of continuously determining a circulating nurse object according to an embodiment of the present application;

[0068] Figure 9 is a schematic view of the perimeter coordinates of a surgical bed according to an embodiment of the present application;

[0069] Figure 10 is a schematic view of monitoring a surgical process according to an embodiment of the present application;

[0070] Figure 11 is a schematic view of a system for early warning of pressure injury according to an embodiment of the present application;

[0071] Figure 12 is a structural block diagram of an early warning device for pressure injury according to an embodiment of the present application;

[0072] Figure 13 is a structural schematic diagram of a computer device provided by an optional embodiment of the present application. DETAILED DESCRIPTION

[0073] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0074] The embodiments of the present application provide an early warning method for pressure injury, solve the problem that artificial inspection of a medical object by a medical care object may not be timely, thereby causing pressure injury, achieve timely early warning of pressure injury, guarantee the health of the medical object, and meanwhile avoid adjustment of a whole air cushion device in the related art, thereby reducing interference with a surgical process.

[0075] According to the embodiments of the present application, an early warning method for pressure injury is provided. It should be noted that the steps shown in the flowchart of the drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown herein.

[0076] In the present embodiment, an early warning method for pressure injury is provided, which can be used in the mobile terminal described above, such as an operating computer in an operating room, Figure 1 is a flowchart of an early warning method for pressure injury according to an embodiment of the present application, as shown in Figure 1 The flowchart includes the following steps:

[0077] Step S101, for any medical object, obtain a plurality of air chamber units required by the medical object for surgery. Specifically, when the medical object is operated, a gas exchange device with a plurality of air chamber units is used to prevent pressure injury. However, the basic data such as height and weight of different medical objects are different, and the type of surgery and the corresponding surgical position are also different, so the air chamber units required by different medical objects during surgery are different. It should be noted that the air chamber unit can be configured by the medical object based on experience, or can be obtained from a position database, or other methods that can determine the required air chamber unit of the medical object, and the embodiments of the present application do not limit this. Among them, the position database has a plurality of air chamber units corresponding to the position. Optionally, taking the type of surgery of the medical object as cholecystectomy and the surgical position as supine position as an example, the air chamber unit required by the medical object is determined by any of the above methods, and Table 1 below is the air chamber unit configuration data of the medical object.

[0078] Table 1: Example of air chamber unit configuration data of medical object

[0079]

[0080]

[0081] Among them, all the air chamber units in the gas exchange device are numbered in the order from top to bottom and from left to right, and the air chamber unit configuration data shown in Table 1 can be obtained by any of the above methods. Then, based on the configuration data, start the air chamber unit that needs to work. It should be noted that the air chamber unit can also be numbered by other methods, and the embodiments of the present application do not limit this. By obtaining the air chamber unit required by the medical object, non-pressure point air chamber units are avoided from working invalidly, and the influence on the surgical process is reduced.

[0082] Step S102, determine a plurality of groups of pressure warning data and a plurality of groups of temperature warning data of the medical object based on a threshold determination model, and the threshold determination model is trained based on the relationship between the pressure data, temperature data and whether the pressure injury occurs of the required air chamber units of a plurality of sample medical objects. Specifically, different medical objects have different sensitivities to pressure injury, so the pressure warning data and temperature warning data of different medical objects are different. At the same time, the length of time that the medical object remains motionless under different pressures until pressure injury occurs is different, and the length of time that the medical object remains motionless under different temperatures until pressure injury occurs is also different. Therefore, by determining the pressure warning data and temperature warning data of the medical object through the threshold determination model, the data of the occurrence of pressure injury can be obtained, so as to effectively warn the pressure injury.

[0083] Step S103, based on the multiple sets of pressure warning data and the multiple sets of temperature warning data, adjusting the multiple air chamber units to warn of pressure injury. Specifically, for any air chamber unit, when the air chamber unit meets the triggering condition of the pressure warning data or the temperature warning data, the air chamber unit is adjusted to avoid pressure injury of the medical object. It should be noted that the medical object only uses part of the air chamber units of the gas exchange device, and the pressure and temperature of different air chamber units are different, so only the air chamber unit that meets the triggering condition of the pressure warning data or the temperature warning data is adjusted. By automatically adjusting the air chamber unit, the problem that manual inspection of the medical object is not timely and pressure injury occurs is solved, timely warning of pressure injury is realized, the health of the medical object is ensured, and adjustment of all air chamber units is avoided, reducing interference with the surgical process.

[0084] The embodiment of the present application provides a pressure injury warning method, which adjusts the air chamber unit used by the medical object based on the pressure warning data and the temperature warning data of the medical object, solves the problem that manual inspection of the medical object by the medical staff may not be timely, thereby causing pressure injury, realizes timely warning of pressure injury, ensures the health of the medical object, and avoids adjustment of the overall air cushion device, thereby reducing interference with the surgical process.

[0085] In the embodiment, a pressure injury warning method is provided, which can be applied to the mobile terminal described above, such as an operating computer in an operating room, Figure 2 is a flowchart of another pressure injury warning method according to the embodiment of the present application, as shown in Figure 2 , comprising the following steps:

[0086] Step S201, for any medical object, obtaining multiple air chamber units required for the medical object to perform surgery. For details, refer to step S101 of the embodiment shown in Figure 1 , which will not be repeated here.

[0087] Step S202, determining multiple sets of pressure warning data and multiple sets of temperature warning data of the medical object based on a threshold determination model, the threshold determination model being trained based on the relationship between the pressure data, the temperature data and whether pressure injury occurs of the required air chamber units of multiple sample medical objects. For details, refer to step S102 of the embodiment shown in Figure 1 , which will not be repeated here.

[0088] In some optional embodiments, the training process of the threshold determination model comprises the following steps:

[0089] Step a1, obtaining body position data, preoperative diagnosis information and sample medical result of a plurality of sample medical objects, the sample medical result indicating whether the sample medical object has pressure injury. Specifically, the body position data and preoperative diagnosis information of any sample medical object used for training the threshold determination model are shown in Table 2 as follows:

[0090] Table 2. Examples of body position data and preoperative diagnosis information of sample medical objects

[0091] Field Remark Example patient_id Sample medical object id 100000922 Height Height (cm) 176 Weight Weight (kg) 70 Sex Sex (male: 1, female: 2) 1 Age Age 50 Race Race Han pre_diagnosis Preoperative diagnosis information Gallbladder stones surgery_type Surgery type Cholecystectomy surgical_position Surgical body position Supine position other_disease Other diseases Diabetes

[0092] Wherein, the sample medical object id corresponds to the sample medical object one by one, and is used for uniquely identifying the sample medical object. Since the body position data, preoperative diagnosis information and sensitivity to pressure injury of different medical objects are different, the threshold determination model can be trained based on a plurality of sample medical objects and a plurality of aspects of data, thereby improving the prediction accuracy of the model.

[0093] In some optional embodiments, the sample medical result of the sample medical object can be obtained through an adverse event system, which is a system used by the hospital to record adverse events, and the sample medical result can also be obtained based on other methods, which are not limited by the embodiments of the present application. Taking the sample medical object id 100000922 as an example, the sample medical result of the sample medical object is shown in Table 3 as follows:

[0094] Table 3. Examples of sample medical results of sample medical objects

[0095]

[0096]

[0097] Step a2, for any sample medical object, obtaining sample pressure data and sample temperature data of a plurality of gas chamber units used by the sample medical object during the operation. Optionally, based on the pressure sensors and temperature sensors at different positions in the gas exchange device, the sample pressure data and sample temperature data of the sample medical object are collected. It should be noted that, in order to reduce the influence on the appearance of the gas exchange device, the present application adopts Bluetooth wireless data transmission and real-time storage into the database, and other methods can also be used to obtain data, which are not limited by the embodiments of the present application. Table 4 below takes the gas chamber unit numbered 1 as an example, and the data structure of the sample pressure data and sample temperature data is obtained.

[0098] Table 4. Data structure example of sample pressure data and sample temperature data

[0099]

[0100] The sample pressure data and the sample temperature data are collected once a minute, and can also be collected once every thirty seconds, once every two minutes, or once every three minutes, which is not limited in the embodiments of the present application. Based on the data structure shown in Table 4, the continuously collected minute-level data examples are shown in Table 5.

[0101] Table 5: Examples of sample pressure data and sample temperature data

[0102] Time air_cell type value 2024-03-19 10:00:00 1 pressure 32 2024-03-19 10:00:00 1 temperature 38 2024-03-19 10:00:01 1 pressure 32 2024-03-19 10:00:01 1 temperature 38 2024-03-19 10:00:02 1 pressure 32 2024-03-19 10:00:02 1 temperature 38 2024-03-19 10:00:03 1 Figure 3 32 2024-03-19 10:00:03 1 Figure 3 38

[0103] In step a3, the sample pressure data of the air chamber unit is discretized to obtain a plurality of sets of sample pressure warning data, and the sample pressure warning data includes a sample pressure warning threshold and a sample pressure duration threshold. Specifically, based on expert experience, when the skin pressure of the medical object is greater than or equal to 32 mmHg, the probability of pressure injury is relatively large, but there is no judgment on the duration. Therefore, by discretizing the sample pressure data, grouping according to the air chamber unit number, and counting the continuous duration of the pressure value exceeding 32 mmHg, a plurality of sets of sample pressure warning thresholds and corresponding sample pressure duration thresholds can be obtained. The skin pressure of 32 mmHg is only a reference critical pressure value, and other methods can be used to determine the critical pressure value of pressure injury or specify other arbitrary reference critical pressure values, which are not limited in the embodiments of the present application.

[0104] In step a4, the sample temperature data of the plurality of air chamber units is discretized to obtain a plurality of sets of sample temperature warning data, and the sample temperature warning data includes a sample temperature warning threshold and a sample temperature duration threshold. Specifically, based on expert experience, when the skin temperature of the medical object is greater than or equal to 38 degrees Celsius, the probability of pressure injury is relatively large, but there is no judgment on the duration. Therefore, by discretizing the sample temperature data, grouping according to the air chamber unit number, and counting the continuous duration of the temperature value exceeding 38 degrees Celsius, a plurality of sets of sample temperature warning thresholds and corresponding sample temperature duration thresholds can be obtained. The skin temperature of 38 degrees Celsius is only a reference critical temperature value, and other methods can be used to determine the critical temperature value of pressure injury or specify other arbitrary reference critical temperature values, which are not limited in the embodiments of the present application. Optionally, taking the air chamber unit with number 1 and the sample medical object id 100000922 as an example, the sample pressure warning data and the sample temperature warning data obtained by steps a3 and a4 are shown in Table 6.

[0105] Table 6: Examples of sample pressure warning data and sample temperature warning data

[0106]

[0107]

[0108] Step a5, the body position data of the sample medical object, preoperative diagnosis information and sample medical results are taken as inputs, and multiple sets of sample pressure warning data and multiple sets of sample temperature warning data are taken as labels to train the threshold determination model. Specifically, the embodiment of the application takes the random forest algorithm as an example to train the threshold determination model. The body position data of the sample medical object, preoperative diagnosis information and sample medical results are taken as input feature data of the random forest algorithm, and multiple sets of sample pressure warning data and multiple sets of sample temperature warning data are taken as label data of the random forest algorithm. The label data is used to guide the prediction result that the threshold determination model should output, and the prediction result is the sample pressure warning data and the sample temperature warning data in the application. Figure 4 is a schematic diagram of training the threshold determination model according to the embodiment of the application, as shown in Figure 4 The body position data of the sample medical object includes height information, weight information, gender information, age information and nationality information; the sample medical result indicates whether pressure injury occurs; and the right label data takes the above table 6 as an example. The model fitting is performed based on the random forest algorithm, the threshold determination model is trained, and a model file is formed after the model training is completed.

[0109] In some optional embodiments, Figure 4 is a schematic diagram of applying the threshold determination model according to the embodiment of the application, as shown in Figure 5 For any medical object, the body position data, preoperative diagnosis information, medical result and model file of the threshold determination model of the medical object are taken as inputs of the random forest algorithm to obtain pressure warning data and temperature warning data, Figure 5 The fields provided in the above table 6 are taken as examples. The medical result defaults to pressure injury, so that the pressure warning data and the temperature warning data when the pressure injury occurs are determined based on the threshold determination model.

[0110] Step S203, based on multiple sets of pressure warning data and multiple sets of temperature warning data, adjusting multiple air chamber units to warn pressure injury.

[0111] Specifically, the above step S203 includes:

[0112] Step S2031, for any set of pressure warning data, determine the first duration of the air chamber unit under the pressure warning threshold corresponding to the pressure warning data. Specifically, any medical object corresponds to multiple sets of pressure warning data, which includes multiple pressure warning thresholds and the corresponding pressure duration thresholds. For any air chamber unit, determine the first duration of the air chamber unit under any pressure warning threshold, so as to determine whether to trigger the warning condition according to the duration under the pressure warning threshold, so as to effectively warn the pressure injury.

[0113] Step S2032, for any set of temperature warning data, determine the second duration of the air chamber unit under the temperature warning threshold corresponding to the temperature warning data. Specifically, any medical object corresponds to multiple sets of temperature warning data, which includes multiple temperature warning thresholds and the corresponding temperature duration thresholds. For any air chamber unit, determine the second duration of the air chamber unit under any temperature warning threshold, so as to determine whether to trigger the warning condition according to the duration under the temperature warning threshold, so as to effectively warn the pressure injury.

[0114] Step S2033, in the case where the first duration is not less than the pressure duration threshold, the air chamber unit is deflated, and the pressure injury is warned. Specifically, for any air chamber unit, configure multiple sets of pressure warning data, multiple sets of temperature warning data, pressure expected value and temperature expected value to the air chamber unit. When the first duration is not less than the pressure duration threshold, the skin pressure value of the medical object is large, and the probability of pressure injury is large. Therefore, the air chamber unit with the first duration not less than the pressure duration threshold is automatically deflated until the pressure value reaches the pressure expected value at this time, and the warning information is broadcasted, which is used to prompt the medical staff to perform medical examination on the pressure part of the medical object. Optionally, in order to reserve the buffer time to protect the health of the medical object, the above adjustment can be performed when the first duration is not less than the pressure duration threshold minus a unit granularity. The unit granularity can be 1 minute, 5 minutes or 10 minutes, and the embodiment of the present application does not limit this. By automatically deflating the air chamber unit that meets the pressure warning condition, the problem that the medical object is not timely examined by the medical staff and the medical object is injured by pressure is solved, the occurrence of pressure injury is effectively avoided, and the health of the medical object is protected.

[0115] Step S2034, in the case where the second duration is not less than the temperature duration threshold, the air chamber unit is cooled to pre-warn the pressure injury. Specifically, for any air chamber unit, a plurality of sets of pressure warning data, a plurality of sets of temperature warning data, a pressure expected value and a temperature expected value are configured to the air chamber unit. When the second duration is not less than the temperature duration threshold, the skin temperature value of the medical subject is high, and thus the probability of pressure injury is large. Therefore, the air chamber unit with the second duration not less than the temperature duration threshold is automatically cooled until the temperature value at this time reaches the temperature expected value, and warning information is also voice broadcasted. Optionally, in order to reserve a buffer time to protect the health of the medical subject, the above adjustment operation can be performed when the second duration is not less than the temperature duration threshold minus a unit granularity. By automatically cooling the air chamber unit that meets the temperature warning condition, the problem of pressure injury of the medical subject due to the medical subject not timely medical examination is solved, the occurrence of pressure injury is effectively avoided, and the health of the medical subject is protected.

[0116] In some optional embodiments, taking the unit granularity of 10 minutes, the pressure expected value of 30 mmHg, the temperature expected value of 36 degrees Celsius, and the fields provided in Table 5 above as examples, the data structure configured to the air chamber unit is described as shown in Table 7 below:

[0117] Table 7: Air chamber unit configuration data structure example

[0118]

[0119]

[0120] It should be noted that if all air chamber units used by the medical subject do not perform deflation or cooling operation within a preset time during the operation, that is, the first duration of all air chamber units under any pressure warning threshold is less than the pressure duration threshold and the second duration of all air chamber units under any temperature warning threshold is less than the temperature duration threshold, a timing reminder mechanism is triggered to automatically voice broadcast warning information to remind the medical staff to perform medical examination on the medical subject. The preset time can be 1 hour, 2 hours or 3 hours, etc., and the embodiments of the present application do not limit this. If any air chamber unit used by the medical subject performs deflation or cooling operation within the preset time, it is considered that the medical subject has taken measures to relieve pressure injury, and the timing is restarted. By setting the timing reminder mechanism on the basis of pressure and temperature warning, the probability of pressure injury is reduced, and the health of the medical subject is protected.

[0121] In some optional embodiments, Field is a flow diagram of adjusting the air chamber unit according to an embodiment of the present application, asRemark As shown, for any medical object, multiple sets of pressure warning thresholds and their corresponding pressure duration thresholds, as well as multiple sets of temperature warning thresholds and their corresponding temperature duration thresholds, are obtained through step S202. Then, all air chamber units used by the medical object are grouped according to pressure type and temperature type, as shown in Table 8. Taking air chamber unit numbered 1 as an example, whose data type is pressure, multiple sets of pressure warning thresholds and their corresponding pressure duration thresholds are transmitted to this air chamber unit. For any pressure warning threshold, if the first duration of the air chamber unit under the pressure warning threshold is not less than the pressure duration threshold, a pressure expectation value is transmitted to the air chamber unit, and the air chamber unit is adjusted to release gas until the pressure value reaches the pressure expectation value. Simultaneously, when the data type of the air chamber unit is temperature, multiple sets of temperature warning thresholds and their corresponding temperature duration thresholds are transmitted to the air chamber unit. For any temperature warning threshold, if the second duration of the warning at that temperature threshold is not less than the temperature duration threshold, a desired temperature value is transmitted to the air chamber unit, and the air chamber unit is adjusted to cool down until the temperature reaches the desired value. Wherein, any air chamber unit belongs to at least one of pressure type or temperature type.

[0122] Table 8. Examples of grouping air chamber units by type

[0123] Example air_cell Air cell unit number type Data type (pressure / temperature) 1 pressure Figure 1 Figure 1

[0124] The pressure injury early warning method provided in this invention adjusts the air chamber unit used by the medical subject based on pressure and temperature early warning data. This solves the problem that pressure injuries may occur due to untimely manual examination of the medical subject by medical staff, and achieves timely early warning of pressure injuries, thus protecting the health of the medical subject. At the same time, it avoids the need to adjust the entire air cushion device, reducing interference with the surgical procedure.

[0125] This embodiment provides a method for early warning of pressure injuries, which can be used in the aforementioned mobile terminal, such as an operating computer in an operating room. The method can also provide early warning of pressure injuries through the following steps:

[0126] Step S301: For any given medical patient, obtain the multiple air chamber units required for the surgery. See details below. Figure 2 Step S101 of the illustrated embodiment will not be described again here.

[0127] Step S302, determine the plurality of sets of pressure warning data and the plurality of sets of temperature warning data of the medical object based on a threshold determination model, the threshold determination model is trained based on the relationship between the pressure data, the temperature data of the required gas chamber unit of the plurality of sample medical objects and whether pressure injury occurs. For details, please refer to Figure 6 Step S102 of the embodiment shown, which will not be repeated here.

[0128] Step S303, based on the plurality of sets of pressure warning data and the plurality of sets of temperature warning data, adjust the plurality of gas chamber units to warn of pressure injury. For details, please refer to Figure 6 Step S203 of the embodiment shown, which will not be repeated here.

[0129] Step S304, acquire a plurality of panoramic pictures within a predetermined range of the medical object and gas chamber pressure data of the medical object, the gas chamber pressure data representing the pressure data of the gas chamber unit used by the medical object during the operation. Specifically, steps S201 to S203 reduce the risk of pressure injury through the automatic deflation or cooling of the air cushion exchange device. However, during the operation, the circulating nurse object needs to detect the pressure part of the medical object at a certain period and intervene in time. However, the circulating nurse object is busy and may forget or not perform well, resulting in the occurrence of pressure injury. The above steps S2033 and S2034 remind the circulating nurse object through voice broadcast, but cannot determine whether the circulating nurse object has performed the detection action on the medical object. Therefore, the plurality of panoramic pictures of the operating room are continuously collected through the panoramic camera in the operating room at the beginning of the operation, and the gas chamber pressure data of the medical object is acquired, so as to monitor whether the circulating nurse object has performed effective medical examination. The plurality of panoramic pictures can be collected in each stage of the operation, and the number can be 50, 100 or 150, etc. The embodiment of the present application does not limit this.

[0130] Step S305, identify the first frame of panoramic picture based on the object detection model, respectively determine the circulating nurse object, the operating bed and the operation computer, the object detection model is trained based on a plurality of sample panoramic pictures, and is used to identify the objects and objects in the picture. Specifically, during the operation, the circulating nurse object needs to perform medical examination on the pressure part of the medical object at regular intervals to avoid the occurrence of pressure injury. The object detection model can accurately identify the circulating nurse object, thereby realizing the monitoring of the circulating nurse object.

[0131] Step S306, based on the pressure data of the air chamber of the circulating nurse object, the operating bed and the medical object, the operation process of the medical object is monitored. Specifically, the current pressure condition of the air chamber unit used by the medical object can be reflected through the air chamber pressure data, so that it can be judged whether the circulating nurse object has performed medical examination on the medical object through the air chamber pressure data, so as to effectively avoid the occurrence of pressure injury.

[0132] Step S307, the first frame panoramic picture is identified based on the object detection model, and the circulating nurse object, the operating bed and the operation computer are determined respectively. The object detection model is trained based on a plurality of sample panoramic pictures, and is used to identify the objects and objects in the picture.

[0133] In some optional embodiments, the training process of the object detection model includes the following steps:

[0134] Step b1, a plurality of sample panoramic pictures are obtained. Specifically, the object detection model in the present application is taken as Yolo target detection model as an example. The Yolo (You Only Look Once) target detection pre-training model is trained through COCO (Microsoft Common Objects in Context) or ImageNet data set, and then the pre-training model is trained based on a plurality of sample panoramic pictures to obtain the object detection model.

[0135] Step b2, based on the object detection model, the plurality of sample panoramic pictures are identified to obtain a plurality of sample objects, sample operating beds and sample operation computers. Specifically, Figure 7 is a schematic diagram of sample annotation position according to an embodiment of the present application, as Figure 7 For any sample panoramic picture, the object detection model identifies the sample panoramic picture, and can identify each object and object, and the corresponding sample annotation position. The sample annotation position includes file name, region coordinate, center coordinate and target object.

[0136] Step b3, based on the difference between the sample annotation position and the actual position of the plurality of sample objects, sample operating beds and sample operation computers, the object detection model is trained. Specifically, by training the object detection model based on the difference between the sample annotation position and the actual position, the recognition ability of the object detection model is improved, so as to realize the monitoring of the circulating nurse object.

[0137] Specifically, the above step S307 includes:

[0138] In step S3071, the first frame panoramic picture is recognized based on the object detection model to determine a plurality of first objects, the operating table and the operation computer. Specifically, the object detection model identifies a plurality of objects in the first frame panoramic picture, and each object is regarded as a first object. Figure 8 FIG. 3 is a schematic diagram of object or object recognition according to the object detection model of an embodiment of the present application. Figure 8 As shown in FIG. 3, the object detection model identifies four first objects, the operating table and the operation computer from the first frame panoramic picture.

[0139] In step S3072, the first center coordinates of the operation computer are determined. Specifically, the first center coordinates of the operation computer are the center point of the operation computer region matrix.

[0140] In step S3073, the second center coordinates corresponding to the plurality of first objects are determined respectively. Specifically, since there are a plurality of first objects in the panoramic picture, the second center coordinates of the first objects are determined to determine the circulating nurse object.

[0141] In step S3074, for any first object, the distance between the second center coordinates of the first object and the first center coordinates is determined. Specifically, the circulating nurse object is determined by the distance between the center coordinates, which can be calculated by the Euclidean distance or other methods, and the present embodiment is not limited in this regard.

[0142] In step S3075, the first object corresponding to the smallest distance among the distances corresponding to the plurality of first objects is determined as the circulating nurse object. Specifically, since the circulating nurse object is by default working near the operation computer at the beginning of the operation, the first object closest to the operation object can be determined as the circulating nurse object.

[0143] In step S308, the next frame panoramic picture of the first frame panoramic picture is recognized based on the object detection model to determine a plurality of second objects. Specifically, the object detection model identifies a plurality of objects in the next frame panoramic picture, and each object is regarded as a second object. Since the circulating nurse object is not fixed during the operation, the distance between the circulating nurse object and the operation computer will change, and the circulating nurse object needs to be determined continuously.

[0144] In step S309, the circulating nurse object is feature extracted to obtain a first feature vector of the circulating nurse object. Specifically, the circulating nurse object can be feature extracted based on a CNN (Convolutional Neural Network) to obtain a 128-bit first feature vector of the circulating nurse object. The feature vector of the circulating nurse object can also be obtained by other methods, and the present embodiment is not limited in this regard.

[0145] Step S310, respectively, the plurality of second objects are extracted, and the second feature vector corresponding to the plurality of second objects is obtained. Specifically, the feature extraction is performed on all second objects in the next frame panoramic picture, and the plurality of second feature vectors are obtained, so that the patrol nurse object can be determined continuously based on the feature vector.

[0146] Step S311, for any second feature vector, the similarity value between the second feature vector and the first feature vector is determined based on the cosine similarity algorithm. Specifically, the similarity value between the first feature vector and the second feature vector can also be determined by other algorithms, and the embodiments of the application do not limit this.

[0147] Step S312, the second object corresponding to the second feature vector with the largest similarity value is determined as the new patrol nurse object. Specifically, the second object corresponding to the second feature vector with the largest similarity value, that is, the object most similar to the patrol nurse object, realizes the continuous determination of the patrol nurse object, and guarantees the accuracy of the monitoring of the patrol nurse object. Figure 9 is a flowchart of continuously determining the patrol nurse object according to the embodiments of the application, as shown in Figure 9 The panoramic picture in the operation process is continuously collected, and all second objects in the panoramic picture are recognized. Then, the feature extraction is performed on the patrol nurse object and all second objects, and the first feature vector and the plurality of second feature vectors are obtained. Then, the similarity values of the first feature vector and the plurality of second feature vectors are compared in turn. Then, the second object corresponding to the second feature vector with the largest similarity value is determined as the new patrol nurse object. Then, the second feature vector of the new patrol nurse object is taken as the new first feature vector, and the update of the patrol nurse object is realized.

[0148] Step S313, based on the air chamber pressure data of the patrol nurse object, the operating table and the medical object, the operation process of the medical object is monitored.

[0149] Specifically, the above step S313 includes:

[0150] Step S3131, the peripheral coordinates of the operating table are obtained, and the peripheral coordinates include the top-left corner coordinate, the top-right corner coordinate, the bottom-left corner coordinate and the bottom-right corner coordinate of the operating table. Specifically, Position is a schematic diagram of the peripheral coordinates of the operating table according to the embodiments of the application, as shown in Judgment rule The center coordinates and the peripheral coordinates of the operating table can be determined based on the object detection model.

[0151] Step S3132, if the coordinates of the circulating nurse object appear in the first preset range of the surrounding coordinates within the preset time period, it is determined that the circulating nurse object appears in the second preset range of the operating bed. Specifically, by the coordinates of the circulating nurse object and the surrounding coordinates of the operating bed, it can be determined whether the circulating nurse object moves from the operating computer to the operating bed and walks around the operating bed to check the medical object. Among them, the position relationship shown in Table 9 is used for judgment.

[0152] Table 9 Coordinate judgment rule

[0153] Upper left corner Lower left corner Upper right corner X8 < X3 && Y8 < Y3 Lower right corner X10 < X3 && Y10 > Y3 X11>X3&&Y11>Y3 X9>X3 && Y9X3 Figure 10 Figure 10

[0154] Among them, it is assumed that the center coordinates of the operating bed are (X3, Y3), the upper left corner coordinates are (X8, Y8), the lower left corner coordinates are (X10, Y10), the upper right corner coordinates are (X9, Y9), and the lower right corner coordinates are (X11, Y11). Within the preset time period, when the coordinates of the circulating nurse object coincide with the above four coordinates or are within the first preset range of the above four coordinates, and the judgment rule in the above Table 9 is met, it means that the circulating nurse object is in the second preset range of the operating bed, that is, the circulating nurse object is beside the operating bed, and the circulating nurse object has completed the action of walking around the operating bed. The preset time period can be 3 minutes, 5 minutes or 6 minutes, etc., and the first preset range and the second preset range can be self-defined, and the present embodiment does not limit the above preset values.

[0155] Step S3133, in the case that the air chamber pressure data of the medical object fluctuates, the fluctuation value of the air chamber pressure data is obtained. Specifically, when the air chamber pressure data of the medical object fluctuates, it means that the circulating nurse object starts to perform medical examination on the medical object, so the fluctuation value of the air chamber pressure data can be obtained to determine whether the medical examination of the circulating nurse object is performed in place.

[0156] Step S3134, in the case that the circulating nurse object appears in the second preset range of the operating bed and the fluctuation value is greater than the preset fluctuation threshold, it is determined that the circulating nurse object performs medical examination on the medical object, and the monitoring of the operation process of the medical object is realized. Specifically, taking the fluctuation value as 7 and calculating the preset fluctuation threshold as 20% of the pressure expected value, that is, the preset fluctuation threshold is 6, as an example for illustration. When the circulating nurse object appears in the second preset range of the operating bed, that is, the circulating nurse object has completed the action of walking around the operating bed, in this process, the fluctuation value 7 is greater than the preset fluctuation threshold 6, which means that the circulating nurse object has performed medical examination on the pressure part of the medical object. The preset fluctuation threshold can also be calculated based on other methods or self-set, and the present embodiment does not limit this.

[0157] In some alternative embodiments, Figure 11 is a schematic diagram of monitoring a surgical procedure according to an embodiment of the present application, as Figure 11 indicated, the object detection model is trained through the above steps b1 to b3. By checking whether the circulating nurse object performs a bed check on the medical object and whether the air chamber pressure data of the medical object fluctuates, it is determined whether the circulating nurse object has performed a medical examination, thereby achieving monitoring of the surgical procedure.

[0158] In some alternative embodiments, Figure 12 is a system schematic diagram of a pressure injury early warning method according to an embodiment of the present application, as Figure 12 indicated, the above step S301 can be implemented based on a body position adaptive module, step S302 can be implemented based on a monitoring threshold automatic module, step S303 can be implemented based on an automatic analysis calculation module and an air chamber unit control module, and steps S304 to S313 can be implemented based on a circulating nurse behavior monitoring module.

[0159] The pressure injury early warning method provided in the embodiment of the present application can identify the circulating nurse object based on the object detection model, and can determine whether the circulating nurse object has performed effective medical examination on the medical object by obtaining the air chamber pressure data of the medical object, thereby reducing the risk of pressure injury.

[0160] In the embodiment, a pressure injury early warning device is also provided, which is used to implement the above-mentioned embodiments and preferred embodiments, and will not be described again. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware, or a combination of software and hardware is also possible and contemplated.

[0161] The embodiment provides a pressure injury early warning device, as Figure 13 indicated, comprising:

[0162] The acquisition module 1201 is configured to, for any medical object, acquire a plurality of air chamber units required for the medical object to perform a surgical procedure.

[0163] The determination module 1202 is configured to determine a plurality of sets of pressure warning data and a plurality of sets of temperature warning data of the medical object based on a threshold determination model, wherein the threshold determination model is trained based on the relationship between the pressure data, the temperature data of the required air chamber units of a plurality of sample medical objects, and whether pressure injury occurs.

[0164] The early warning module 1203 is configured to adjust the plurality of air chamber units based on the plurality of sets of pressure warning data and the plurality of sets of temperature warning data, and early warning of pressure injury.

[0165] In an optional implementation, each set of pressure warning data comprises a pressure warning threshold and a pressure duration threshold, and each set of temperature warning data comprises a temperature warning threshold and a temperature duration threshold; the warning module 1203 comprises:

[0166] a first determination unit, configured to determine, for any set of pressure warning data, a first duration of the gas chamber unit at the pressure warning threshold corresponding to the pressure warning data.

[0167] a second determination unit, configured to determine, for any set of temperature warning data, a second duration of the gas chamber unit at the temperature warning threshold corresponding to the temperature warning data.

[0168] a deflation unit, configured to perform a deflation operation on the gas chamber unit to warn of pressure injury if the first duration is not less than the pressure duration threshold.

[0169] a cooling unit, configured to perform a cooling operation on the gas chamber unit to warn of pressure injury if the second duration is not less than the temperature duration threshold.

[0170] In an optional implementation, the training process of the threshold determination model comprises:

[0171] obtaining body position data, preoperative diagnosis information and sample medical results of a plurality of sample medical objects, the sample medical results indicating whether the sample medical objects have pressure injury;

[0172] for any sample medical object, obtaining sample pressure data and sample temperature data of a plurality of gas chamber units used by the sample medical object during the operation;

[0173] performing discretization processing on the sample pressure data of the gas chamber unit to obtain a plurality of sets of sample pressure warning data, the sample pressure warning data comprising a sample pressure warning threshold and a sample pressure duration threshold;

[0174] performing discretization processing on the sample temperature data of the plurality of gas chamber units to obtain a plurality of sets of sample temperature warning data, the sample temperature warning data comprising a sample temperature warning threshold and a sample temperature duration threshold;

[0175] training the threshold determination model by taking the body position data, preoperative diagnosis information and sample medical results of the sample medical object as input and taking the plurality of sets of sample pressure warning data and the plurality of sets of sample temperature warning data as labels.

[0176] In an optional implementation, the apparatus further comprises:

[0177] The picture acquisition module is configured to acquire a plurality of panoramic pictures within a preset range of a medical subject and air chamber pressure data of the medical subject, the air chamber pressure data representing pressure data of an air chamber unit used by the medical subject during a surgery process.

[0178] The first identification module is configured to identify the first panoramic picture based on an object detection model, and determine a circulating nurse object, a surgery bed, and an operation computer, respectively.

[0179] The monitoring module is configured to monitor the surgery process of the medical subject based on the circulating nurse object, the surgery bed, and the air chamber pressure data of the medical subject.

[0180] In an optional implementation, the training process of the object detection model includes:

[0181] acquiring a plurality of sample panoramic pictures;

[0182] identifying the plurality of sample panoramic pictures based on the object detection model to obtain sample labeled positions of a plurality of sample objects, a sample surgery bed, and a sample operation computer, respectively;

[0183] training the object detection model based on differences between the sample labeled positions and actual positions of the plurality of sample objects, the sample surgery bed, and the sample operation computer.

[0184] In an optional implementation, the first identification module includes:

[0185] The identification unit is configured to identify the first panoramic picture based on the object detection model to determine a plurality of first objects, a surgery bed, and an operation computer.

[0186] The third determination unit is configured to determine a first center coordinate of the operation computer.

[0187] The fourth determination unit is configured to determine a second center coordinate corresponding to each of the plurality of first objects, respectively.

[0188] The distance determination unit is configured to determine, for any first object, a distance between the second center coordinate of the first object and the first center coordinate.

[0189] The fifth determination unit is configured to determine, as the circulating nurse object, a first object corresponding to a smallest distance among distances corresponding to the plurality of first objects.

[0190] In an optional implementation, after the first identification module, the apparatus further includes:

[0191] The second identification module is configured to identify a next panoramic picture of the first panoramic picture based on the object detection model to determine a plurality of second objects.

[0192] The first extraction module is configured to perform feature extraction on the circulating nurse object to obtain a first feature vector of the circulating nurse object.

[0193] The second extraction module is configured to perform feature extraction on each of the plurality of second objects to obtain a second feature vector corresponding to each of the plurality of second objects.

[0194] The similarity determination module is configured to determine, for any second feature vector, a similarity value of the second feature vector and the first feature vector based on a cosine similarity algorithm.

[0195] The object determination module is configured to determine, as a new circulating nurse object, a second object corresponding to a second feature vector with the largest similarity value.

[0196] In an optional embodiment, the monitoring module comprises:

[0197] The coordinate acquisition unit is configured to acquire a surrounding coordinate of the operating bed, the surrounding coordinate comprising a top-left corner coordinate, a top-right corner coordinate, a bottom-left corner coordinate and a bottom-right corner coordinate of the operating bed.

[0198] The range determination unit is configured to determine, within a preset time period, that the circulating nurse object appears in a second preset range of the operating bed if a coordinate of the circulating nurse object appears in a first preset range of the surrounding coordinate.

[0199] The fluctuation acquisition unit is configured to acquire a fluctuation value of the air chamber pressure data of the medical object if the air chamber pressure data of the medical object fluctuates.

[0200] The monitoring unit is configured to determine that the circulating nurse object performs a medical examination on the medical object and to monitor a surgery process of the medical object if the circulating nurse object appears in the second preset range of the operating bed and the fluctuation value is greater than a preset fluctuation threshold.

[0201] Further functions of each of the above modules and units are the same as those of the corresponding embodiments described above, and thus will not be described herein.

[0202] The pressure injury early warning device in the embodiment is in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and a memory executing one or more software or fixed programs, and / or other devices capable of providing the above functions.

[0203] The embodiment of the present application also provides a computer device having the pressure injury early warning device shown in the above Figure 13

[0204] Please refer to​Figure 13 , Figure 13 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 13 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). ​ Take a processor 10 as an example.

[0205] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0206] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0207] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0208] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0209] The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 can be connected through a bus or other means, ​ The bus connection is taken as an example in the foregoing description.

[0210] The input device 30 can receive inputted digital or character information, and generate key signal input related to user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 can include a display device, an auxiliary lighting device (e.g., an LED), a tactile feedback device (e.g., a vibration motor), etc. The display device includes but is not limited to a liquid crystal display, a light-emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0211] The embodiments of the present application also provide a computer readable storage medium, and the method according to the embodiments of the present application can be implemented in hardware, firmware, or recorded in a storage medium, or stored in a remote storage medium or a non-transitory machine readable storage medium and downloaded from a network and stored in a local storage medium by computer code, so that the method described herein can be processed by such software on a storage medium using a general purpose computer, a special purpose processor or programmable or special hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk or a solid state disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that the computer, processor, microprocessor controller or programmable hardware includes a storage component that can store or receive software or computer code, when the software or computer code is accessed and executed by the computer, processor or hardware, the method shown in the above embodiments is implemented.

[0212] Although the embodiments of the present application are described in conjunction with the accompanying drawings, various modifications and changes can be made by those skilled in the art without departing from the spirit and scope of the present application, and such modifications and changes fall within the scope defined by the appended claims.

Claims

1. A pressure injury early warning device, characterized by, The device comprises: an acquisition module configured to acquire, for any medical subject, a plurality of air chamber units required by the medical subject for surgery; a determination module configured to determine, based on a threshold determination model, a plurality of sets of pressure warning data and a plurality of sets of temperature warning data of the medical subject, the threshold determination model being trained based on relationships of pressure data, temperature data and whether pressure injury occurs of required air chamber units of a plurality of sample medical subjects; a warning module configured to adjust the plurality of air chamber units based on the plurality of sets of pressure warning data and the plurality of sets of temperature warning data to warn of pressure injury, each set of pressure warning data comprising a pressure warning threshold and a pressure duration threshold, and each set of temperature warning data comprising a temperature warning threshold and a temperature duration threshold; wherein the warning module comprises: a first determination unit configured to determine, for any set of pressure warning data, a first duration of the air chamber unit at the pressure warning threshold corresponding to the set of pressure warning data; a second determination unit configured to determine, for any set of temperature warning data, a second duration of the air chamber unit at the temperature warning threshold corresponding to the set of temperature warning data; a deflation unit configured to perform a deflation operation on the air chamber unit to warn of pressure injury if the first duration is not less than the pressure duration threshold; a cooling unit configured to perform a cooling operation on the air chamber unit to warn of pressure injury if the second duration is not less than the temperature duration threshold; the training process of the threshold determination model comprises: acquiring body position data, preoperative diagnosis information and sample medical results of the plurality of sample medical subjects, the sample medical results indicating whether the sample medical subjects have pressure injury; for any sample medical subject, acquiring sample pressure data and sample temperature data of a plurality of air chamber units used by the sample medical subject during surgery; performing discretization processing on the sample pressure data of the air chamber units to obtain a plurality of sets of sample pressure warning data, the sample pressure warning data comprising a sample pressure warning threshold and a sample pressure duration threshold; performing discretization processing on the sample temperature data of the plurality of air chamber units to obtain a plurality of sets of sample temperature warning data, the sample temperature warning data comprising a sample temperature warning threshold and a sample temperature duration threshold; training the threshold determination model by taking the body position data, the preoperative diagnosis information and the sample medical results of the sample medical subjects as input and taking the plurality of sets of sample pressure warning data and the plurality of sets of sample temperature warning data as labels.

2. The apparatus of claim 1, wherein, The device further comprises: a picture acquisition module configured to acquire a plurality of panoramic pictures within a predetermined range of a medical subject and air chamber pressure data of the medical subject, the air chamber pressure data representing pressure data of air chamber units used by the medical subject during surgery; a first identification module configured to identify a first panoramic picture based on an object detection model to determine a circulating nurse object, a surgical bed and an operation computer, respectively, the object detection model being trained based on a plurality of sample panoramic pictures to identify objects and objects in a picture; The monitoring module is configured to monitor the operation process of the medical object based on the circulating nurse object, the operating bed, and the chamber pressure data of the medical object.

3. The apparatus of claim 2, wherein, The training process of the object detection model comprises: obtaining the plurality of sample panoramic pictures; identifying the plurality of sample panoramic pictures based on the object detection model to obtain sample labeled positions of a plurality of sample objects, a sample operating bed, and a sample operating computer respectively; training the object detection model based on differences between the sample labeled positions and actual positions of the plurality of sample objects, the sample operating bed, and the sample operating computer.

4. The apparatus of claim 2, wherein, The first identification module comprises: an identification unit configured to identify the first frame of panoramic picture based on the object detection model to determine a plurality of first objects, the operating bed, and the operating computer; a third determination unit configured to determine a first center coordinate of the operating computer; a fourth determination unit configured to determine second center coordinates corresponding to the plurality of first objects respectively; a distance determination unit configured to determine, for any first object, a distance between the second center coordinate of the first object and the first center coordinate; a fifth determination unit configured to determine, as the circulating nurse object, a first object corresponding to a smallest distance among distances corresponding to the plurality of first objects.

5. The apparatus of claim 2, wherein, After the first identification module, the device further comprises: a second identification module configured to identify a next frame of panoramic picture of the first frame of panoramic picture based on the object detection model to determine a plurality of second objects; a first extraction module configured to perform feature extraction on the circulating nurse object to obtain a first feature vector of the circulating nurse object; a second extraction module configured to perform feature extraction on the plurality of second objects respectively to obtain second feature vectors corresponding to the plurality of second objects; a similarity determination module configured to determine, for any second feature vector, a similarity value of the second feature vector and the first feature vector based on a cosine similarity algorithm; an object determination module configured to determine, as a new circulating nurse object, a second object corresponding to a second feature vector with a largest similarity value.

6. The apparatus of claim 2, wherein, The monitoring module comprises: a coordinate acquisition unit configured to acquire peripheral coordinates of the operating bed, the peripheral coordinates comprising a top-left corner coordinate, a top-right corner coordinate, a bottom-left corner coordinate, and a bottom-right corner coordinate of the operating bed; a range determination unit configured to determine, within a preset time period, that the circulating nurse object appears in a second preset range of the operating bed if a coordinate of the circulating nurse object appears in a first preset range of the peripheral coordinates; a fluctuation acquisition unit configured to acquire a fluctuation value of the chamber pressure data of the medical object in a case where the chamber pressure data of the medical object fluctuates; a monitoring unit configured to determine that the circulating nurse object performs medical examination on the medical object and implement monitoring of the operation process of the medical object in a case where the circulating nurse object appears in the second preset range of the operating bed and the fluctuation value is greater than a preset fluctuation threshold.

Images