A medical power system adaptive hibernation and fast wake-up method

Abstract

The application discloses a medical power supply system adaptive hibernation and rapid wake-up method, relates to the technical field of power networks, and comprises the following steps: acquiring external interaction data in a hospital, obtaining the type of medical equipment required by a patient and medical activities; generating an intention preparation signal, so that the medical power supply in the medical equipment enters a light hibernation state; in the light hibernation state, acquiring the operation intention and operation time point of the medical equipment; in the light hibernation state, generating a secondary wake-up instruction; setting an ultra-fast response capacitor, supplying power, and after the medical power supply is completely woken up, exiting the power supply and charging, and waiting for the next wake-up; collecting real-time load data of the medical power supply in real time, generating a hibernation instruction, and controlling the medical power supply and the medical equipment to enter a deep hibernation state. The application has the effects of improving the intelligentization and adaptive level of hibernation and wake-up control of the medical power supply system, and improving the overall operation efficiency and reliability of the medical power supply system.

Description

Technical Field

[0001] This application relates to the technical field of power networks, and in particular to an adaptive sleep and fast wake-up method for a medical power supply system. Background Technology

[0002] In various medical diagnosis and treatment activities in hospitals, the medical power supply system and the corresponding medical equipment are key infrastructures to ensure the safety of patients and the smooth operation of medical work. The reliability of its power supply and the timeliness of its response directly affect the overall efficiency and safety of medical treatment.

[0003] In existing technologies, to ensure the immediate availability of medical equipment in emergencies, hospital medical power systems typically employ either a continuous power supply mode or a basic standby / sleep mode. However, using a continuous power supply mode results in significant energy waste, increasing hospital operating costs, and accelerating the aging of medical power supplies and internal components. Traditional standby / sleep modes often rely on manual button activation or simple sensor triggering, which frequently involves lengthy startup times. In time-critical medical scenarios such as emergency care and sudden surgeries, delays in power supply and equipment activation can easily cause medical staff to miss optimal treatment opportunities, posing a serious threat to patient safety. Furthermore, existing sleep / wake mechanisms lack intelligent perception of the complex hospital environment and genuine medical intentions, often resulting in frequent invalid wake-ups due to accidental touches by medical staff or movement of unrelated personnel. This not only fails to achieve true energy savings but also reduces system stability. Therefore, improving the intelligence and adaptability of medical power system sleep and wake-up mechanisms to achieve high energy efficiency while ensuring rapid response has become a pressing issue in this field. Summary of the Invention

[0004] The purpose of this invention is to provide an adaptive sleep and fast wake-up method for a medical power supply system to solve the problems mentioned in the background art.

[0005] This application provides an adaptive sleep and fast wake-up method for a medical power system, the method comprising: By acquiring and combining audio data, image and video data, and medical system data from the hospital, external interaction data is obtained. Based on the external interaction data, the types of medical equipment and medical activities that the patient needs to use are determined. Based on the type of medical device and the medical activity, an intention preparation signal is generated and sent to the corresponding medical device, causing the medical power supply in the medical device to enter a shallow sleep state. In the shallow sleep state, target image data around the medical device is acquired, and the intention to operate the medical device and the time point of operation are obtained based on the target image data. In the shallow sleep state, the current change rate and AC impedance waveform of the medical device are collected and compared to obtain the load mutation characteristics. A secondary wake-up command is generated by combining the operation intention, the operation time point and the load mutation characteristics. A high-speed response capacitor is connected in parallel with the medical power supply in the medical device. Based on the secondary wake-up command, the high-speed response capacitor supplies power and waits for the medical power supply to be fully woken up before disconnecting the power supply and charging, waiting for the next wake-up. The real-time load data of the medical power supply is collected in real time. After cross-validating the real-time load data with the external interaction data, a sleep command is generated. The medical power supply and the medical device are controlled to enter a deep sleep state according to the sleep command.

[0006] Preferably, sound data, image and video data, and medical system data from the hospital are acquired and combined to obtain external interaction data. Based on the external interaction data, the type of medical equipment needed by the patient and the steps of the medical activity are determined, specifically as follows: Acquire sound data, image and video data, and medical system data from the hospital, and combine the sound data, image and video data, and medical system data to obtain external interaction data; The audio data includes hospital conversation data and ambulance siren sound data; the image and video data includes video image information of people walking in the hospital entrance, corridors, and outpatient entrances; and the medical system data includes outpatient doctors' diagnostic data, emergency data, and surgical data. Extract patient-related target interaction data from the external interaction data, and extract target human voice conversation data, target person walking video image information, and target diagnostic data from the target interaction data; Based on the target human voice conversation data, the patient's first target device preference information is obtained, and based on the target person's walking video image information, the patient's second target device preference information is obtained; By combining the first target device preference information and the second target device preference information, data verification is performed on the target diagnostic data to obtain the type of medical device required by the patient and the medical activities to be performed.

[0007] Preferably, after obtaining the type of medical device and the steps of the medical activity required by the patient based on the external interaction data, the method further includes: Based on the sound data, extract the sound fluctuation features of the sound data and the human voice communication keywords present in the sound data; Based on the sound fluctuation characteristics, the noise level value of the sound is obtained. Combining the noise level value with the human voice communication keywords, the patient's current first urgency level value and first possible disease type set are obtained. Based on the image and video data, extract the personnel density value, first aid action type, and facial expression of medical personnel from the image and video data; Based on the facial expression image, the tension level of medical staff regarding the patient's current condition is obtained. Combining the tension level value and the personnel density value, the patient's current second urgency level value is obtained. Based on the type of emergency rescue action, the patient's current second possible disease type set is obtained. By combining the first urgency value and the second urgency value, the patient's current urgency value is obtained. The intersection of the first set of possible disease types and the second set of possible disease types is extracted to obtain the patient's current multiple possible target disease types. Based on the urgency level value and multiple possible target disease types, determine the type of medical device the patient currently needs.

[0008] Preferably, the step of generating an intent preparation signal based on the type of medical device and the medical activity, and sending the intent preparation signal to the corresponding medical device to cause the medical power supply in the medical device to enter a shallow sleep state, specifically includes: Based on the type of medical device, extract the device usage information corresponding to the type of medical device; Based on the medical activity, extract the corresponding diagnostic and treatment target information; Combining the device usage information and the diagnostic and treatment target information, an intent preparation signal is generated, and the intent preparation signal is sent to the medical device corresponding to the medical device type and the medical power supply in the medical device; After receiving the intention preparation signal, the medical power supply pre-wakes up its external power supply terminal, putting the external power supply terminal into a shallow sleep state. After receiving the intent preparation signal, the medical device pre-wakes up the internal electrical components and power input terminal, putting the internal electrical components and power input terminal into a shallow sleep state. The shallow sleep state involves shutting down the main power circuit, keeping the drive circuit in microcurrent preheating mode, and maintaining an electrical connection between the power input terminal and the medical power supply.

[0009] Preferably, in the shallow sleep state, the steps of acquiring target image data around the medical device and obtaining the operation intention and operation time point of the medical device based on the target image data are as follows: In the shallow sleep state, video monitoring is performed on the area surrounding the medical device to obtain target image data of the area surrounding the medical device; Extract the action images of medical staff and patients from the target image data; Based on the patient's movements, determine whether the patient has completed the body positioning according to the preset target movement requirements; If it is determined that the patient has completed the body positioning according to the target action requirements, then the action image of the medical staff is identified to obtain the action operation trend and action operation speed of the medical staff on the medical equipment; Based on the action operation trend, the intention to operate the medical device is obtained. Based on the action operation speed, the time point when the medical device is operated is predicted to obtain the operation time point.

[0010] Preferably, in the shallow sleep state, the step of acquiring and comparing the current change rate and AC impedance waveform of the medical device to obtain load mutation characteristics, and generating a secondary wake-up command by combining the operation intention, the operation time point, and the load mutation characteristics, specifically includes: In the shallow sleep state, the current change rate and AC impedance waveform of the medical device are collected; Construct a time axis, place the current change rate and the AC impedance waveform on the time axis, and record the current change trend and impedance change trend of the current change rate and the AC impedance waveform at the same time point; Determine whether the trend of current change and the trend of impedance change are consistent; If it is determined that they are consistent, then the target time period in which the current change rate and the AC impedance waveform maintain a consistent trend is extracted. Within the target time period, the target time points at which the current change rate and the AC impedance waveform simultaneously undergo abrupt changes are recorded, and the current abrupt change characteristics and impedance abrupt change characteristics at the target time points are extracted. By combining the current abrupt change characteristics and impedance abrupt change characteristics, the load abrupt change characteristics are obtained. Based on the stated intention and the stated time of the operation, it is determined whether the medical staff accidentally touched the device. If it is determined that there was no accidental touch by medical staff, the magnitude and duration of the load mutation feature are extracted, and a secondary wake-up command is generated based on the magnitude and duration of the mutation.

[0011] Preferably, a high-speed response capacitor is connected in parallel with the medical power supply in the medical device. Based on the secondary wake-up command, the high-speed response capacitor supplies power and waits for the medical power supply to fully wake up before disconnecting power and charging, awaiting the next wake-up. Specifically, the steps are as follows: A high-speed response capacitor is provided in the medical device, the high-speed response capacitor is connected in parallel with the medical power supply, and the high-speed response capacitor is electrically connected to the power input terminal of the medical device. The secondary wake-up command is sent to the medical device and the medical power supply. After receiving the secondary wake-up command, the high-speed response capacitor supplies power to the power input terminal and waits for the medical power supply to be fully woken up. Before the medical power supply is fully activated, the high-speed response capacitor continuously supplies power to the power input terminal. Once the medical power supply is fully activated, the high-speed response capacitor stops supplying power and switches to the medical power supply for power. The high-speed response capacitor then recharges and waits for the next activation.

[0012] Preferably, the steps of collecting real-time load data of the medical power supply, cross-validating the real-time load data with the external interaction data, generating a sleep command, and controlling the medical power supply and the medical device to enter a deep sleep state according to the sleep command are as follows: Real-time load data of the medical power supply is collected, and the operation is determined based on the real-time load data. If it is determined that the work has ended, the time range of the next work time is obtained based on the external interaction data. Extract the rest time length between the starting point of the time range and the end time of the current work, and determine whether the rest time length exceeds a preset minimum duration threshold. If it is determined that the rest time exceeds the minimum duration threshold, a hibernation command is generated and sent to the medical power supply and the medical device to control the medical power supply and the medical device to enter a deep hibernation state.

[0013] In summary, this application includes at least one of the following beneficial technical effects: By combining audio, video, and medical system data from the hospital, external interaction data is obtained. Based on this, the type of medical equipment and medical activity required by the patient are analyzed, generating an intent preparation signal that is sent to the medical equipment, causing its internal power supply to enter a shallow sleep state. Then, in the shallow sleep state, target image data around the device is acquired to determine the intended operation and the timing of the operation. Simultaneously, the device's current change rate and AC impedance waveform are compared to obtain load change characteristics, and a secondary wake-up command is generated based on this information. Then, using a high-speed response capacitor connected in parallel with the medical power supply, power is supplied first based on the secondary wake-up command. Once the medical power supply is fully awakened, power supply is withdrawn and the capacitor is charged, awaiting the next wake-up. Finally, real-time load data is collected and cross-validated with external interaction data to generate a sleep command that controls the medical power supply and medical equipment to enter a deep sleep state. This improves the intelligence and adaptability of the medical power supply system's sleep and wake-up control, as well as the overall operating efficiency and reliability of the medical power supply system. Attached Figure Description

[0014] Figure 1 This application provides a step flow diagram of an adaptive sleep and fast wake-up method for a medical power system. Detailed Implementation

[0015] The following combination Figure 1 This application will be described in further detail, but the embodiments of the present invention are not limited thereto.

[0016] This application discloses an adaptive sleep and fast wake-up method for a medical power supply system.

[0017] In this embodiment, an adaptive sleep and fast wake-up method for a medical power system is provided, the method comprising: S100: Acquires and combines audio data, image and video data and medical system data in the hospital to obtain external interaction data, and determines the type of medical equipment and medical activities that the patient needs to use based on the external interaction data; S200: Based on the type of medical device and the medical activity, generate an intention preparation signal and send the intention preparation signal to the corresponding medical device, so that the medical power supply in the medical device enters a shallow sleep state. S300: In shallow sleep mode, acquire target image data around the medical device, and obtain the operation intention and operation time of the medical device based on the target image data; S400: In shallow sleep mode, the current change rate and AC impedance waveform of the medical device are collected and compared to obtain the load change characteristics. Combined with the operation intention, operation time point and load change characteristics, a second wake-up command is generated. S500: A high-speed response capacitor is set in parallel with the medical power supply in the medical device. Based on the second wake-up command, the high-speed response capacitor supplies power and waits for the medical power supply to be fully woken up before disconnecting the power supply and charging, waiting for the next wake-up. S600: Real-time load data of medical power supply is collected. After cross-validating the real-time load data with external interaction data, a sleep command is generated. Based on the sleep command, the medical power supply and medical equipment are controlled to enter a deep sleep state.

[0018] By acquiring and combining audio, image, video, and medical system data from the hospital, external interaction data is obtained. Based on this external interaction data, the types of medical equipment needed by the patient and the steps of the medical procedure are determined. Specifically: Acquire audio data, image and video data, and medical system data from the hospital, and combine these data to obtain external interactive data; The audio data includes data on conversations in the hospital and ambulance siren sounds; the video data includes video images of people walking around the hospital entrance, corridors, and outpatient entrances; and the medical system data includes data on outpatient doctors' diagnoses of patients, emergency data, and surgical data. Extract patient-related target interaction data from external interaction data, and extract target human voice conversation data, target personnel walking video image information, and target diagnostic data from the target interaction data; The patient's first target device preference information is obtained based on the target human voice conversation data, and the patient's second target device preference information is obtained based on the target person's walking video image information. By combining the first target device preference information and the second target device preference information, data verification is performed in the target diagnostic data to obtain the type of medical equipment required by the patient and the medical activities to be performed.

[0019] In this application, we take the X-ray machine in the emergency room of a central city hospital as an example. First, the system acquires sound data, image and video data, and medical system data from the hospital. These three types of data are then transmitted via network to the processor and combined to obtain the total external interaction data. Specifically, the sound data includes ambulance siren sound data at 90 decibels and voice conversation data of "preparing for bone X-ray." The image and video data includes video images of five nurses quickly pushing a cart in the hospital emergency room entrance and corridor. The medical system data includes emergency data of "multiple fractures from a car accident" that the doctor has just entered into the computer. Next, the target interaction data directly related to this fracture patient is extracted. Then, based on the voice conversation of "preparing for bone X-ray," the system analyzes and concludes that the patient's primary target device preference information is the X-ray machine. Based on the image information of the nurses pushing the cart towards Emergency Room 3, the system analyzes and concludes that the patient's secondary device preference information is also the X-ray machine. The first and second preference information are then compared side-by-side to confirm complete consistency. Finally, the X-ray machine result is cross-checked against the diagnostic data of "multiple fractures from a car accident" for verification. Because fractures require imaging examinations using an X-ray machine, the verification was successful, and the final calculation showed that the type of medical equipment the patient needed was an X-ray machine, and the upcoming medical activity was a fracture imaging examination.

[0020] After obtaining the types of medical devices and procedures required by the patient based on external interaction data, the following is also included: Based on sound data, we extract the sound fluctuation features of sound data and the keywords of human voice communication in sound data; Based on the characteristics of sound fluctuations, the noise level of the sound is obtained. Combining the noise level value with keywords in human voice communication, the patient's current first urgency level value and the first set of possible disease types are obtained. Based on image and video data, extract personnel density values, types of first aid actions, and facial expressions of medical personnel from the image and video data; Based on facial expression images, the stress level of medical staff regarding the patient's current condition is obtained. Combining the stress level value and the personnel density value, the patient's current second urgency level value is obtained. Based on the type of emergency rescue action, the patient's current second possible disease type set is obtained. By combining the first urgency value and the second urgency value, the patient's current urgency value is obtained. The intersection of the first set of possible disease types and the second set of possible disease types is extracted to obtain the patient's current multiple possible target disease types. Based on the urgency level and multiple possible target disease types, determine the type of medical equipment the patient currently needs.

[0021] In this application, we take the X-ray machine in the emergency room of a central city hospital as an example. First, based on sound data, we extract the sound fluctuation characteristics and the keywords of human communication within it. For example, if the fluctuation is 85 decibels and the keyword is "broken leg," the noise level is calculated to be level 8. Combining the level 8 noise level with the keyword, we arrive at the patient's primary urgency level score of 8, with the primary possible disease type being fracture and trauma. Next, based on image and video data, we extract the nurse's personnel density value as 3 people per square meter, the type of emergency action as immobilizing the thigh, and the facial expression as a furrowed brow and slight sweating. Based on the furrowed brow and slight sweating facial expression assessment, we determine that the nurse is currently in a state of tension, anxiety, and urgency. The system's internal expression judgment model maps the feature of "furrowed brow and slight sweating" to a specific tension score, directly assigning a value of 9. To unify the standard with the tension level, the system needs to convert the physical density into a 10-point criticality score. The assessment combines a 9-point stress level with 3 people per square meter. In this hypothetical model, 0 people per square meter is defined as 0 points, and 4 or more people per square meter is considered an extremely crowded situation, scored out of 10. Based on this range, 3 people per square meter corresponds to a critical score of 8 points. When assessing the emergency situation at the scene, the system assigns equal importance to "personnel emotional state (stress level)" and "scene crowding state (personnel density)," each with a weighting coefficient of 50%. The patient's second urgency score is calculated to be 8.5 points. Based on the immobilized thigh movement, the second possible disease type set is calculated to be fracture and muscle tear. Then, the first and second urgency scores are added together and divided by 2, resulting in a total urgency score of 8.25 points. The intersection of the repeated diseases in the first and second sets is identified, revealing the patient's current target disease type as fracture. Finally, based on the urgency score of 8.25 points and the fracture type, the required medical equipment for the patient is determined to be an X-ray machine.

[0022] Based on the type of medical device and the medical activity, the process of generating an intent preparation signal and sending it to the corresponding medical device to cause the medical power supply in the medical device to enter a shallow sleep state is as follows: Based on the type of medical equipment, extract the equipment usage information corresponding to that type of medical equipment. Based on medical activities, extract the corresponding diagnostic and treatment goal information; Combining equipment usage information and treatment target information, an intent preparation signal is generated and sent to the medical device corresponding to the medical device type and the medical power supply in the medical device. After receiving the intention preparation signal, the medical power supply will pre-wake up its external power supply end, putting the external power supply end into a shallow sleep state. After receiving the intention preparation signal, the medical device will pre-wake up the internal electrical components and power input terminals, putting them into a shallow sleep state. In the shallow sleep state, the main power circuit is turned off, the drive circuit is kept warm with a small current, and the power input terminal is kept electrically connected to the medical power supply.

[0023] In this application, we take the X-ray machine in the emergency room of a central city hospital as an example. First, based on the type of medical equipment, the system looks up its corresponding purpose information as internal radiographic fluoroscopy. Next, based on the medical activity of fracture imaging examination, the system analyzes and extracts the corresponding diagnostic target information as clearly seeing the specific location of the femoral head fracture. Then, combining the purpose of radiographic fluoroscopy with the target of seeing the bone fracture, the computer generates an intent preparation signal containing an activation command. The system sends this intent preparation signal to the X-ray machine via a wired network, and simultaneously to the medical power supply inside the machine. Upon receiving the signal, the medical power supply pre-wakes up its external power supply end, keeping it in a shallow sleep state. Upon receiving the signal, the X-ray machine also pre-wakes up its internal electrical control components and power input end, keeping these two parts in a shallow sleep state as well. In this shallow sleep state, the main power circuit of the equipment is completely shut down, but the drive circuit maintains a microcurrent of 0.1 amps for preheating. Meanwhile, the power input terminal of the X-ray machine and the external power supply terminal of the medical power supply will always maintain a normal physical conductive connection, ready to be powered on and put into operation at any time.

[0024] In a shallow sleep state, the steps for acquiring target image data around the medical device and determining the intended operation and timing of the operation based on this data are as follows: In a shallow sleep state, video monitoring is performed on the area around the medical equipment to obtain target image data around the medical equipment; Extract the action footage of medical staff and patients from the target image data; Based on the patient's movements, determine whether the patient has completed the body positioning according to the preset target movement requirements; If it is determined that the patient has completed the body positioning according to the target action requirements, the action image of the medical staff is identified to obtain the trend and speed of the medical staff's operation of the medical equipment. Based on the trend of action operation, the intention to operate the medical device is obtained. Based on the speed of action operation, the timing of operation of the medical device is predicted, and the timing of operation is obtained.

[0025] In this application, taking the X-ray machine in the emergency room of a central city hospital as an example, while the X-ray machine is in a shallow sleep state with microcurrent preheating, the system uses a wall-mounted camera to monitor the surrounding area in real time, capturing target image data. Then, using image recognition technology, it extracts the nurse's hand movements and the patient's movements from the target image data. Based on the patient's movements, it checks whether the patient's legs are properly positioned according to the doctor's instructions for lying flat. If it determines that the patient has indeed positioned their legs correctly, the system carefully examines the nurse's movements. By analyzing the nurse's arm movement trajectory, it calculates that the nurse's intention to operate the X-ray machine is to immediately press the switch to take an X-ray. Finally, based on a speed of 1.5 meters per second, the system uses the formula of distance divided by speed to predict and calculate that the actual time when the nurse's finger presses the switch is 2 seconds later.

[0026] In a shallow sleep state, the current change rate and AC impedance waveform of the medical device are collected and compared to obtain the load mutation characteristics. The steps for generating a secondary wake-up command, combining the operation intent, the operation time point, and the load mutation characteristics, are as follows: In shallow sleep mode, the current change rate and AC impedance waveform of the medical device are collected; Construct a time axis, place the current rate of change and AC impedance waveform on the time axis, and record the current change trend and impedance change trend of the current rate of change and AC impedance waveform at the same time point; Determine whether the trends of current change and impedance change are consistent; If it is determined that the current change rate and AC impedance waveform are consistent, then extract the target time period in which the trend is consistent. Within the target time period, record the target time points where the current rate of change and AC impedance waveform simultaneously change abruptly, and extract the current change characteristics and impedance change characteristics at the target time points. Combine the current change characteristics and impedance change characteristics to obtain the load change characteristics. Based on the intended action and the timing of the action, determine whether medical staff accidentally touched the device. If it is determined that there was no accidental touch by medical staff, the amplitude and duration of the load mutation feature are extracted, and a secondary wake-up command is generated based on the amplitude and duration of the mutation.

[0027] In this application, an X-ray machine in the emergency room of a central city hospital is used as an example. When the machine is in a light sleep state, internal sensors collect real-time data on the rate of change of current and the AC impedance waveform. A timeline is then created in the computer software, and the rate of change of current and the AC impedance waveform are plotted on the timeline, recording their trends at the same time points. The computer compares these trends to determine if they are consistent. If they are, the system extracts a 3-second period where the trends completely overlap as the target time segment. Within this 3-second target time segment, the system records the target time point where both trends suddenly change simultaneously in the second second. Then, the upward surge in current and the downward decrease in impedance in this second second are extracted and combined to obtain an overall load change characteristic. This, combined with the nurse's intention to take a bone X-ray and the expected operation time point of pressing the button 2 seconds later, leads to the conclusion that the nurse is operating the machine normally and there was no accidental hand contact. Since it was not a mis-touch, the system extracted the magnitude of the 2-amp surge and the duration of the 0.5-second surge from the load mutation characteristics. Based on these two data points, the system ultimately generated a secondary wake-up command to fully power on the machine.

[0028] A high-speed response capacitor is connected in parallel with the medical power supply in the medical device. Based on the secondary wake-up command, the high-speed response capacitor supplies power and waits for the medical power supply to fully wake up before disconnecting from the power supply and charging, waiting for the next wake-up. The specific steps are as follows: A fast-response capacitor is installed in the medical device. The fast-response capacitor is connected in parallel with the medical power supply and is electrically connected to the power input terminal of the medical device. The second wake-up command is sent to the medical device and the medical power supply. After receiving the second wake-up command, the medical power supply responds quickly to the power input terminal by the capacitor and waits for the medical power supply to be fully woken up. Before the medical power supply is fully awakened, the high-speed response capacitor continuously supplies power to the power input terminal. Once the medical power supply is fully activated, the high-speed response capacitor stops supplying power and switches to the medical power supply. The high-speed response capacitor then recharges and waits for the next activation.

[0029] In practice, let's take an X-ray machine in the emergency room of a central city hospital as an example. First, technicians pre-install a 100-farad fast-response capacitor inside the X-ray machine's casing. In the circuit connection, this fast-response capacitor is connected in parallel with the machine's main medical power supply, and it's also directly connected to the X-ray machine's internal power input. When the system generates the aforementioned secondary wake-up command, it immediately sends this command to the X-ray machine and the medical power supply. Upon receiving the secondary wake-up command, the medical power supply, due to its slow startup, immediately uses the 100-farad fast-response capacitor to directly discharge power to the X-ray machine's power input, while waiting for the main medical power supply to fully power up all circuits and fully wake up the machine. During this 1.5-second window before the medical power supply fully wakes up, the fast-response capacitor continuously supplies power to the power input at a current of 3 amps. Once 1.5 seconds have passed and the medical power supply is fully awake and capable of full power output, the rapid response capacitor will automatically disconnect and stop supplying power. The machine's power supply will immediately switch back to normal operation via the medical power supply. Finally, the rapidly response capacitor, which has been de-energized, will recharge itself until it is fully charged, ready for the next wake-up command.

[0030] The process involves real-time acquisition of load data from the medical power supply, cross-validation of this data with external interaction data, generation of a sleep command, and control of the medical power supply and medical equipment into a deep sleep state based on the sleep command. The specific steps are as follows: Real-time load data of the medical power supply is collected, and the operation is terminated based on the real-time load data. If the task is determined to be completed, the time range for the next task is obtained based on external interaction data. Extract the length of the rest period between the starting point of the time range and the end time of the current work, and determine whether the length of the rest period exceeds the preset minimum duration threshold. If the rest period is determined to exceed the minimum duration threshold, a hibernation command is generated and sent to the medical power supply and medical equipment to control them to enter a deep hibernation state.

[0031] In practice, let's take the X-ray machine in the emergency room of a central city hospital as an example. After finishing its work, the system collects real-time load data of the internal medical power supply. For example, it might find that the current power output has dropped to 5 watts. Based on this low load data of 5 watts, the system can clearly determine that the X-ray machine has completely finished its imaging work. If it determines that the work has ended, the system will check the external interaction data of the registration queue in the background. From the data records, it finds that the next patient's scheduled imaging time is between 10:20 AM and 10:30 AM, thus obtaining the time range for the next work session. Next, the system extracts the starting point of this time range, which is 10:20 AM. Then, it calculates the difference between 10:20 AM and the current end time of the work session, 10:05 AM. This calculation determines that there is a 15-minute break before the machine is turned on again. Subsequently, the computer system determines whether this 15-minute break exceeds a pre-set minimum threshold of 10 minutes. If it is determined that the 15 minutes does exceed the 10-minute threshold, the system will automatically generate a hibernation command. Finally, the system sends a hibernation command to the medical power supply and the X-ray machine, and controls the medical power supply to disconnect from the entire medical equipment according to the command, safely entering a deep hibernation state.

[0032] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method for adaptive sleep and fast wake-up in a medical power supply system, characterized in that, include: By acquiring and combining audio data, image and video data, and medical system data from the hospital, external interaction data is obtained. Based on the external interaction data, the types of medical equipment and medical activities that the patient needs to use are determined. Based on the type of medical device and the medical activity, an intention preparation signal is generated and sent to the corresponding medical device, causing the medical power supply in the medical device to enter a shallow sleep state. In the shallow sleep state, target image data around the medical device is acquired, and the intention to operate the medical device and the time point of operation are obtained based on the target image data. In the shallow sleep state, the current change rate and AC impedance waveform of the medical device are collected and compared to obtain the load mutation characteristics. A secondary wake-up command is generated by combining the operation intention, the operation time point and the load mutation characteristics. A high-speed response capacitor is connected in parallel with the medical power supply in the medical device. Based on the secondary wake-up command, the high-speed response capacitor supplies power and waits for the medical power supply to be fully woken up before disconnecting the power supply and charging, waiting for the next wake-up. The real-time load data of the medical power supply is collected in real time. After cross-validating the real-time load data with the external interaction data, a hibernation command is generated. The medical power supply and the medical device are controlled to enter a deep hibernation state according to the hibernation command.

2. The adaptive sleep and fast wake-up method for a medical power supply system according to claim 1, characterized in that, By acquiring and combining audio, image, video, and medical system data from the hospital, external interaction data is obtained. Based on this external interaction data, the types of medical equipment the patient needs to use and the steps of the medical procedure are determined. Specifically: Acquire sound data, image and video data, and medical system data from the hospital, and combine the sound data, image and video data, and medical system data to obtain external interaction data; The audio data includes hospital conversation data and ambulance siren sound data; the image and video data includes video image information of people walking in the hospital entrance, corridors, and outpatient entrances; and the medical system data includes outpatient doctors' diagnostic data, emergency data, and surgical data. Extract patient-related target interaction data from the external interaction data, and extract target human voice conversation data, target person walking video image information, and target diagnostic data from the target interaction data; Based on the target human voice conversation data, the patient's first target device preference information is obtained, and based on the target person's walking video image information, the patient's second target device preference information is obtained; By combining the first target device preference information and the second target device preference information, data verification is performed on the target diagnostic data to obtain the type of medical device that the patient needs to use and the medical activities that will be performed.

3. The adaptive sleep and fast wake-up method for a medical power supply system according to claim 2, characterized in that, After obtaining the type of medical device and the steps of the medical procedure required by the patient based on the external interaction data, the procedure further includes: Based on the sound data, extract the sound fluctuation features of the sound data and the human voice communication keywords present in the sound data; Based on the sound fluctuation characteristics, the noise level value of the sound is obtained. Combining the noise level value with the human voice communication keywords, the patient's current first urgency level value and first possible disease type set are obtained. Based on the image and video data, extract the personnel density value, first aid action type, and facial expression of medical personnel from the image and video data; Based on the facial expression image, the tension level of medical staff regarding the patient's current condition is obtained. Combining the tension level value and the personnel density value, the patient's current second urgency level value is obtained. Based on the type of emergency rescue action, the patient's current second possible disease type set is obtained. By combining the first urgency value and the second urgency value, the patient's current urgency value is obtained. The intersection of the first set of possible disease types and the second set of possible disease types is extracted to obtain the patient's current multiple possible target disease types. Based on the urgency level value and multiple possible target disease types, determine the type of medical device the patient currently needs.

4. The adaptive sleep and fast wake-up method for a medical power supply system according to claim 3, characterized in that, Based on the type of medical device and the medical activity, the step of generating an intent preparation signal and sending the intent preparation signal to the corresponding medical device, causing the medical power supply in the medical device to enter a shallow sleep state, specifically includes: Based on the type of medical device, extract the device usage information corresponding to the type of medical device; Based on the medical activity, extract the corresponding diagnostic and treatment target information; Combining the device usage information and the diagnostic and treatment target information, an intent preparation signal is generated, and the intent preparation signal is sent to the medical device corresponding to the medical device type and the medical power supply in the medical device; After receiving the intention preparation signal, the medical power supply pre-wakes up its external power supply terminal, putting the external power supply terminal into a shallow sleep state. After receiving the intent preparation signal, the medical device pre-wakes up the internal electrical components and power input terminal, putting the internal electrical components and power input terminal into a shallow sleep state. The shallow sleep state involves shutting down the main power circuit, keeping the drive circuit in microcurrent preheating mode, and maintaining an electrical connection between the power input terminal and the medical power supply.

5. The adaptive sleep and fast wake-up method for a medical power supply system according to claim 4, characterized in that, In the shallow sleep state, the steps of acquiring target image data around the medical device and obtaining the operation intention and operation time point of the medical device based on the target image data are as follows: In the shallow sleep state, video monitoring is performed on the area surrounding the medical device to obtain target image data of the area surrounding the medical device; Extract the action images of medical staff and patients from the target image data; Based on the patient's movements, determine whether the patient has completed the body positioning according to the preset target movement requirements; If it is determined that the patient has completed the body positioning according to the target action requirements, then the action image of the medical staff is identified to obtain the action operation trend and action operation speed of the medical staff on the medical equipment; Based on the action operation trend, the intention to operate the medical device is obtained. Based on the action operation speed, the time point when the medical device is operated is predicted to obtain the operation time point.

6. The adaptive sleep and fast wake-up method for a medical power supply system according to claim 5, characterized in that, In the shallow sleep state, the steps of acquiring and comparing the current change rate and AC impedance waveform of the medical device to obtain load mutation characteristics, and generating a secondary wake-up command by combining the operation intention, the operation time point, and the load mutation characteristics are as follows: In the shallow sleep state, the current change rate and AC impedance waveform of the medical device are collected; Construct a time axis, place the current change rate and the AC impedance waveform on the time axis, and record the current change trend and impedance change trend of the current change rate and the AC impedance waveform at the same time point; Determine whether the trend of current change and the trend of impedance change are consistent; If it is determined that they are consistent, then the target time period in which the current change rate and the AC impedance waveform maintain a consistent trend is extracted. Within the target time period, the target time points at which the current change rate and the AC impedance waveform simultaneously undergo abrupt changes are recorded, and the current abrupt change characteristics and impedance abrupt change characteristics at the target time points are extracted. By combining the current abrupt change characteristics and impedance abrupt change characteristics, the load abrupt change characteristics are obtained. Based on the stated intention and the stated time of the operation, it is determined whether the medical staff accidentally touched the device. If it is determined that there was no accidental touch by medical staff, the magnitude and duration of the load mutation feature are extracted, and a secondary wake-up command is generated based on the magnitude and duration of the mutation.

7. The adaptive sleep and fast wake-up method for a medical power supply system according to claim 6, characterized in that, A high-speed response capacitor is connected in parallel with the medical power supply in the medical device. Based on the secondary wake-up command, the high-speed response capacitor supplies power and waits for the medical power supply to fully wake up before disconnecting power and charging, awaiting the next wake-up. Specifically, the steps are as follows: A high-speed response capacitor is provided in the medical device, the high-speed response capacitor is connected in parallel with the medical power supply, and the high-speed response capacitor is electrically connected to the power input terminal of the medical device. The secondary wake-up command is sent to the medical device and the medical power supply. After receiving the secondary wake-up command, the high-speed response capacitor supplies power to the power input terminal and waits for the medical power supply to be fully woken up. Before the medical power supply is fully activated, the high-speed response capacitor continuously supplies power to the power input terminal. Once the medical power supply is fully activated, the high-speed response capacitor stops supplying power and switches to the medical power supply for power. The high-speed response capacitor then recharges and waits for the next activation.

8. The adaptive sleep and fast wake-up method for a medical power supply system according to claim 7, characterized in that, The steps of collecting real-time load data of the medical power supply, cross-validating the real-time load data with the external interaction data, generating a sleep command, and controlling the medical power supply and the medical device to enter a deep sleep state according to the sleep command are as follows: Real-time load data of the medical power supply is collected, and the operation is determined based on the real-time load data. If it is determined that the work has ended, the time range of the next work time is obtained based on the external interaction data. Extract the rest time length between the starting point of the time range and the end time of the current work, and determine whether the rest time length exceeds a preset minimum duration threshold. If it is determined that the rest time exceeds the minimum duration threshold, a hibernation command is generated and sent to the medical power supply and the medical device to control the medical power supply and the medical device to enter a deep hibernation state.

Images