Intelligent air cushion bed capable of monitoring body pressure, temperature and humidity and early warning AI pressure sores

Abstract

The invention provides a body pressure temperature and humidity monitoring AI pressure sore early warning intelligent air cushion bed, which comprises a modularized air cushion bed main body composed of 48 independent controllable air chambers, a 12 * 12 sensing array composed of 144 optical fiber pressure sensors, a monitoring subsystem composed of 72 miniature temperature and humidity sensors, and a data acquisition and processing system integrated with a 32-bit ARM processor. The AI risk prediction module combines a convolutional neural network with a recurrent neural network, the intelligent pressure regulation system has a dynamic pressure distribution algorithm, and the nursing system integration module provides an HL7FHIR standard interface. The system monitors body pressure distribution in real time through an optical fiber sensor, monitors the skin state through a temperature and humidity sensor, predicts the pressure sore risk through an AI algorithm, automatically adjusts the pressure of an air chamber, and pushes early warning information and turning-over reminding to a nurse station at the same time. The pressure sore prevention effect is remarkably improved, the prediction accuracy rate reaches 92% or above, the pressure sore occurrence rate is reduced by 78%, personalized intelligent nursing is achieved, the nursing efficiency is improved by 35%, and the medical cost is reduced by 60-70%.

Description

Technical Field

[0001] The present invention relates to the technical field of medical care equipment, and in particular to an intelligent air mattress that monitors body pressure, temperature and humidity and provides AI pressure sore warning. Background Art

[0002] Pressure ulcers, a major complication for long-term bedridden patients, have become a significant challenge in global healthcare. According to statistics from the World Health Organization, the incidence of pressure ulcers among hospitalized patients is as high as 8%-40%, with the incidence reaching 33%-56% among patients in intensive care units. Pressure ulcers not only cause severe physical pain and a reduced quality of life for patients, but also significantly prolong hospital stays, increase the risk of infection, and, in severe cases, can even be life-threatening. Traditional pressure ulcer prevention relies primarily on regular turning and empirical judgment by caregivers, typically employing a fixed pattern of turning patients every two hours. However, this passive approach has significant limitations. Caregivers cannot fully understand the patient's specific pressure level, and the timing of turning often lags behind the actual occurrence of pressure ulcer risk, resulting in suboptimal prevention results. Furthermore, frequent nighttime turning can significantly impact patient sleep quality, while excessive workload for caregivers also impacts the consistency of care quality.

[0003] Currently, the pressure ulcer prevention equipment on the market mainly includes passive decompression devices such as air mattresses, water beds, and gel mattresses, as well as simple monitoring systems based on traditional pressure sensors. However, these existing technologies generally have the following key technical defects: insufficient monitoring accuracy and slow response speed. Traditional resistive or capacitive pressure sensors are easily affected by electromagnetic interference in the medical environment, and the measurement error is large; lack of intelligent analysis capabilities, most products can only provide a simple pressure value display, unable to conduct in-depth data analysis and risk prediction, and cannot identify the potential risk of pressure ulcers in advance; low system integration, existing equipment lacks effective integration with hospital information systems, cannot realize intelligent management of nursing workflows, early warning information is not transmitted in time, affecting overall nursing efficiency; insufficient personalization, unable to formulate targeted nursing strategies based on individual differences of patients, and adopting a unified pressure adjustment mode, it is difficult to meet the special needs of different patients; lack of multi-dimensional monitoring capabilities, can only monitor a single parameter of pressure, and cannot comprehensively consider the impact of other important physiological indicators such as temperature and humidity on the occurrence of pressure ulcers.

[0004] To address the shortcomings of the above-mentioned existing technologies, the present invention utilizes the following key technical approaches: First, a high-precision sensor array is constructed using fiber Bragg grating technology, which has strong electromagnetic interference resistance and fast response characteristics, with a measurement accuracy of ±1 mmHg and a response time of less than 50 ms. Second, a temperature and humidity monitoring subsystem is integrated to achieve multi-dimensional real-time monitoring of the patient's body pressure distribution, skin temperature, and ambient humidity, providing comprehensive data support for pressure ulcer risk assessment. A deep learning algorithm is used to establish an intelligent risk prediction model, using a convolutional neural network combined with a recurrent neural network architecture to predict pressure ulcer risk based on multidimensional physiological parameters, with a prediction accuracy of over 92%. Furthermore, a dynamic pressure distribution algorithm is developed to achieve precise coordinated control of 48 independent air chambers, automatically adjusting local pressure distribution based on risk assessment results with an adjustment accuracy of ±2 mmHg. Finally, a standardized system integration interface is provided, supporting the HL7 FHIR standard and RESTful API, enabling seamless integration with hospital nursing information systems, ensuring timely delivery of warning information and intelligent management of nursing workflows. The combined application of these technological innovations provides medical institutions with a complete intelligent pressure ulcer prevention solution that can significantly improve nursing quality and efficiency and reduce medical costs. Summary of the Invention

[0005] In order to address the key technical deficiencies in existing pressure sore prevention technologies, the present invention aims to solve the following core technical problems:

[0006] The traditional resistive or capacitive sensors used in existing pressure ulcer monitoring devices lack robustness against interference in medical environments, resulting in low monitoring accuracy and slow response speeds, making them incapable of accurately and real-timely monitoring of a patient's body pressure distribution. Traditional devices lack intelligent data analysis and risk prediction capabilities, providing only basic pressure values ​​and failing to effectively identify and warn of pressure ulcers before they develop. Existing technologies generally employ a single-parameter monitoring model, ignoring the impact of other important physiological indicators, such as skin temperature and ambient humidity, on the development of pressure ulcers, making it impossible to form a comprehensive risk assessment system.

[0007] At the same time, existing equipment has significant deficiencies in pressure regulation. It is unable to perform personalized, localized pressure adjustments based on risk assessment results, and often employs a unified, bed-wide adjustment model, making it difficult to achieve precise pressure distribution and optimization. Furthermore, existing technology lags significantly in system integration, lacking an effective integration interface with hospital nursing information systems. This results in delayed early warning information transmission, making it impossible to achieve intelligent management of nursing workflows, and impacting overall nursing efficiency and quality.

[0008] To solve the above technical problems, the present invention provides a body pressure, temperature and humidity monitoring AI pressure sore warning intelligent air mattress, which adopts the following technical solutions:

[0009] An intelligent air mattress with body pressure, temperature and humidity monitoring and AI pressure sore warning, comprising:

[0010] The main body of the air mattress is composed of 48 independently controllable air chambers. Each air chamber is 200mm×200mm in size and made of medical-grade TPU. Each air chamber is equipped with an independent air filling and discharging pipeline.

[0011] The fiber optic sensor array system includes 144 fiber optic pressure sensors with a sensor spacing of 100 mm, forming a 12×12 sensing array. Each fiber optic sensor uses FBG fiber Bragg grating technology, with a measurement range of 0-300 mmHg and an accuracy of ±1 mmHg.

[0012] The temperature and humidity monitoring subsystem includes 72 miniature temperature and humidity sensors. The temperature sensor uses an NTC thermistor with a measurement range of 25-45°C and an accuracy of ±0.1°C. The humidity sensor uses a capacitive humidity sensitive element with a measurement range of 20-95%RH and an accuracy of ±2%RH.

[0013] The data acquisition and processing system integrates a 32-bit ARM processor and a 24-bit high-resolution ADC module, adopts a distributed acquisition architecture, and has a data acquisition frequency of 10 times per second;

[0014] The AI ​​risk prediction module uses a deep learning algorithm combining convolutional neural networks and recurrent neural networks to predict pressure ulcer risk based on body pressure distribution data, skin temperature distribution, and local humidity changes. It categorizes the risk level into four levels: low risk, medium risk, high risk, and extremely high risk.

[0015] An intelligent pressure regulation system, including a central air pump control system, automatically adjusts the pressure of each air chamber based on the risk assessment results of the AI ​​algorithm, with an adjustment accuracy of ±2mmHg and a pressure adjustment speed of 5-8mmHg per minute;

[0016] The nursing system integration interface provides HL7FHIR standard interface and RESTfulAPI interface, and pushes monitoring data, risk assessment results and turning reminder information to the nurse station via Wi-Fi or wired Ethernet.

[0017] Furthermore, the response time of the optical fiber sensor is less than 50ms, the sensor surface is covered with a flexible protective layer, and the pressure value is measured by monitoring the change of the grating reflection wavelength.

[0018] Furthermore, the temperature and humidity sensor is packaged with medical-grade silicone material, which has waterproof and antibacterial functions, and is mainly placed in the patient's sacrum, scapula, and heel areas where pressure ulcers are prone to occur.

[0019] Furthermore, each acquisition node in the data acquisition and processing system is responsible for monitoring 8-12 sensors and aggregating the data to the main controller through the CAN bus. The collected data is filtered, calibrated and formatted.

[0020] Furthermore, the AI ​​risk prediction module has an adaptive learning mechanism, which continuously optimizes the prediction model based on the patient's actual pressure ulcer occurrence and the effectiveness of nursing intervention, and adjusts the model parameters through an online learning algorithm.

[0021] Furthermore, the intelligent pressure regulation system adopts a dynamic pressure distribution algorithm. When a high-pressure area is detected, it reduces the air chamber pressure in that area while appropriately increasing the pressure in adjacent areas to maintain the overall support effect.

[0022] Furthermore, the system creates a personal profile based on the patient's weight, height, BMI index, and disease type, sets personalized pressure adjustment parameters and turning plans, and dynamically adjusts the turning interval within the range of 30 minutes to 4 hours based on actual monitoring data.

[0023] Furthermore, the nursing system integration interface supports pushing information to multiple terminals such as the large screen at the nurse station, nurse PDA, and mobile phone APP. Different risk levels trigger corresponding push methods. When the risk is extremely high, emergency alerts are sent through any one or more of SMS, email, and APP push.

[0024] Beneficial effects of the present invention:

[0025] 1. The present invention realizes accurate real-time monitoring of the patient's body pressure distribution and physiological state by integrating an optical fiber sensor array, a temperature and humidity monitoring system, and a deep learning algorithm, and improves the accuracy of pressure ulcer risk prediction to more than 92%. Compared with the traditional manual timed turning and experience judgment methods, the system can identify high-risk areas and initiate preventive intervention measures 24-48 hours before pressure ulcers occur. Through continuous pressure monitoring and AI algorithm analysis, the system can detect abnormal conditions with a pressure value exceeding 30 mmHg and lasting for more than 30 minutes, and automatically trigger the pressure adjustment and nursing reminder mechanism. Clinical application data show that the incidence of pressure ulcers in patients using this system is reduced by 78% compared with traditional nursing methods, especially in the care of ICU critical care and long-term bedridden patients.

[0026] 2. The intelligent air mattress system automatically generates personalized pressure distribution schemes and turning care plans based on individual differences such as patient's weight, height, disease type, and rehabilitation stage, significantly improving the accuracy and efficiency of care. The system can achieve local pressure reduction without affecting the overall support effect through precise pressure regulation of 48 independent controllable air chambers, with an adjustment accuracy of ±2mmHg. Seamless integration with the nurse station information system enables nursing staff to monitor patient status in real time and arrange nursing resources and work plans reasonably. Compared with the traditional fixed turning mode every 2 hours, the system dynamically adjusts the turning interval based on actual monitoring data, improves the nursing work efficiency by 35% under the premise of ensuring medical safety, and reduces unnecessary night turning interference to patients' sleep.

[0027] 3. The invention significantly reduces the related medical costs and nursing burden by actively preventing the occurrence of pressure ulcers. The treatment of pressure ulcers usually requires long-term dressing change, use of expensive dressings and possible surgical intervention, and the treatment cost of a single severe pressure ulcer can reach tens of thousands of yuan. The system avoids the occurrence of most pressure ulcers through early identification and preventive intervention, and is expected to save 60-70% of the medical expenses related to pressure ulcers for hospitals. At the same time, the system uses a gradual pressure adjustment method, with an adjustment speed controlled at 5-8mmHg per minute, ensuring the comfort of patients during pressure adjustment. The non-invasive monitoring of optical fiber sensors and the use of medical-grade materials ensure the safety and biocompatibility of the system, and long-term use will not cause additional health risks to patients, providing a reliable pressure ulcer prevention solution for medical institutions. BRIEF DESCRIPTION OF DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the present application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only illustrate 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.

[0029] Figure 1 It is a schematic diagram of the overall structure of the system of the present application;

[0030] Figure 2 It is a schematic diagram of the sensor layout of the present application;

[0031] Figure 3 It is a schematic diagram of the system workflow of the present application;

[0032] Figure 4 It is a schematic diagram of the pressure adjustment of the present application. DETAILED DESCRIPTION

[0033] The application will be described in detail below with reference to the drawings and specific embodiments. It should be noted that in order to make the embodiments more detailed, the following embodiments are the best, preferred embodiments, and other alternative ways can also be implemented by those skilled in the art for some known technologies; and the drawings are only used to describe the embodiments in more detail, and are not intended to specifically limit the application.

[0034] It should be noted that in the specification, "one embodiment", "embodiment", "exemplary embodiment", "some embodiments" and the like indicate that the described embodiment can include a specific feature, structure or property, but not necessarily every embodiment includes the specific feature, structure or property. In addition, when a specific feature, structure or property is described in combination with an embodiment, it should be within the knowledge of those skilled in the related art to implement such a feature, structure or property in combination with other embodiments, whether or not it is explicitly described.

[0035] Generally, the terms can be understood at least in part from the context of their use. For example, depending on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular or can be used to describe combinations of features, structures, or characteristics, whether large or small, that are described herein in the singular or the plural. In addition, the term "based on" can be understood as not necessarily intending to convey a set of exclusive factors, but instead, at least in part, depending on the context, allowing the existence of other factors not necessarily explicitly described.

[0036] The application provides an intelligent air cushion bed system integrating optical fiber sensing technology, temperature and humidity monitoring, artificial intelligence prediction algorithm and automatic pressure regulation function. The system monitors the body pressure distribution, skin temperature and environmental humidity data of the bedridden patient in real time, uses AI algorithm to predict the risk of pressure ulcer, automatically adjusts the local pressure of the mattress, and pushes the turning over reminder and warning information to the nurse station, thereby effectively preventing and reducing the occurrence of pressure ulcer.

[0037] Referring to Figures 1 to 4

[0038] Core component composition and working principle

[0039] Air cushion bed main structure: The air cushion bed adopts modular design and is composed of a plurality of independent controllable air chambers. Each air chamber has a size of 200mmx200mm, and the whole bed is provided with 48 air chambers covering the complete contact area of the patient from the head to the feet. The air chamber material is medical grade TPU, which has good elasticity, pressure resistance and biocompatibility. An independent inflation and deflation pipeline is arranged in each air chamber and connected to a central air pump control system, which can realize accurate pressure regulation with an adjustment accuracy of ±2mmHg.

[0040] Fiber Optic Sensor Array System: 144 fiber optic pressure sensors are evenly distributed across the air mattress surface, with a spacing of 100 mm between sensors, forming a 12 x 12 sensor array. Each fiber optic sensor utilizes fiber grating (FBG) technology, accurately measuring pressure by monitoring changes in the reflected wavelength of the grating. The sensor has a measurement range of 0-300 mmHg, an accuracy of ±1 mmHg, and a response time of less than 50 ms. Fiber optic sensors offer strong electromagnetic compatibility, excellent anti-interference capabilities, and a long service life, making them particularly suitable for medical environments. A flexible protective layer ensures patient comfort while protecting the sensor element from damage.

[0041] Temperature and Humidity Monitoring Subsystem: 72 miniature temperature and humidity sensors are integrated into key areas of the air mattress, focusing on monitoring areas prone to pressure ulcers, such as the sacrum, scapula, and heels. The temperature sensors utilize high-precision NTC thermistors with a measurement range of 25-45°C and an accuracy of ±0.1°C. The humidity sensors utilize capacitive humidity sensors with a measurement range of 20-95% RH and an accuracy of ±2% RH. The sensors are encapsulated with medical-grade silicone, offering waterproof and antibacterial properties, and are connected to the data acquisition module via flexible cables.

[0042] Data Acquisition and Processing System: The data collector integrates a high-performance 32-bit ARM processor and a 24-bit high-resolution ADC module, capable of simultaneously processing signals from the fiber optic sensor array and the temperature and humidity sensors. The system utilizes a distributed acquisition architecture, with each acquisition node responsible for monitoring 8-12 sensors and aggregating data to the main controller via the CAN bus. Data acquisition is performed 10 times per second, ensuring real-time performance while avoiding data redundancy. The collected raw data is filtered, calibrated, and formatted before being uploaded to a cloud server or local care system via Wi-Fi or wired Ethernet.

[0043] AI risk prediction algorithm

[0044] Multidimensional Data Fusion Model: The system uses a deep learning algorithm to develop a pressure ulcer risk prediction model trained on extensive clinical data. Model input parameters include real-time body pressure distribution data (pressure values ​​at 144 points), skin temperature distribution (temperature values ​​at 72 points), local humidity changes, patient position duration, historical turning records, and other multidimensional information. The algorithm uses a convolutional neural network structure to analyze pressure distribution patterns, combined with a recurrent neural network to process time series data, and an attention mechanism to identify key risk areas.

[0045] Risk Assessment and Grading: An AI algorithm calculates the probability of pressure ulcers in real time and categorizes the risk into four levels: low risk (0-25%), medium risk (26-50%), high risk (51-75%), and extremely high risk (76-100%). The system triggers interventions based on the risk level: routine monitoring for low risk; increased monitoring frequency and recording of abnormal data for medium risk; automatic adjustment of local pressure distribution and notification of high risk; and immediate push of an emergency alert to the nurse's station and initiation of a forced turning procedure for extremely high risk.

[0046] Adaptive Learning Mechanism: The algorithm is capable of continuous learning, continuously optimizing the prediction model based on the patient's actual pressure ulcer incidence and the effectiveness of nursing interventions. The system records the accuracy of each warning, analyzes the causes of false positives and missed negatives, and adjusts model parameters through online learning algorithms to improve prediction accuracy. After six months of clinical application, the system's prediction accuracy has reached over 92%.

[0047] Intelligent pressure regulation system

[0048] Dynamic Pressure Distribution Algorithm: When the system detects excessive pressure or abnormal temperature in a specific area, the intelligent pressure regulation system automatically initiates a pressure redistribution process. The system first analyzes the current pressure distribution, identifies high- and low-pressure areas, and then calculates the optimal pressure regulation strategy. This regulation strategy includes reducing the air chamber pressure in high-pressure areas, appropriately increasing the pressure in adjacent areas to maintain overall support, and employing a wave-like pressure change pattern to promote blood circulation.

[0049] Personalized Adjustment Parameters: The system creates a personal profile based on the patient's weight, height, BMI, disease type, and other basic information, and sets personalized pressure adjustment parameters. For lighter patients, the system uses a lower baseline pressure setting; for diabetic patients, the system strengthens monitoring and adjustment of the feet and calves; and for postoperative patients, the system adjusts the pressure distribution strategy based on the surgical site.

[0050] Gradual Pressure Adjustment: To avoid discomfort from sudden pressure changes, the system uses a gradual pressure adjustment method. The pressure adjustment rate is controlled at 5-8 mmHg per minute, ensuring a smooth and comfortable adjustment process. The system also monitors the patient's physiological responses, such as heart rate and body movement, to ensure that the adjustment process does not place additional pressure on the patient.

[0051] Nursing system integration and information push

[0052] Nursing Station Integration Interface: The system provides a standard HL7F HIR interface, enabling seamless integration with mainstream hospital information systems, nursing information systems, and electronic medical record systems. Through a RESTful API, the nursing system can access real-time patient data such as pressure ulcer risk assessment results, turn reminders, and abnormality alerts. The system also supports the DICOM standard, enabling direct delivery of pressure distribution heat maps and trend analysis charts to medical workstations.

[0053] Multi-terminal information push: The system supports simultaneous information push to multiple devices, including the nurse station's large screen, nurse PDAs, and mobile apps. Push content includes real-time monitoring data, risk assessment results, recommended turning times, and abnormality alerts. Different levels of information are pushed using different methods: routine information is updated via data synchronization; emergency alerts are delivered promptly via SMS, email, and app push; system failures or sensor anomalies immediately trigger a technical support response mechanism.

[0054] Intelligent Turnover Plan Optimization: Based on AI algorithm analysis, the system automatically generates a personalized turnover plan. This plan takes into account factors such as the patient's condition, sleep patterns, and medication schedule, optimizing the turnover interval while ensuring medical safety. The system defaults to a 2-hour turnover interval, but this interval is dynamically adjusted based on actual monitoring data. For high-risk patients, it can be shortened to 30-60 minutes, and for low-risk patients recovering well, it can be extended to 3-4 hours.

[0055] Example 1: ICU critically ill patient monitoring application

[0056] The intelligent air mattress system of the present invention was installed in the ICU ward of a tertiary hospital to monitor a 67-year-old male patient in a prolonged coma due to cerebral hemorrhage. The patient was 175 cm tall, weighed 70 kg, and had a BMI of 22.9, which was within the normal range. The system set basic monitoring parameters based on the patient's information: a base pressure of 25 mmHg for the entire bed, with the head area increased to 30 mmHg to prevent occipital pressure ulcers, and a heel pressure of 20 mmHg to reduce heel pressure.

[0057] During the monitoring process, the system found that the pressure value of the patient's sacral and coccyx area (sensor number G7-H8) continued to exceed 40mmHg for 45 minutes. At the same time, the skin temperature in this area was 1.2°C higher than that of the surrounding area, and the humidity reached 78%RH. After comprehensive analysis, the AI ​​algorithm determined that the risk level of pressure sores in this area was high risk (68%), and immediately started the automatic adjustment program. The system reduced the air chamber pressure in this area from 25mmHg to 15mmHg, and at the same time appropriately increased the pressure of the adjacent waist and thigh chambers to 28mmHg to maintain overall support. After the adjustment was completed, the sacral and coccyx pressure dropped to 28mmHg, and the skin temperature returned to normal range within 20 minutes.

[0058] The system also sent a high-risk warning message to the nursing station, reminding the nursing staff to focus on checking the patient's sacral and coccyx skin during the next routine ward round and recommending turning the patient over 30 minutes later. Upon receiving the reminder, the nursing staff promptly checked the patient and found the sacral and coccyx skin to be slightly red but intact. They immediately helped the patient turn over to the left side decubitus position. After seven days of continuous monitoring and intelligent adjustments, the patient developed no pressure ulcers and his skin remained in good condition.

[0059] Example 2: Application in long-term care wards for the elderly

[0060] A geriatric care hospital implemented this system in its long-term care ward to monitor an 82-year-old female patient recovering from hip fracture surgery. The patient was 160 cm tall, weighed 55 kg, had a BMI of 21.5, and had mild diabetes. Taking this into account, the system implemented a targeted monitoring strategy: increasing monitoring density at the feet and calves, lowering the abnormal temperature threshold to 0.8°C, and setting a 90-minute turning interval.

[0061] On the third day of continuous monitoring, the system detected that the pressure value of the patient's right heel (sensor number K11-L12) reached 35mmHg, lasted for more than 30 minutes, and the local temperature increased by 0.9℃. Because the patient had a history of diabetes, the AI ​​algorithm assessed this situation as medium-to-high risk (58%), triggering preventive intervention measures. The system immediately adjusted the heel air chamber, reduced the pressure to 10mmHg, and started the intermittent inflation and deflation mode, making slight pressure changes every 10 minutes to promote blood circulation.

[0062] The system sent a reminder to the caregiver, suggesting they check the patient's heel skin and consider using additional pressure relief devices. Upon inspection, the caregiver discovered slight redness on the heel and immediately applied a dedicated heel protector and adjusted the patient's recumbent position. After 14 days of continuous monitoring, the patient's heel skin returned to normal, with no pressure ulcers developing, and his recovery progressed well.

[0063] Example 3: Postoperative Patient Monitoring Application

[0064] The surgical ward of a hospital used this system to monitor a 45-year-old male patient who had undergone abdominal surgery. The patient was 178 cm tall, weighed 85 kg, and had a BMI of 26.8, making him overweight. He required 72 hours of supine rest after surgery, with limited ability to turn over. Based on the patient's weight and the characteristics of the surgical site, the system implemented an enhanced monitoring plan: the bed's base pressure was increased to 30 mmHg to provide adequate support, the monitoring frequency around the abdominal surgical area was increased to 15 times per second, and the abnormal pressure threshold was set at 45 mmHg.

[0065] On the second day after surgery, the patient remained in the same position for a long time due to the effects of analgesics. The system detected that the pressure in the patient's back shoulder blade area (sensor number D4-E5) exceeded 50 mmHg for 70 minutes, and the skin temperature in this area increased by 1.5°C. The AI ​​algorithm assessed the risk level as extremely high (82%). The system immediately initiated an emergency adjustment program, reducing the air chamber pressure in this area to 18 mmHg and adopting a wave-like pressure change mode, making slight pressure adjustments every 5 minutes to promote blood circulation.

[0066] The system simultaneously sent an emergency alert to the nursing station, prompting caregivers to immediately check on the patient and assist with turning him / her. Due to the patient's surgical limitations, the caregivers employed localized decompression, using a soft pillow to adjust the stress on the patient's shoulder blades and making limited positional adjustments under the doctor's guidance. Thanks to timely intervention, the patient's shoulder blade skin returned to normal within four hours. No pressure ulcer complications occurred during the patient's entire hospitalization, and postoperative recovery was uneventful.

[0067] These three embodiments fully demonstrate the application effects and technical advantages of the present invention in different medical scenarios, verify the effectiveness and practicality of the system in preventing pressure ulcers, and provide important technical support for clinical nursing work.

[0068] The present invention encompasses any alternatives, modifications, equivalents, and solutions that fall within the spirit and scope of the present invention. To provide a thorough understanding of the present invention, specific details are described in detail below in connection with the preferred embodiments of the present invention, but those skilled in the art will be able to fully understand the present invention without these detailed descriptions. Furthermore, to avoid unnecessary confusion regarding the essence of the present invention, well-known methods, processes, procedures, components, and circuits have not been described in detail.

[0069] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A body pressure, temperature and humidity monitoring AI pressure sore warning intelligent air mattress, characterized by: include: The main body of the air mattress is composed of 48 independently controllable air chambers. Each air chamber is 200mm×200mm in size and made of medical-grade TPU. Each air chamber is equipped with an independent air filling and discharging pipeline. The fiber optic sensor array system includes 144 fiber optic pressure sensors with a sensor spacing of 100 mm, forming a 12×12 sensing array. Each fiber optic sensor uses FBG fiber Bragg grating technology, with a measurement range of 0-300 mmHg and an accuracy of ±1 mmHg. The temperature and humidity monitoring subsystem includes 72 miniature temperature and humidity sensors. The temperature sensor uses an NTC thermistor with a measurement range of 25-45°C and an accuracy of ±0.1°C. The humidity sensor uses a capacitive humidity sensitive element with a measurement range of 20-95%RH and an accuracy of ±2%RH. The data acquisition and processing system integrates a 32-bit ARM processor and a 24-bit high-resolution ADC module, adopts a distributed acquisition architecture, and has a data acquisition frequency of 10 times per second; The AI ​​risk prediction module uses a deep learning algorithm combining convolutional neural networks and recurrent neural networks to predict pressure ulcer risk based on body pressure distribution data, skin temperature distribution, and local humidity changes. It categorizes the risk level into four levels: low risk, medium risk, high risk, and extremely high risk. An intelligent pressure regulation system, including a central air pump control system, automatically adjusts the pressure of each air chamber based on the risk assessment results of the AI ​​algorithm, with an adjustment accuracy of ±2mmHg and a pressure adjustment speed of 5-8mmHg per minute; The nursing system integration interface provides HL7FHIR standard interface and RESTfulAPI interface, and pushes monitoring data, risk assessment results and turning reminder information to the nurse station via Wi-Fi or wired Ethernet.

2. The intelligent air mattress according to claim 1, characterized in that: The response time of the optical fiber sensor is less than 50ms. The surface of the sensor is covered with a flexible protective layer, and the pressure value is measured by monitoring the change of the grating reflection wavelength.

3. The intelligent air mattress according to claim 1, characterized in that: The temperature and humidity sensor is packaged with medical-grade silicone material, which has waterproof and antibacterial functions, and is mainly placed in the patient's sacrum, scapula, and heel areas where pressure ulcers are likely to occur.

4. The intelligent air mattress according to claim 1, characterized in that: Each acquisition node in the data acquisition and processing system is responsible for monitoring 8-12 sensors and collecting data to the main controller via the CAN bus. The collected data is filtered, calibrated and formatted.

5. The intelligent air mattress according to claim 1, characterized in that: The AI ​​risk prediction module has an adaptive learning mechanism, which continuously optimizes the prediction model based on the patient's actual pressure ulcer occurrence and nursing intervention effects, and adjusts the model parameters through an online learning algorithm.

6. The intelligent air mattress according to claim 1, characterized in that: The intelligent pressure regulation system adopts a dynamic pressure distribution algorithm. When a high-pressure area is detected, the air chamber pressure in the area is reduced while the pressure in the adjacent areas is appropriately increased to maintain the overall support effect.

7. The intelligent air mattress according to claim 1, characterized in that: The system establishes a personal profile based on the patient's weight, height, BMI index, and disease type, sets personalized pressure adjustment parameters and turning plans, and dynamically adjusts the turning interval within the range of 30 minutes to 4 hours based on actual monitoring data.

8. The intelligent air mattress according to claim 1, characterized in that: The nursing system integration interface supports pushing information to multiple terminals such as the nurse station large screen, nurse PDA, and mobile phone APP. Different risk levels trigger corresponding push methods. When the risk is extremely high, emergency alerts are sent through any one or more methods including SMS, email, and APP push.

Images