Intelligent decompression mattress system based on blood perfusion multi-mode monitoring

Abstract

The invention provides an intelligent decompression mattress system based on blood perfusion multi-mode monitoring, comprising: an array type air bag layer comprising wedge-shaped air bags which are sequentially arranged along a central axis in the length direction of a bed body, and an electromagnetic valve is arranged in an air pipe of each wedge-shaped air bag; the multi-mode sensor monitoring layer comprises a pressure sensor, a temperature sensor, a subcutaneous humidity scanner, a laser Doppler probe and an ultrasonic probe which are arranged on the array type air bag layer; the multi-modal data fusion and clinical decision-making layer comprises a multi-channel data acquisition module and a control module, the multi-channel data acquisition module acquires the multi-modal monitoring data uploaded by the multi-modal sensor monitoring layer, and the control module performs data fusion processing based on the multi-modal monitoring data and establishes an early warning model; the control module sends a risk prompt signal based on a prediction result of the early warning model and controls an electromagnetic valve of the array type air bag layer to conduct inflation and deflation operation, the state of a patient can be effectively judged in real time according to monitored multi-mode data, and therefore the aim of assisting the patient in improving the body position is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent nursing equipment, and particularly relates to an intelligent decompression mattress system based on multi-modal monitoring of blood perfusion. BACKGROUND

[0002] Blood perfusion refers to the amount of circulating blood flowing through the capillaries of an organ or tissue per unit time, and reflects the ability of blood to provide oxygen and nutrients for local tissues and to carry away metabolic waste, and is an important indicator for evaluating organ function. Long-term bedridden patients are particularly prone to blood perfusion problems at pressure sites such as the sacrococcygeal region, heel, and lower limb venous system. Studies have shown that 2 hours of bed rest under local continuous pressure can cause a decrease in skin blood perfusion, such as a decrease in sacrococcygeal blood flow of more than 40, and a pressure of more than 32 mmHg at the bony protuberance can block capillary blood flow, causing tissue hypoxia and necrosis, and further leading to pressure ulcers.

[0003] With the development of technology, the prior art has related products of air bag nursing mattresses, which use sensors to detect long-term bedridden patients and assist patients in adjusting their body positions through a series of air bags. However, the prior art generally uses a single type of sensor for detection, which is single in means and lacks the collection, summarization, and analysis of multi-modal data, resulting in an inability to more accurately analyze the blood perfusion state of patients and to effectively adjust the decompression mattress. SUMMARY

[0004] The purpose of the present application is to provide an intelligent decompression mattress system based on multi-modal monitoring of blood perfusion, which can collect multi-modal detection data to monitor the blood perfusion state of patients in all directions, thereby more timely and accurately assisting patients in adjusting their body positions.

[0005] In order to achieve the above purpose, the present application provides an intelligent decompression mattress system based on multi-modal monitoring of blood perfusion, comprising:

[0006] The arrayed air bag layer comprises wedge-shaped air bags arranged in sequence along the center axis of the bed body in the length direction, the wedge-shaped air bags are divided into left and right groups, and the left and right groups are connected with the air cylinder through the air pipe, and the air pipe is provided with an electromagnetic valve;

[0007] The multi-modal sensor monitoring layer comprises a pressure sensor, a temperature sensor, a subcutaneous humidity scanner, a laser Doppler probe, and an ultrasonic probe arranged on the arrayed air bag layer;

[0008] The multi-modal data fusion and clinical decision layer includes a multi-channel data acquisition module and a control module. The multi-channel acquisition module acquires multi-modal monitoring data transmitted by the multi-modal sensor monitoring layer. The control module performs data fusion processing based on the multi-modal monitoring data, and establishes a warning model. The control module sends a risk prompt signal based on the prediction result of the warning model, and controls the electromagnetic valve of the array air bag layer to perform inflation and deflation operation.

[0009] Further, each group of wedge-shaped air bags can deflate and inflate at the same time, and the control of the electromagnetic valve realizes the simultaneous decompression of the air bags on both sides, the obvious inflation of one side, the non-inflation of one side, the local inflation and deflation, so as to change the pressure of different body parts and assist the patient in body position transformation.

[0010] Further, the pressure sensor is made of flexible material and is installed on each wedge-shaped air bag to collect the local pressure change of the pressure part in real time. The skin temperature sensor is evenly distributed on the wedge-shaped air bag according to the head, chest, abdomen and limbs of the human body, and is installed on the wedge-shaped air bag to collect temperature data periodically. The subcutaneous humidity sensor is covered on each wedge-shaped air bag in a grid shape to periodically scan the subcutaneous humidity and analyze the skin texture and edema condition. The laser Doppler probe is fixed on the surface of the air bag through a flexible support to detect the key part of blood circulation.

[0011] Further, the controller performs data fusion after acquiring the multi-modal data collected by each sensor. The data fusion includes the following steps:

[0012] S101, data preprocessing, data cleaning is performed on the collected multi-modal data, denoising, missing value filling and normalization processing are completed, and a time stamp is set to time align the multi-modal data;

[0013] S102, extracting pressure features, the pressure features include the average pressure of each air bag area, the rate of change of pressure with time, and the peak pressure of each air bag area;

[0014] S103, extracting temperature features, the temperature features include the average temperature, temperature change rate and temperature anomaly of each air bag area;

[0015] S104, extracting humidity features, calculating the average skin humidity, humidity change rate and humidity peak value of each air bag area;

[0016] S105, extracting blood flow features, calculating the average value, change rate and abnormal value of blood flow velocity and flow rate of the key part;

[0017] S106, extracting tissue damage features, marking the area and degree of damage occurrence by ultrasonic detection.

[0018] Further, the early warning model constructs a first index for evaluating tissue metabolic load (TML) according to the pressure feature, the temperature feature and the humidity feature, constructs a second index for evaluating microcirculation efficiency (MES) according to the blood flow feature and the tissue damage feature, and judges the risk level of the patient according to the first index and the second index.

[0019] Further, the risk level of the patient includes:

[0020] low risk, the first index < 0.3, the second index > 0.9, and the damage area = 0;

[0021] medium risk, 0.3 < the first index < 0.6, 0.7 < the second index < 0.9, and the local damage area < 2 cm 2 ;

[0022] high risk, the first index > 0.6, the second index < 0.7, and the local damage area > 2 cm 2 .

[0023] Further, the calculation process of the first index includes:

[0024] TML = a · Pressure norm + β · ΔTemp + γ · ΔHumidity

[0025] wherein a, β, γ are weight coefficients, Pressure norm is the pressure feature, and

[0026]

[0027] wherein P crit is the critical closing pressure, taking a value of 32 mmHg, ΔP is the change rate, and ΔP max is the maximum allowable change rate;

[0028]

[0029] ΔTemp is the temperature feature, T avg is the average temperature, ΔT is the temperature change rate, and T anom / T total is the proportion of abnormal temperature points;

[0030]

[0031] ΔHumidity is the humidity feature, H avg is the average humidity, ΔH is the humidity change rate, and H peak is the humidity peak value.

[0032] Further, the calculation process of the second index includes:

[0033]

[0034] Wherein, BF is blood flow comprehensive index, BF base is blood flow baseline, Injury score is injury degree;

[0035]

[0036] Wherein, Vel avg is blood flow velocity average, Vel base is blood flow velocity baseline, Flow avg is blood flow average, Flow base is blood flow baseline, |Delta Fow / Delta t| is blood flow change rate, Flow range is blood flow change range.

[0037] The application provides an intelligent decompression mattress system based on blood perfusion multi-modal monitoring, which comprises an array type air bag layer, a multi-modal sensor monitoring layer and a multi-modal data fusion and clinical decision layer. BRIEF DESCRIPTION OF DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced below, and obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0039] Figure 1 It is an architecture diagram of an intelligent decompression mattress system based on blood perfusion multi-modal monitoring according to an embodiment of the present application.

[0040] Figure 2 It is a method flow chart of multi-modal data fusion according to an embodiment of the present application.

[0041] Figure 3 is a flow chart of a method for constructing a warning model according to an embodiment of the present application. DETAILED DESCRIPTION

[0042] The technical solutions in the embodiments of the present application will be clearly and completely described with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0043] In the following description, "some embodiments" are described, which describe a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict.

[0044] If similar descriptions of "first / second" appear in the application file, the following description is added. In the following description, the terms "first\second\third" are only to distinguish similar objects, and do not represent a specific order of the objects. It can be understood that "first\second\third" can be interchanged in a specific order or sequence as allowed, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0045] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0046] Figure 1 is a schematic diagram of an intelligent decompression mattress system based on blood perfusion multi-modal monitoring according to an embodiment of the present application. As shown in the figure, the system includes an array air bag layer 1, a multi-modal sensor detection layer 2, and a multi-modal data fusion and clinical decision layer 3. Figure 1 As shown in the figure, the system includes an array air bag layer 1, a multi-modal sensor detection layer 2, and a multi-modal data fusion and clinical decision layer 3.

[0047] Specifically, in the present embodiment, the array air bag layer 1 includes wedge-shaped air bags arranged in sequence along the central axis of the bed body in the length direction. The wedge-shaped air bags are divided into two groups, which are connected to the air cylinder through the air pipe, and the air pipe is provided with an electromagnetic valve. The wedge-shaped air bags are matched with each other, and the outer surface is covered with soft material to protect the sensor.

[0048] Each group of wedge-shaped air bags can be inflated and deflated at the same time, and the control of the electromagnetic valve realizes the simultaneous decompression of the air bags on both sides, the obvious inflation of the air bags on one side, the non-inflation of the air bags on one side, the local inflation and deflation, so as to change the pressure of different body parts and assist the patient in body position transformation.

[0049] Specifically, in this embodiment, the multimodal sensor detection layer 2 includes a pressure sensor, a temperature sensor, a subcutaneous humidity scanner, a laser Doppler probe, and an ultrasonic probe disposed on the array-type airbag layer.

[0050] The pressure sensors, made of flexible material, are installed on each wedge-shaped airbag to collect real-time changes in local pressure at the pressure points. Skin temperature sensors are evenly distributed across the head, chest, abdomen, and limbs of the body and are installed on the wedge-shaped airbags to periodically collect temperature data. Subcutaneous humidity sensors are arranged in a grid pattern on each wedge-shaped airbag to periodically scan subcutaneous humidity and analyze skin texture and edema. The laser Doppler probe is fixed to the surface of the airbags by a flexible bracket to detect key areas of blood circulation.

[0051] Specifically, in this embodiment, the multimodal data fusion and clinical decision-making layer 3 includes a multi-channel data acquisition module and a control module. The multi-channel acquisition module acquires multimodal monitoring data uploaded by the multimodal sensor monitoring layer 2. The control module performs data fusion processing based on the multimodal monitoring data and establishes an early warning model. Based on the prediction results of the early warning model, the control module issues a risk warning signal and controls the solenoid valves of the array-type airbag layer to perform inflation and deflation operations.

[0052] The following is combined with Figure 2 and Figure 3 This embodiment describes the steps involved in multimodal data fusion, data fusion processing of the clinical decision layer 3, and the construction of the early warning model.

[0053] like Figure 2 As shown, the multimodal data fusion method in this embodiment includes the following steps:

[0054] S101, Data Preprocessing.

[0055] In this embodiment, a low-pass filtering algorithm is used to remove noise from the sensors. Missing data points are filled in using linear interpolation. Subsequently, the data from different sensors are standardized to the same dimension, normalized to the 0.1 range, for subsequent fusion processing. To ensure consistent timestamps across all sensor data, time synchronization is used to align the data from different sensors.

[0056] S102, extract pressure features, including the average pressure of each airbag region, the rate of change of pressure over time, and the peak pressure of each airbag region.

[0057] S103, extract temperature features, including the average temperature, temperature change rate, and temperature anomalies of each airbag region.

[0058] S104, extract humidity features, calculate the average skin humidity, humidity change rate and humidity peak of each airbag area.

[0059] S105 extracts blood flow characteristics and calculates the average value, rate of change, and outliers of blood flow velocity and flow rate at key locations.

[0060] S106 extracts tissue damage characteristics and uses ultrasound to detect and mark the area and extent of damage.

[0061] like Figure 2 As shown, the process of establishing the early warning model includes:

[0062] S201 is a first index for evaluating tissue metabolic load (TML) based on pressure, temperature and humidity characteristics.

[0063] Specifically, the calculation process for the first index includes:

[0064] TML = α·Pressure norm +β·ΔTemp+γ·ΔHumidity

[0065] Where α, β, and γ are weighting coefficients, Pressure norm As a pressure characteristic;

[0066]

[0067] Among them, P crit The critical closure pressure is taken as 32 mmHg, and ΔP is the rate of change. max The maximum permissible rate of change;

[0068]

[0069] ΔTemp is a temperature characteristic, T avg Let T be the average temperature, ΔT be the rate of temperature change, and T be the average temperature. anom / T total This represents the percentage of abnormal temperature points.

[0070]

[0071] ΔHumidity is a characteristic of humidity, H avg Let H be the average humidity, ΔH be the rate of change of humidity, and H be the average humidity. peak This represents the peak humidity level.

[0072] S202 is a second index constructed based on blood flow characteristics and tissue damage characteristics to evaluate microcirculation efficiency (MES).

[0073] The calculation process for the second index includes:

[0074]

[0075] BF = (BF baseline - BF average) / BF baseline base BF baseline = blood flow baseline score Injury = injury degree

[0076]

[0077] Vel average = blood flow velocity average avg Vel baseline = blood flow velocity baseline base Flow average = blood flow average avg Flow baseline = blood flow baseline base |ΔFlow / Δt| = blood flow change rate range Flow range = blood flow change range

[0078] S203, determining the risk level of the patient according to the first index and the second index.

[0079] The risk level of the patient includes:

[0080] Low risk, the first index < 0.3, the second index > 0.9, and the injury area = 0.

[0081] Medium risk, 0.3 < the first index < 0.6, 0.7 < the second index < 0.9, and the local injury area < 2 cm 2 .

[0082] High risk, the first index > 0.6, the second index < 0.7, and the local injury area > 2 cm 2 .

[0083] Specifically, the system adds a pressure injury risk prompt indicator light at the bedside: I-level warning - green indicator light; II-level warning - yellow indicator light; III-level warning - orange indicator light; IV-level warning - red indicator light.

[0084] I-level warning: normal monitoring state; II-level warning: low risk registration warning; III-level warning: medium risk level warning; IV-level warning: high risk level warning.

[0085] Although the embodiments of the present application have been shown and described above, it should be understood that the above-mentioned embodiments are exemplary and should not be construed as limiting the present application, and those skilled in the art can make changes, modifications, replacements and variations to the above-mentioned embodiments within the scope of the present application. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. An intelligent decompression mattress system based on blood perfusion multi-modal monitoring, characterized in that, The application relates to a multi-modal pressure support system for patients, which comprises the following parts: an arrayed airbag layer, which comprises wedge-shaped airbags arranged along the central axis of the bed body, the wedge-shaped airbags being divided into two groups, and the airbags being connected with air cylinders through air pipes, and electromagnetic valves being arranged in the air pipes; a multi-modal sensor monitoring layer, which comprises pressure sensors, temperature sensors, subcutaneous humidity scanners, laser Doppler probes and ultrasonic probes arranged on the arrayed airbag layer; a multi-modal data fusion and clinical decision layer, which comprises a multi-channel data acquisition module and a control module, the multi-channel data acquisition module acquires multi-modal monitoring data uploaded by the multi-modal sensor monitoring layer, the control module performs data fusion processing based on the multi-modal monitoring data, establishes a warning model, sends a risk prompt signal based on the prediction result of the warning model, and controls the electromagnetic valves of the arrayed airbag layer to perform air charging and discharging operations.

2. An intelligent reduced pressure wound mattress system based on blood perfusion multi-modal monitoring as defined in claim 1, wherein, Each group of wedge-shaped airbags can simultaneously perform air discharging and air charging, and the control of the electromagnetic valves can realize the following functions: simultaneously reducing the pressure of airbags on both sides, obviously charging the airbags on one side, not charging the airbags on the other side, locally charging and discharging, so as to change the pressure of different body parts and assist patients in position transformation.

3. An intelligent reduced pressure wound mattress system based on blood perfusion multi-modal monitoring as defined in claim 2, wherein, The pressure sensors are made of flexible materials, are arranged on each wedge-shaped airbag, and acquire the local pressure changes of the pressure receiving parts in real time; the skin temperature sensors are evenly distributed on the wedge-shaped airbags according to the head, chest, abdomen and limbs of the human body, acquire temperature data regularly, the subcutaneous humidity sensors are arranged on each wedge-shaped airbag in a grid shape, regularly scan the subcutaneous humidity, analyze the skin texture and edema condition, and the laser Doppler probes are fixed on the surface of the airbags through flexible supports and detect the key parts of blood circulation.

4. An intelligent reduced pressure wound mattress system based on blood perfusion multi-modal monitoring as defined in claim 3, wherein, The controller performs data fusion after acquiring the multi-modal data collected by the sensors, and the data fusion comprises the following steps: S101, data preprocessing, data cleaning is performed on the collected multi-modal data, denoising, missing value filling and normalization processing are completed, and time stamps are set to time-align the multi-modal data; S102, pressure feature extraction, the pressure features include the average pressure of each airbag area, the pressure change rate with time and the peak pressure of each airbag area; S103, temperature feature extraction, the temperature features include the average temperature, temperature change rate and temperature anomaly of each airbag area; S104, humidity feature extraction, the average skin humidity, humidity change rate and humidity peak value of each airbag area are calculated; S105, blood flow feature extraction, the average value, change rate and abnormal value of the blood flow velocity and flow of the key parts are calculated; S106, tissue damage feature extraction, the area and degree of damage occurrence are marked through ultrasonic detection.

5. An intelligent reduced pressure wound mattress system based on blood perfusion multi-modal monitoring as defined in claim 4, wherein, The warning model constructs a first index for evaluating tissue metabolic load (TML) according to the pressure features, temperature features and humidity features, constructs a second index for evaluating microcirculation efficiency according to the blood flow features and tissue damage features, and judges the risk level of the patient according to the first index and the second index.

6. An intelligent reduced pressure wound mattress system based on blood perfusion multi-modal monitoring as defined in claim 5, wherein, The risk level of the patient comprises: low risk, the first index < 0.3, the second index > 0.9, and the damage area = 0; Moderate risk, 0.3 < first index < 0.6, 0.7 < second index < 0.9, local damage area < 2 cm 2 ; High risk, first index > 0.6, second index < 0.7, local damage area > 2 cm 2 .

7. An intelligent reduced pressure wound mattress system based on blood perfusion multi-modal monitoring as defined in claim 5, wherein, the calculation process of the first index comprises: TML = a - Pressure norm + β - ΔTemp + γ - ΔHumidity where a, β, γ are weight coefficients, Pressure norm is a pressure characteristic, where P crit is the critical closing pressure, having a value of 32 mmHg, ΔP is the rate of change, and ΔP max is the maximum allowable rate of change; ΔTemp is the temperature feature, T avg is the average temperature, ΔT is the temperature change rate, T anom / T total is the proportion of abnormal temperature points; ΔHumidity is the humidity characteristic, H avg is the average humidity, ΔH is the humidity change rate, H peak is the humidity peak.

8. An intelligent reduced pressure wound mattress system based on blood perfusion multi-modal monitoring as defined in claim 5, wherein, the calculation process of the second index comprises: wherein BF is the blood flow composite index, BF base is the blood flow baseline, Injury score is the injury degree; wherein Vel avg is the average blood flow velocity, Vel base is the baseline blood flow velocity, Flow avg is the average blood flow, Flow base is the baseline blood flow, |ΔFow / Δt| is the blood flow rate of change, Flow range is the blood flow rate of change range.

Images