Wireless nursing monitoring device based on infrared identification of stroke patient and application

Abstract

The invention relates to a wireless nursing monitoring device based on cerebral apoplexy patient infrared identification and application, and relates to the technical field of monitoring equipment. The device comprises an infrared thermal imager, an upper computer, an image processing module and a three-dimensional calculation module, the motion state of the stroke patient in a ward can be monitored in real time, the lying state, the sitting state and the walking state can be accurately recognized, and the walking ability is evaluated through the stride difference. Infrared thermal imaging is innovatively adopted to protect privacy, image data is efficiently processed in combination with a specific algorithm, and the computing resource demand and cost are reduced. A cloud module and a mobile terminal are integrated, and remote monitoring and data management are achieved. By monitoring physiological parameters such as body temperature, heart rate, blood oxygen and the like, medical staff are reminded of multi-stage feedback when abnormity occurs. Nursing safety and rehabilitation effect are improved, and wide application prospects are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of monitoring equipment, and in particular to a wireless nursing monitoring device based on infrared identification of stroke patients and its application. Background Art

[0002] During the care of stroke patients, real-time monitoring of their activities within the ward is crucial. Existing patient monitoring equipment often uses traditional cameras for video surveillance, which presents numerous drawbacks. On the one hand, while cameras can clearly capture a patient's facial expressions and body details, this can easily lead to privacy breaches in private settings like wards, causing psychological anxiety and impacting their willingness to receive care and their recovery progress. On the other hand, traditional cameras capture massive amounts of image data, placing high demands on storage and transmission resources. They also require complex facial recognition and behavioral analysis algorithms for image processing, which not only consumes significant computing resources but also places high demands on the device's hardware performance, making it difficult to reduce device costs.

[0003] Furthermore, such devices typically only provide two-dimensional image information, limiting their ability to accurately determine a patient's spatial position and motion status. This makes it difficult for medical staff to conduct detailed analysis and timely intervention. Furthermore, most existing monitoring devices are unable to monitor physiological parameters such as a patient's body temperature in real time, making it impossible to detect physiological abnormalities and provide rapid feedback. Therefore, there is an urgent need to develop a new type of nursing monitoring device that can effectively protect the privacy of stroke patients while enabling accurate, real-time monitoring of a patient's motion status with minimal resource consumption and cost investment, providing strong support for improving nursing quality and promoting patient recovery. Summary of the Invention

[0004] In order to solve the above problems, the present invention provides a wireless nursing monitoring device based on infrared recognition of stroke patients, comprising an infrared thermal imager, a host computer, an image processing module and a three-dimensional calculation module;

[0005] The host computer is connected to the image processing module, and the image processing module is connected to the infrared thermal imager and the three-dimensional calculation module;

[0006] The number of infrared thermal imagers is 2 or more, which are used to collect infrared thermal imaging images of patients in the ward; the images collected by the infrared thermal imager are sent to the image processing module for image recognition, and the patient area in the image is extracted and the position is marked according to the patient area; then the three-dimensional calculation module obtains the patient's motion parameters; the image processing module sends the infrared thermal imaging image and the patient's motion parameters to the host computer; the host computer analyzes the patient's motion status according to the patient's motion parameters.

[0007] Furthermore, in one embodiment, the image processing module identifies the patient area and marks the positions of the head, shoulders, hips, and feet of the human body. The identification method is as follows:

[0008] The captured infrared thermal imaging images are preprocessed by using a median filter algorithm to remove salt and pepper noise from the images. The images are then grayscaled to convert the color images into grayscale images. An adaptive threshold segmentation algorithm is used to automatically determine the threshold based on the grayscale value distribution of different areas in the image, segmenting the image into foreground and background parts, and preliminarily extracting the area where the human body is located.

[0009] Perform morphological processing on the segmented human body area, using dilation and erosion operations to remove small noise points and fill the holes in the human body area, calculate the connected areas of the human body area, and determine the outline of the human body;

[0010] Using a joint recognition algorithm based on thermal imaging features, the patient's head, shoulders, hips, and feet are identified and marked, including:

[0011] When identifying the head, the highest point of the human body outline is found and combined with the thermal distribution characteristics, the head is determined to be located on the condition that the head is one of the areas with higher temperature on the human body and appears as a brighter area in the thermal image;

[0012] When identifying the shoulder, based on the characteristic that the heat distribution of the neck is narrow and that of the shoulder is wide, the search is expanded to both sides of the neck, looking for the area with the wide heat distribution as the shoulder location;

[0013] When identifying the crotch, look for an area in the lower area of ​​the human body with a relatively narrow heat distribution and located near the center line of the human body as the crotch location;

[0014] When identifying the feet, the area at the bottom of the human body outline is used as the location of the feet; the identified positions of the head, shoulders, hips, and feet are marked.

[0015] Furthermore, in one embodiment, the three-dimensional calculation module uses two infrared thermal imagers to perform three-dimensional position calculation, specifically in the following manner:

[0016] The imaging planes of the two infrared thermal imagers are calibrated in a pre-set spatial coordinate system within the ward to determine their positions and attitude parameters within the spatial coordinate system. Based on the imaging parameters and relative positional relationship of the two infrared thermal imagers, a spatial grid model is established to divide the monitoring area within the ward into multiple regular spatial grid cells, each with a unique three-dimensional coordinate identifier.

[0017] The image processing module extracts the patient area and the positions of the head, shoulders, hips, and feet, and then marks the position information in the images captured by the two infrared thermal imagers. The 3D calculation module calculates the 3D position coordinates of the patient and each of their joints in the spatial coordinate system based on the marked positions in the images of the two infrared thermal imagers, combined with the calibrated position and posture parameters, using the principle of triangulation.

[0018] The calculated three-dimensional position information is matched with the pre-established spatial grid model to determine the grid cells where the patient and his or her joints are located.

[0019] Furthermore, in one embodiment, the parameters of the infrared thermal imager have the following limitations:

[0020] The image resolution of an infrared thermal imager ranges from [X1×Y1] pixels to [X2×Y2] pixels, where X1 and Y1 represent the minimum resolution at which the outline of a human body can be accurately extracted, and X2 and Y2 represent the maximum resolution at which facial features cannot be clearly distinguished.

[0021] Furthermore, in one embodiment, the infrared thermal imager includes an infrared thermal imaging sensor and a matrix light-transmitting plate arranged in front of the infrared thermal imaging sensor, the matrix light-transmitting plate is composed of a plurality of microlenses, and the microlenses are arranged horizontally and vertically to form a microlens array; the number of rows and columns of the microlens array is between [X1×Y1] and [X2×Y2] to ensure that the image resolution of the infrared thermal imager is between [X1×Y1] pixels and [X2×Y2] pixels.

[0022] Furthermore, in one embodiment,

[0023] The calculation formula for the lower limit of resolution [X1×Y1] is:

[0024] X1 = ceil((W × n) / L);

[0025] Y1 = ceil((H × n) / H);

[0026] ceil represents rounding up; W and H are the width and height of the ward, respectively, to ensure that the thermal imager can cover the entire ward area; L and H are the minimum horizontal and vertical dimensions of the human body, which are used to determine the minimum pixel coverage of the human body in the image; n is the number of pixels required to capture the temperature difference, calculated based on the thermal imager's thermal sensitivity NETD and the minimum temperature difference ΔT between the human body and the background, n = ceil(ΔT / NETD);

[0027] The calculation formula for the upper limit of resolution [X2×Y2] is:

[0028] X2=floor((D×θ) / (d×π / 180));

[0029]

[0030] floor means rounding down; D is the monitoring distance; θ and are the horizontal and vertical field of view of the thermal imager; d is the diameter of the facial key features, which is used to determine the maximum pixel size of the facial features in the image to ensure that the face cannot be distinguished.

[0031] Furthermore, in one embodiment, the minimum horizontal dimension of the human body is H = 0.5 m, the height of the ward is H = 2 m, the minimum vertical dimension of the human body is H = 1 m, the temperature difference between the human body and the background is ΔT = 0.5° C., and the NETD of the thermal imager is 50 mK, i.e., 0.05° C., then n = ceil(0.5 / 0.05) = 10; and it is calculated that X1 = ceil(4×10 / 0.5) = 80 pixels, and Y1 = ceil(2×10 / 1) = 20 pixels;

[0032] Monitoring distance D = 2m, horizontal field of view angle θ = 45°, vertical field of view angle Facial feature diameter d = 2 cm = 0.02 m.

[0033] Furthermore, in one embodiment, the host computer performs motion state analysis on the patient's motion parameters. The host computer receives the infrared thermal imaging image and the patient's motion parameters sent by the image processing module and sends them to the data processing module. The data processing module determines whether the patient is in bed, sitting or walking state based on the patient's motion parameters.

[0034] When the position of the patient's head, shoulders, hips and feet relative to the bed does not change significantly, the patient is judged to be in bed; when the patient's feet are in contact with the bed or the ground, and the patient's hips are in contact with the bed or chair, and the patient's head, shoulders and hips are relatively stable without significant displacement, the patient is judged to be in a sitting state; when the patient's feet are detected to alternately leave the ground and touch the ground, and the body's center of gravity moves back and forth regularly, the patient is judged to be in a walking state;

[0035] In the walking state, the patient's walking ability is further analyzed: the walking ability is assessed by calculating the stride difference between the patient's two feet. The stride difference is defined as the absolute value of the difference between the stride lengths of each step of the left and right feet. The calculation formula is:

[0036] S diff =|S left -S right ∣;

[0037] S left is the left footstep length, S right is the right foot step; the stride is calculated by the change in the landing position of the same foot in consecutive frames; if the stride difference S diffExceeds the set threshold S th , it is determined that the patient's walking ability is abnormal and requires further attention and intervention. The abnormal walking ability is divided into 3-5 levels according to the range of stride difference.

[0038] Furthermore, in one embodiment, a cloud module and a mobile terminal are also provided;

[0039] The cloud module is used to store and manage the data collected by the monitoring device, including the patient's infrared thermal imaging images, motion parameters and motion status analysis results; users can remotely access the cloud module through a mobile terminal to view the patient's monitoring data;

[0040] The mobile terminal is used to receive patient movement status information and abnormal reminders sent by the host computer; when the host computer determines that the patient's walking ability is abnormal, it sends reminder signals of varying degrees to the mobile terminal according to the degree of abnormality, so as to timely understand the patient's abnormal condition and take corresponding measures.

[0041] The present invention also protects the use of a wireless nursing monitoring device based on infrared identification of stroke patients for monitoring in a stroke patient ward.

[0042] This device is designed for stroke patient wards and is mainly used to monitor patients' daily activities in real time in the ward. It is particularly suitable for the following scenarios:

[0043] Early stage of recovery: The patient's body functions have not yet fully recovered and his mobility is limited. His small movements in the ward need to be closely monitored to prevent accidents such as falls due to sudden situations.

[0044] Nighttime rest: When patients are sleeping at night, it is difficult for medical staff to observe their condition in real time. This device can monitor continuously and issue timely reminders if the patient shows abnormal activities or is at risk of getting up.

[0045] Daily activities: When patients perform simple rehabilitation activities or move independently in the ward, the device can accurately capture their movement status, providing data support for medical staff to assess the patient's safety and health.

[0046] The beneficial effects of the present invention are:

[0047] The present invention uses an infrared thermal imager to capture images and, through specific image processing technology, only extracts information on the human body outline and key parts, avoiding the clear imaging of private parts such as the patient's face by traditional cameras, effectively protecting the patient's privacy. The invention is particularly suitable for use in private spaces such as wards, so that patients do not need to worry about the leakage of personal privacy while being monitored. It can be used not only for monitoring in wards, but also for monitoring in spaces such as bathrooms, thereby improving patient acceptance and comfort.

[0048] The advantages of the algorithm presented in this paper are primarily reflected in its efficient utilization of computing resources and cost control. First, the algorithm utilizes preprocessing steps such as median filtering, grayscale conversion, and adaptive threshold segmentation, significantly reducing image data volume and computational complexity while ensuring accurate human contour extraction. These preprocessing steps remove salt-and-pepper noise and redundant information from the image, reducing the computational burden of subsequent processing and, therefore, lowering hardware resource requirements. The algorithm also incorporates a joint recognition method based on thermal imaging features to identify and label key body parts, such as the head, shoulders, hips, and feet. This approach avoids the need for complex deep learning models or high-precision sensors, reducing computing resource consumption. Furthermore, the algorithm leverages thermal distribution characteristics and knowledge of human anatomy to effectively identify key parts in low-resolution images, further reducing image resolution requirements and storage and transmission costs. The algorithm employs triangulation principles to calculate the three-dimensional position of the patient and their joints, combining image data from two infrared thermal imagers. This approach not only improves position calculation accuracy but also avoids the use of expensive 3D imaging equipment, reducing the overall system cost. The algorithm's high efficiency and low resource usage enable the entire system to run on standard computing devices, eliminating the need for high-performance processors or dedicated graphics processing units (GPUs), thereby reducing hardware costs. Furthermore, the algorithm's design takes into account the data transmission and storage requirements of real-world applications, reducing data transmission and storage space usage by optimizing the data processing process.

[0049] The device can capture a patient's movement status in real time. By rapidly processing and analyzing infrared thermal images, it can accurately identify whether the patient is bedridden, sitting, or walking. Especially for stroke patients, it can promptly detect abnormal activity during critical periods such as the early stages of recovery and at night, providing a powerful safeguard against accidents like falls and effectively improving the safety and timeliness of care.

[0050] As the patient walks, the device calculates key indicators such as stride length difference to quantitatively assess their walking ability and issue graded warnings based on the degree of abnormality. This not only helps medical staff more accurately understand the patient's rehabilitation progress, but also provides a scientific basis for developing personalized rehabilitation training plans, helping to improve the effectiveness of rehabilitation treatment and accelerate the patient's recovery.

[0051] By combining cloud modules and mobile terminals, medical staff can access patient status information and historical data anytime, anywhere, regardless of time and space constraints. This remote monitoring model not only improves nursing efficiency and reduces the workload of medical staff, but also ensures that patients receive timely attention and response even when they are away from their families, enabling continuous and dynamic monitoring of patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0053] Attachment Figure 1 This is a block diagram of the overall structure of the device system of the present invention;

[0054] Attachment Figure 2 Infrared thermal imaging image and marking position of the patient of the present invention;

[0055] Attachment Figure 3 Marking positions and grids for the patient's posture of the present invention;

[0056] Attachment Figure 4 This is a structural diagram of the infrared thermal imager according to embodiment 2 of the present invention.

[0057] Reference numerals: 1 matrix light-transmitting plate, 2 infrared thermal imaging sensor. DETAILED DESCRIPTION

[0058] Example 1:

[0059] See also Figures 1 to 3 , the present invention provides a wireless nursing monitoring device based on infrared recognition of stroke patients, including an infrared thermal imager, a host computer, an image processing module and a three-dimensional calculation module;

[0060] The host computer is connected to the image processing module, and the image processing module is connected to the infrared thermal imager and the three-dimensional calculation module;

[0061] The number of infrared thermal imagers is 2 or more, which are used to collect infrared thermal imaging images of patients in the ward; the images collected by the infrared thermal imager are sent to the image processing module for image recognition, and the patient area in the image is extracted and the position is marked according to the patient area; then the three-dimensional calculation module obtains the patient's motion parameters; the image processing module sends the infrared thermal imaging image and the patient's motion parameters to the host computer; the host computer analyzes the patient's motion status according to the patient's motion parameters.

[0062] Furthermore, in one embodiment, the image processing module identifies the patient area and marks the positions of the head, shoulders, hips, and feet of the human body. The identification method is as follows:

[0063] The captured infrared thermal imaging images are preprocessed by using a median filter algorithm to remove salt and pepper noise from the images. The images are then grayscaled to convert the color images into grayscale images. An adaptive threshold segmentation algorithm is used to automatically determine the threshold based on the grayscale value distribution of different areas in the image, segmenting the image into foreground and background parts, and preliminarily extracting the area where the human body is located.

[0064] Perform morphological processing on the segmented human body area, using dilation and erosion operations to remove small noise points and fill the holes in the human body area, calculate the connected areas of the human body area, and determine the outline of the human body;

[0065] Using a joint recognition algorithm based on thermal imaging features, the patient's head, shoulders, hips, and feet are identified and marked, including:

[0066] When identifying the head, the highest point of the human body outline is found and combined with the thermal distribution characteristics, the head is determined to be located on the condition that the head is one of the areas with higher temperature on the human body and appears as a brighter area in the thermal image;

[0067] When identifying the shoulder, based on the characteristic that the heat distribution of the neck is narrow and that of the shoulder is wide, the search is expanded to both sides of the neck, looking for the area with the wide heat distribution as the shoulder location;

[0068] When identifying the crotch, look for an area in the lower area of ​​the human body with a relatively narrow heat distribution and located near the center line of the human body as the crotch location;

[0069] When identifying the feet, the area at the bottom of the human body outline is used as the location of the feet; the identified positions of the head, shoulders, hips, and feet are marked.

[0070] Furthermore, in one embodiment, the three-dimensional calculation module uses two infrared thermal imagers to perform three-dimensional position calculation, specifically in the following manner:

[0071] The imaging planes of the two infrared thermal imagers are calibrated in a pre-set spatial coordinate system within the ward to determine their positions and attitude parameters within the spatial coordinate system. Based on the imaging parameters and relative positional relationship of the two infrared thermal imagers, a spatial grid model is established to divide the monitoring area within the ward into multiple regular spatial grid cells, each with a unique three-dimensional coordinate identifier.

[0072] The image processing module extracts the patient area and the positions of the head, shoulders, hips, and feet, and then marks the position information in the images captured by the two infrared thermal imagers. The 3D calculation module calculates the 3D position coordinates of the patient and each of their joints in the spatial coordinate system based on the marked positions in the images of the two infrared thermal imagers, combined with the calibrated position and posture parameters, using the principle of triangulation.

[0073] The calculated three-dimensional position information is matched with the pre-established spatial grid model to determine the grid cells where the patient and his or her joints are located.

[0074] Furthermore, in one embodiment, the parameters of the infrared thermal imager have the following limitations:

[0075] The image resolution of an infrared thermal imager ranges from [X1×Y1] pixels to [X2×Y2] pixels, where X1 and Y1 represent the minimum resolution at which the outline of a human body can be accurately extracted, and X2 and Y2 represent the maximum resolution at which facial features cannot be clearly distinguished.

[0076] Furthermore, in one embodiment,

[0077] The calculation formula for the lower limit of resolution [X1×Y1] is:

[0078] X1 = ceil((W × n) / L);

[0079] Y1 = ceil((H × n) / H);

[0080] ceil represents rounding up; W and H are the width and height of the ward, respectively, to ensure that the thermal imager can cover the entire ward area; L and H are the minimum horizontal and vertical dimensions of the human body, which are used to determine the minimum pixel coverage of the human body in the image; n is the number of pixels required to capture the temperature difference, calculated based on the thermal imager's thermal sensitivity NETD and the minimum temperature difference ΔT between the human body and the background, n = ceil(ΔT / NETD);

[0081] The calculation formula for the upper limit of resolution [X2×Y2] is:

[0082] X2=floor((D×θ) / (d×π / 180));

[0083]

[0084] floor means rounding down; D is the monitoring distance; θ and are the horizontal and vertical field of view of the thermal imager; d is the diameter of the facial key features, which is used to determine the maximum pixel size of the facial features in the image to ensure that the face cannot be distinguished.

[0085] Furthermore, in one embodiment, the minimum horizontal dimension of the human body is H = 0.5 m, the height of the ward is H = 2 m, the minimum vertical dimension of the human body is H = 1 m, the temperature difference between the human body and the background is ΔT = 0.5° C., and the NETD of the thermal imager is 50 mK, i.e., 0.05° C., then n = ceil(0.5 / 0.05) = 10; and it is calculated that X1 = ceil(4×10 / 0.5) = 80 pixels, and Y1 = ceil(2×10 / 1) = 20 pixels;

[0086] Monitoring distance D = 2m, horizontal field of view angle θ = 45°, vertical field of view angle Facial feature diameter d = 2 cm = 0.02 m.

[0087] Furthermore, in one embodiment, the host computer performs motion state analysis on the patient's motion parameters. The host computer receives the infrared thermal imaging image and the patient's motion parameters sent by the image processing module and sends them to the data processing module. The data processing module determines whether the patient is in bed, sitting or walking state based on the patient's motion parameters.

[0088] When the position of the patient's head, shoulders, hips and feet relative to the bed does not change significantly, the patient is judged to be in bed; when the patient's feet are in contact with the bed or the ground, and the patient's hips are in contact with the bed or chair, and the patient's head, shoulders and hips are relatively stable without significant displacement, the patient is judged to be in a sitting state; when the patient's feet are detected to alternately leave the ground and touch the ground, and the body's center of gravity moves back and forth regularly, the patient is judged to be in a walking state;

[0089] In the walking state, the patient's walking ability is further analyzed: the walking ability is assessed by calculating the stride difference between the patient's two feet. The stride difference is defined as the absolute value of the difference between the stride lengths of each step of the left and right feet. The calculation formula is:

[0090] S diff =|S left -S right ∣;

[0091] S left is the left footstep length, S right is the right foot step; the stride is calculated by the change in the landing position of the same foot in consecutive frames; if the stride difference S diff Exceeds the set threshold S th , it is determined that the patient's walking ability is abnormal and requires further attention and intervention. The abnormal walking ability is divided into 3-5 levels according to the range of stride difference.

[0092] Furthermore, in one embodiment, a cloud module and a mobile terminal are also provided;

[0093] The cloud module is used to store and manage the data collected by the monitoring device, including the patient's infrared thermal imaging images, motion parameters and motion status analysis results; users can remotely access the cloud module through a mobile terminal to view the patient's monitoring data;

[0094] The mobile terminal is used to receive patient movement status information and abnormal reminders sent by the host computer; when the host computer determines that the patient's walking ability is abnormal, it sends reminder signals of varying degrees to the mobile terminal according to the degree of abnormality, so as to timely understand the patient's abnormal condition and take corresponding measures.

[0095] The present invention also protects the use of a wireless nursing monitoring device based on infrared identification of stroke patients for monitoring in a stroke patient ward.

[0096] Example 2:

[0097] See also Figure 4 The resolution of an infrared thermal imager and its price are not in a linear relationship, but in a curvilinear relationship due to the manufacturing cost and the market supply, that is, the price of equipment with moderate resolution is low, while the prices of equipment with extremely high and extremely low resolution are high; since the resolution of infrared thermal imaging on the current market is generally high for the application of the present invention, customizing a low-resolution thermal imager will result in higher costs, so the infrared thermal imager includes an infrared thermal imaging sensor and a matrix light-transmitting plate arranged in front of the infrared thermal imaging sensor, the matrix light-transmitting plate is composed of a plurality of microlenses, and the microlenses are arranged horizontally and vertically to form a microlens array; the number of rows and columns of the microlens array is between [X1×Y1] and [X2×Y2], so as to ensure that the image resolution of the infrared thermal imager is between [X1×Y1] pixels and [X2×Y2] pixels.

[0098] This ensures that when the resolution of the lowest-cost thermal imager currently on the market does not meet the requirements, this method can ensure that the imaging resolution meets the requirements without leaking patient privacy.

[0099] Example 3:

[0100] This embodiment specifically introduces the usage method and precautions of the device of the present invention.

[0101] (1) Equipment installation and debugging

[0102] Infrared thermal imager installation

[0103] Depending on the ward layout, two or more infrared thermal imagers are installed on the ceiling or walls of the ward to ensure full coverage of the patient's activity area, including beds, chairs, walkways, etc. The installation height is generally around 2-2.5 meters to ensure clear thermal imaging of the patient's entire body.

[0104] The parameters of the infrared thermal imager are set to ensure that its image resolution is within the calculated range of [X1×Y1] pixels to [X2×Y2] pixels, which can accurately extract the human body contour while not being able to clearly distinguish facial details, effectively protecting the patient's privacy.

[0105] Other device connections and configurations

[0106] The image processing module, 3D calculation module and host computer are connected via a wireless network to ensure the stability and real-time performance of data transmission. At the same time, the host computer is connected and configured with the cloud module so that the monitoring data can be uploaded to the cloud storage in a timely manner.

[0107] Install a dedicated monitoring application on a mobile terminal (such as a medical staff's smartphone or tablet), complete the pairing connection with the host computer and cloud module, and ensure that medical staff can receive patient status information anytime and anywhere.

[0108] (2) Patient monitoring process

[0109] Real-time monitoring and status identification

[0110] The infrared thermal imager continuously collects thermal images in the ward and sends the images to the image processing module at a rate of about 30 frames per second.

[0111] The image processing module uses a preset image recognition algorithm to process and analyze the collected images in real time, accurately identifying the patient's body contours and precisely marking the positions of key parts such as the head, shoulders, hips, and feet.

[0112] Three-dimensional position calculation and motion parameter acquisition

[0113] The three-dimensional calculation module calculates the three-dimensional position coordinates of the patient and its joints in the spatial coordinate system based on the patient position information marked in the images collected by the two infrared thermal imagers, combined with the pre-calibrated thermal imager position and posture parameters, and using the triangulation principle.

[0114] By analyzing the changes in three-dimensional position coordinates in continuous frames, the patient's motion parameters, including motion trajectory, speed, acceleration and other information, are obtained and sent to the host computer.

[0115] Movement status analysis and abnormality judgment

[0116] After receiving the motion parameters, the host computer immediately starts the motion status analysis program and determines whether the patient is in bed, sitting or walking every second based on the preset judgment rules.

[0117] When the patient is walking, the host computer further analyzes his walking ability and evaluates the degree of abnormality in the patient's walking ability by calculating indicators such as the difference in stride length between the left and right feet. Depending on the degree of abnormality, different levels of reminder signals are sent to the mobile terminal.

[0118] Data storage and remote monitoring

[0119] The host computer uploads the patient's thermal imaging images, motion parameters, and motion status analysis results to the cloud module for real-time storage. The cloud module has a large storage capacity and can store patient monitoring data over a long period of time, providing complete historical data for subsequent review and analysis by medical staff.

[0120] Medical staff can use the monitoring app on their mobile devices to view the patient's current movement status in real time and receive abnormal alerts. They can also access historical data stored in the cloud at any time to understand the patient's recovery progress and condition trends, allowing them to adjust treatment and care plans in a timely manner.

[0121] (3) Regular data evaluation and nursing adjustment

[0122] Medical staff systematically evaluate patient monitoring data stored in the cloud weekly, analyzing trends in the patient's movement status over time. Based on the patient's physical condition and rehabilitation goals, they assess whether the patient's recovery progress is meeting expectations. Based on the evaluation results, nursing and rehabilitation plans are adjusted every two weeks to provide patients with more targeted rehabilitation training and care measures to promote the recovery of their physical functions.

[0123] Example 4:

[0124] Based on the above, this embodiment further describes how the device of the present invention detects physiological parameters such as the patient's body temperature and promptly reports abnormal conditions.

[0125] (1) Body temperature monitoring function

[0126] In the ward, infrared thermal imagers not only capture thermal images of patients for motion monitoring but also utilize their thermal imaging capabilities to monitor their body temperature in real time. They accurately measure the patient's surface temperature and analyze the temperature distribution across different parts of the image, particularly key areas like the head and chest. They calculate the average temperature every 30 seconds as the current temperature data and transmit it to a host computer.

[0127] (2) Monitoring of other physiological parameters

[0128] The device also integrates a small physiological parameter monitoring module for monitoring the patient's heart rate, blood oxygen saturation, and other physiological parameters. This module uses a non-contact or minimally invasive sensor, installed in a location easily accessible to the patient (such as a finger-clip blood oxygen sensor) to avoid causing additional discomfort. The monitoring module collects heart rate and blood oxygen saturation data once a minute and transmits it wirelessly to a host computer.

[0129] (3) Abnormal judgment and feedback mechanism

[0130] The host computer has pre-stored thresholds for normal body temperature range (e.g. 36-37°C), normal heart rate range (e.g. 60 to 100 beats / minute), and normal blood oxygen saturation range (e.g. 95% to 100%). When the body temperature, heart rate, or blood oxygen saturation data received by the host computer exceeds the set normal range, the abnormal feedback mechanism is immediately activated:

[0131] Alarm in the ward: The host computer sends out an audible and visual alarm through the connected alarm device to remind the medical staff in the ward to check the patient's condition in time.

[0132] Mobile terminal reminder: At the same time, the host computer sends detailed abnormal information to the medical staff's mobile terminal, including the specific values ​​of the abnormal physiological parameters, the time of the abnormality and the patient's current movement status (such as lying in bed, walking, etc.), so that medical staff can quickly assess the urgency of the abnormality and take corresponding measures.

[0133] Cloud data marking: Abnormal data will be specially marked when stored in the cloud, which is convenient for subsequent review and analysis, and provides a reference for doctors to adjust treatment plans.

[0134] Through this real-time monitoring and multi-level feedback mechanism, the device of the present invention can monitor the patient's movement status while paying comprehensive attention to the patient's physiological health status, ensuring timely detection and treatment of potential health risks, and further improving the quality of care and safety level for stroke patients.

[0135] Through the above application methods, this device can effectively meet the needs of stroke patients in ward monitoring, provide patients with all-round, real-time safety monitoring, and at the same time provide medical staff with accurate and timely patient status information, helping to improve the quality of care and rehabilitation effects of stroke patients.

[0136] Thus far, the description of the above-described embodiments has been provided for the purpose of illustration and description. It is not intended to be exhaustive or to limit the present disclosure. The individual elements or features of a particular embodiment are generally not limited to the particular embodiment, but when applicable, they can be interchanged and used for selected embodiments even if not specifically shown or described. In many aspects, the same elements or features can also be changed. Such changes are not considered to depart from the present disclosure, and all such modifications are intended to be included within the scope of the present disclosure.

[0137] Example embodiments are provided so that the present disclosure will be thorough and will fully convey the scope to those skilled in the art. In order to thoroughly understand the embodiments of the present disclosure, numerous details are set forth, such as examples of specific parts, devices, and methods. It will be apparent to those skilled in the art that specific details need not be used, and the example embodiments may be implemented in many different forms, and neither should be construed as limiting the scope of the present disclosure. In certain example embodiments, well-known processes, well-known device structures, and well-known technologies are not described in detail.

[0138] Here, professional vocabulary is used only for the purpose of describing specific example embodiments and is not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms "a" and "the" used herein may be intended to include the plural forms as well. The terms "including" and "having" are inclusive and therefore specify the presence of the claimed features, wholes, steps, operations, elements and / or components, but do not exclude the presence or additional presence of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof. Unless the order of execution is explicitly indicated, the method steps, processes and operations described herein are not to be interpreted as necessarily needing to be performed in the specific order discussed and shown. It should also be understood that additional or optional steps may be adopted.

Claims

1. A wireless nursing monitoring device based on infrared recognition of stroke patients, comprising an infrared thermal imager, a host computer, an image processing module, and a three-dimensional calculation module; characterized by: The host computer is connected to the image processing module, and the image processing module is connected to the infrared thermal imager and the three-dimensional calculation module; The number of infrared thermal imagers is 2 or more, which are used to collect infrared thermal imaging images of patients in the ward; after the images collected by the infrared thermal imager are sent to the image processing module for image recognition, the patient area in the image is extracted and the position is marked according to the patient area; The three-dimensional calculation module then obtains the patient's motion parameters; the image processing module sends the infrared thermal imaging image and the patient's motion parameters to the host computer; and the host computer analyzes the patient's motion status based on the patient's motion parameters.

2. The wireless nursing monitoring device based on infrared recognition of stroke patients according to claim 1 is characterized in that: The image processing module identifies the patient area and marks the positions of the head, shoulders, hips, and feet of the human body. The identification method is as follows: The captured infrared thermal imaging images are preprocessed by using a median filter algorithm to remove salt and pepper noise from the images. The images are then grayscaled to convert the color images into grayscale images. An adaptive threshold segmentation algorithm is used to automatically determine the threshold based on the grayscale value distribution of different areas in the image, segmenting the image into foreground and background parts, and preliminarily extracting the area where the human body is located. Perform morphological processing on the segmented human body area, using dilation and erosion operations to remove small noise points and fill the holes in the human body area, calculate the connected areas of the human body area, and determine the outline of the human body; Using a joint recognition algorithm based on thermal imaging features, the patient's head, shoulders, hips, and feet are identified and marked, including: When identifying the head, the highest point of the human body outline is found and combined with the thermal distribution characteristics, the head is determined to be located on the condition that the head is one of the areas with higher temperature on the human body and appears as a brighter area in the thermal image; When identifying the shoulder, based on the characteristic that the heat distribution of the neck is narrow and that of the shoulder is wide, the search is expanded to both sides of the neck, looking for the area with the wide heat distribution as the shoulder location; When identifying the crotch, look for an area in the lower area of ​​the human body with a relatively narrow heat distribution and located near the center line of the human body as the crotch location; When identifying the feet, the area at the bottom of the human body outline is used as the location of the feet; the identified positions of the head, shoulders, hips, and feet are marked.

3. The wireless nursing monitoring device based on infrared recognition of stroke patients according to claim 2 is characterized in that: The 3D calculation module uses two infrared thermal imagers to perform 3D position calculations in the following ways: The imaging planes of the two infrared thermal imagers are calibrated in a pre-set spatial coordinate system within the ward to determine their positions and attitude parameters within the spatial coordinate system. Based on the imaging parameters and relative positional relationship of the two infrared thermal imagers, a spatial grid model is established to divide the monitoring area within the ward into multiple regular spatial grid cells, each with a unique three-dimensional coordinate identifier. The image processing module extracts the patient area and the positions of the head, shoulders, hips, and feet, and then marks the position information in the images captured by the two infrared thermal imagers. The 3D calculation module calculates the 3D position coordinates of the patient and each of their joints in the spatial coordinate system based on the marked positions in the images of the two infrared thermal imagers, combined with the calibrated position and posture parameters, using the principle of triangulation. The calculated three-dimensional position information is matched with the pre-established spatial grid model to determine the grid cells where the patient and his or her joints are located.

4. The wireless nursing monitoring device based on infrared recognition of stroke patients according to claim 1 is characterized in that: The parameters of the infrared thermal imager have the following limitations: The image resolution of an infrared thermal imager ranges from [X1×Y1] pixels to [X2×Y2] pixels, where X1 and Y1 represent the minimum resolution at which the human body contour can be accurately extracted, and X2 and Y2 represent the maximum resolution at which facial features cannot be clearly distinguished.

5. The wireless nursing monitoring device based on infrared recognition of stroke patients according to claim 1 is characterized in that: The infrared thermal imager includes an infrared thermal imaging sensor and a matrix light-transmitting plate arranged in front of the infrared thermal imaging sensor. The matrix light-transmitting plate is composed of a plurality of microlenses, which are arranged horizontally and vertically to form a microlens array. The number of rows and columns of the microlens array is between [X1×Y1] and [X2×Y2] to ensure that the image resolution of the infrared thermal imager is between [X1×Y1] pixels and [X2×Y2] pixels.

6. A wireless nursing monitoring device based on infrared recognition of stroke patients according to claim 4 or 5, characterized in that: The calculation formula for the lower limit of resolution [X1×Y1] is: X1 = ceil((W × n) / L); Y1 = ceil((H × n) / H); ceil means rounding up; W and H are the width and height of the ward, respectively, to ensure that the thermal imager can cover the entire ward area; L and H are the minimum horizontal and vertical dimensions of the human body, which are used to determine the minimum pixel coverage of the human body in the image; n is the number of pixels required to capture temperature changes, calculated based on the thermal sensitivity NETD of the thermal imager and the minimum temperature difference ΔT between the human body and the background, n = ceil(ΔT / NETD); The calculation formula for the upper limit of resolution [X2×Y2] is: X2=floor((D×θ) / (d×π / 180)); floor means rounding down; D is the monitoring distance; θ and are the horizontal and vertical field of view of the thermal imager; d is the diameter of the facial key features, which is used to determine the maximum pixel size of the facial features in the image to ensure that the face cannot be distinguished.

7. The wireless nursing monitoring device based on infrared recognition of stroke patients according to claim 6 is characterized in that: The minimum horizontal dimension of a human body is H = 0.5 m, the ward height is H = 2 m, the minimum vertical dimension of a human body is H = 1 m, the temperature difference between the human body and the background is ΔT = 0.5°C, and the NETD of the thermal imager is 50 mk, or 0.05°C. Therefore, n = ceil(0.5 / 0.05) = 10. The calculation yields X1 = ceil(4×10 / 0.5) = 80 pixels, and Y1 = ceil(2×10 / 1) = 20 pixels. Monitoring distance D = 2m, horizontal field of view angle θ = 45°, vertical field of view angle Facial feature diameter d = 2 cm = 0.02 m.

8. The wireless nursing monitoring device based on infrared recognition of stroke patients according to claim 1 or 2, characterized in that: The host computer performs motion state analysis on the patient's motion parameters. The host computer receives the infrared thermal imaging image and the patient's motion parameters sent by the image processing module and sends them to the data processing module. The data processing module determines whether the patient is in bed, sitting or walking state based on the patient's motion parameters; When the position of the patient's head, shoulders, hips and feet relative to the bed does not change significantly, the patient is judged to be in bed; when the patient's feet are in contact with the bed or the ground, and the patient's hips are in contact with the bed or chair, and the patient's head, shoulders and hips are relatively stable without significant displacement, the patient is judged to be in a sitting state; when the patient's feet are detected to alternately leave the ground and touch the ground, and the body's center of gravity moves back and forth regularly, the patient is judged to be in a walking state; In the walking state, the patient's walking ability is further analyzed: the walking ability is assessed by calculating the stride difference between the patient's two feet. The stride difference is defined as the absolute value of the difference between the stride lengths of each step of the left and right feet. The calculation formula is: S diff =∣S left -S right ∣; S left is the left footstep length, S right is the right foot stride; the stride is calculated by the change in the landing position of the same foot in consecutive frames; if the stride difference S diff Exceeds the set threshold S th , it is determined that the patient's walking ability is abnormal and requires further attention and intervention. The abnormal walking ability is divided into 3-5 levels according to the range of stride difference.

9. The wireless nursing monitoring device based on infrared recognition of stroke patients according to claim 1 or 2, characterized in that: There are also cloud modules and mobile terminals; The cloud module is used to store and manage the data collected by the monitoring device, including the patient's infrared thermal imaging images, motion parameters and motion status analysis results; users can remotely access the cloud module through a mobile terminal to view the patient's monitoring data; The mobile terminal is used to receive patient movement status information and abnormal reminders sent by the host computer; when the host computer determines that the patient's walking ability is abnormal, it sends reminder signals of varying degrees to the mobile terminal according to the degree of the abnormality, so as to timely understand the patient's abnormal condition and take corresponding measures.

10. Use of the wireless nursing monitoring device based on infrared recognition of stroke patients according to any one of claims 1 to 9 for monitoring in a stroke patient ward.

Images