Surgical patient rehabilitation assisting system based on neural network model

Abstract

The invention relates to the technical field of thoracic surgery patient rehabilitation guidance, in particular to a surgical patient rehabilitation assisting system based on a neural network model, which comprises a biosensor module, a motion sensor module, an emotion information analysis module and a rehabilitation effect evaluation module. A patient is monitored in real time through the biosensor module, the treatment effect evaluation unit detects a rehabilitation training value of the patient through the motion sensor module and compares the rehabilitation training value with a rest value of the patient detected by the biosensor module, the recovery condition of the patient is judged, and then a doctor adjusts the medication condition of the patient according to an analysis result. The metabolism state evaluation unit detects the absorption effect of the medicine after the patient takes the medicine, meanwhile, when the patient conducts rehabilitation training, the movement posture of the patient is predicted through the movement track prediction unit, and when the movement posture is wrong, the alarm unit gives an alarm to a worker in time.

Description

Technical Field

[0001] The present invention relates to the technical field of rehabilitation guidance for thoracic surgery patients, and in particular to a surgical patient rehabilitation auxiliary system based on a neural network model. Background Art

[0002] Thoracic surgery is a medical specialty that focuses on the study of organs within the chest cavity, primarily the diagnosis and treatment of esophageal, lung, and mediastinal lesions. The field of breast surgery is also included in this specialty, with lung surgery and esophageal surgery being the main focus. Diseases in the field of thoracic surgery often occur in life-related organs, such as the heart and lungs.

[0003] The existing monitoring rehabilitation assistance system judges the patient's symptoms by detecting the patient's different recovery states and the patient's oral description of the changes in body function. However, when the patient loses his or her language ability, it is impossible to judge the patient's recovery through oral description and detection. At this time, the existing system lacks judgment basis, which will lead to inaccurate judgment. At the same time, when the patient is undergoing rehabilitation training, the posture summarized in the patient's training process cannot be judged in time, which will lead to poor rehabilitation training effect. In addition, due to the patient's loss of language ability during training, the patient's training intensity cannot be adjusted in time, which will also increase the patient's rehabilitation time. In view of this, we propose a surgical patient rehabilitation assistance system based on a neural network model. Summary of the invention

[0004] The purpose of the present invention is to provide a surgical patient rehabilitation assistance system based on a neural network model, which monitors the patient in real time through a biosensor module. A treatment effect evaluation unit detects the value of the patient's rehabilitation training through a motion sensor module and compares it with the value detected by the biosensor module when the patient is resting, so as to judge the patient's recovery status. The communication assistance unit can communicate with patients who have lost their language ability. During the communication, the system can analyze the value output by the motion sensor module with the biosensor module to change the training intensity in the patient's motion sensor module.

[0005] To achieve the above-mentioned object, the present invention provides a surgical patient rehabilitation auxiliary system based on a neural network model, comprising a biosensor module, a motion sensor module, an emotional information analysis module and a rehabilitation effect evaluation module;

[0006] The biosensor module includes a physiological parameter detection unit, a metabolic state assessment unit and an infection prediction unit;

[0007] The physiological parameter detection unit is used to detect the physiological parameters of the patient and output the detected physiological parameters to the treatment and rehabilitation effect evaluation module and the motion sensor module;

[0008] Physiological parameters include: heart rate, blood pressure, blood oxygen saturation;

[0009] Detect the patient's heart rate and observe the light absorption characteristics of the hemoglobin molecules in the blood. The detection is carried out specifically through a photoelectric sensor. The working principle is as follows:

[0010] Step 1: The photoelectric sensor emits infrared light and is composed of infrared LEDs;

[0011] Step 2: The light travels through the skin and is absorbed by the blood. The amount absorbed is proportional to the concentration of hemoglobin molecules in the blood and the blood flow.

[0012] Step 3: The electrical signal generated by the photoelectric sensor is inversely proportional to the light intensity, so as the blood flow changes, the electrical signal will also change;

[0013] The patient's blood pressure is measured by analyzing the pulse waveform to estimate the heart rate. The calculation method is as follows:

[0014] Heart rate = 60 / heart cycle time

[0015] Cardiac cycle time: calculated by measuring the time from the beginning of one cardiac cycle (rising peak) to the beginning of the next cardiac cycle (next rising peak);

[0016] Heart rate: Heart rate is equal to the number of heart beats per minute;

[0017] The blood oxygen saturation of patients is measured by using the light absorption characteristics of hemoglobin. Hemoglobin has two forms: oxygenated hemoglobin and reduced hemoglobin. These two forms have different absorption rates for light of specific wavelengths. The calculation formula is as follows:

[0018]

[0019] SpO2 is the blood oxygen saturation, HbO2 is the concentration of oxygenated hemoglobin, and Hb is the concentration of reduced hemoglobin;

[0020] After the above detection value is detected, the discrete information represented by the digital value is converted into a continuous voltage or current signal and then the detection value is output to the treatment and rehabilitation effect evaluation module, specifically a digital-to-analog conversion method:

[0021] In the digital-to-analog conversion method, each detected value has a corresponding weight, which determines the contribution to the output analog signal. The structure of the digital-to-analog conversion method is an RR ladder network, in which each node has two resistors, one with a resistance value of R and the other with R, allowing each bit to control a specific current value through its corresponding switch. The current value is proportional to the weight of the bit. The current or voltage of each bit is added according to its weight to produce a total analog output signal. The calculation formula is as follows:

[0022]

[0023] in:

[0024] V out is the output voltage of the DAC, V ref is the reference voltage, the digital input value is the digital signal input to the DAC, and the DAC resolution refers to the number of discrete levels that the DAC can represent, expressed in bits;

[0025] The metabolic state evaluation unit is used to detect the degree of absorption of the drug in the patient's body; after the patient takes the drug, the changes in the heart rate and blood pressure detected by the physiological parameter detection unit are observed and compared with those before taking the drug;

[0026] Blood pressure change rate:

[0027]

[0028] BP post is the blood pressure after drug administration, BP pre is the blood pressure before drug administration;

[0029] Heart rate changes:

[0030]

[0031] HR post is the heart rate after drug administration, HR pre is the heart rate before drug administration;

[0032] The infection prediction unit is used to detect biological indicators of patients, so as to detect inflammation and infection at an early stage and adjust the treatment plan in time;

[0033] The formula for calculating white blood cells is as follows:

[0034] White blood cell count = (total white blood cell count / total blood volume) × 10^6

[0035] When inflammation occurs in a patient's body, the immune system is activated and inflammatory cells migrate to the site of inflammation, at which time changes occur in white blood cells.

[0036] The biosensor module also includes a data cleaning unit;

[0037] The data cleaning unit is used to assist the staff to screen out the error values ​​and abnormal values ​​in the test results of the physiological parameter detection unit through the bubble sort algorithm, so as to improve the overall quality of the data set;

[0038] The bubble sort algorithm compares adjacent elements in a data set, and if the current element is larger than the next element, they swap their positions, repeating the process until the elements in the data set are sorted;

[0039] Then the work checks whether the data is within a reasonable range. For example, a person’s age will not exceed years old. If years old appears in the data, it is wrong data.

[0040] The motion sensor module includes a rehabilitation training guidance unit and a motion trajectory prediction unit; the rehabilitation training guidance unit is used for the physiological parameter detection unit to output various values ​​of the patient to the inside of the rehabilitation training guidance unit through the digital-to-analog conversion method in the physiological parameter detection unit, and then determine the appropriate training heart rate interval according to the patient's age, heart rate recovery ability and health status to customize the rehabilitation training plan for the patient, specifically a health algorithm, where the calculation formula is as follows:

[0041] Safe maximum heart rate = 220-age

[0042] Training heart rate lower limit = safe maximum heart rate × 0.5

[0043] Training heart rate upper limit = safe maximum heart rate × 0.85

[0044] At this time, the rehabilitation training plan customized for the patient should be between the patient's lower and upper training heart rate limits;

[0045] When the patient's heart rate during training exceeds the range of the lower limit and the upper limit of the training heart rate, the heart rate detection can be output to the emotion information analysis module through the digital-to-analog conversion method in the biosensor module;

[0046] The motion trajectory prediction unit is used to simulate the patient's physical characteristics and motion laws through a physical simulation method, predict the patient's motion state and predict the patient's incorrect motion mode. The calculation formula of the physical simulation method is as follows:

[0047] Newton's second law: F = ma, where F is force, m is mass, and a is acceleration;

[0048] Dynamic equation: F = dp / dt, where F is force, p is momentum, and t is time;

[0049] When an incorrect movement pattern of the patient is predicted, the error signal is sent to the emotion information analysis module through the digital-to-analog conversion method in the biosensor module.

[0050] The motion sensor module also includes a pain sensing unit;

[0051] The pain sensing unit is used to calculate the pain value endured by the patient when the patient undergoes rehabilitation training through the rehabilitation training guidance unit in combination with the numerical value output by the biosensor module;

[0052] Pain value = (heart rate change + blood pressure change + blood oxygen saturation change) / 3

[0053] At the same time, the patient's physiological indicators show abnormalities. The doctor asks the patient whether he feels pain and uses a pain assessment tool to determine the degree of pain.

[0054] The emotional information analysis module includes an emotion detection unit, a communication assistance unit and an alarm unit;

[0055] The emotion detection unit is used to monitor the patient's psychological emotions in real time through the heart rate variability method, and then output the monitoring signal to the rehabilitation effect evaluation module;

[0056] The heart rate variability method collects the patient's physiological signals, which can reflect the patient's emotional state. For example, an increased heart rate may indicate anxiety or excitement, and changes in skin electrical response may be related to emotional arousal. The specific algorithm is as follows:

[0057]

[0058] x i is the time series data of consecutive heartbeat intervals, N is the total number of data points, and RMSSD is an indicator of HRV, which is related to emotional state;

[0059] The communication auxiliary unit is used to monitor the patient's brain waves by establishing a brain wave model library, and then compare the frequency of the patient's brain waves with the data in the brain wave model library, so that the patient who has lost the ability to communicate with language can communicate. If the patient feels uncomfortable during training through the motion sensor module, the communication auxiliary unit can output the discomfort signal to the alarm unit in time;

[0060] The alarm unit is used to receive the alarm signal of the motion sensor module, and to send a control signal from one device to another device through wireless communication to alarm the doctor through the remote reporting method. The transmission of the output signal of the remote reporting method needs to meet certain power and signal strength requirements;

[0061] The output power is calculated as follows:

[0062] Pout=Pin+Gt+Gr-Lf-Lp

[0063] Output power refers to the signal strength sent from the master device to the slave device, Pin is the input power, Gt is the transmit antenna gain, Gr is the receive antenna gain, Lf is the signal propagation loss, and Lp is the signal loss;

[0064] The signal strength is calculated as follows:

[0065] S=Pout-Lp-10·log10(d)

[0066] Pout is the output power, Lp is the signal propagation loss, and d is the distance.

[0067] The rehabilitation effect evaluation module includes a data receiving unit, a treatment effect evaluation unit and an adjustment unit;

[0068] The data receiving unit is used to receive the data output by the biosensor module, the motion trajectory prediction unit and the emotion detection unit, and to determine whether the patient is infected;

[0069] After the data receiving unit receives the value detected by the physiological parameter detection unit and the actual number of white blood cells of the patient detected by the infection prediction unit;

[0070] During the treatment of the patient, the doctor determines the ideal value of the patient's leukocyte concentration based on the value detected by the physiological parameter detection unit, and then the staff compares the actual number of the patient's leukocytes detected by the infection prediction unit with the ideal value determined by the doctor based on the detection result of the physiological parameter detection unit. If the actual number of leukocytes is greater than the ideal value, it means that the patient is at risk of infection at this time. After comparing the patient's infection risk at this time, the doctor outputs the patient's medication situation to the inside of the adjustment unit through the digital-to-analog conversion method in the physiological parameter detection unit;

[0071] The treatment effect evaluation unit evaluates the patient's recovery degree through the data received by the data receiving unit, specifically, when the actual number of white blood cells in the patient is greater than the ideal value;

[0072] Normal white blood cell count range for adults: 4,000-10,000 / μL;

[0073] Mildly elevated: WBC count between 10,001-20,000 / μL;

[0074] Moderately elevated: WBC count between 20,001-50,000 / μL;

[0075] Severely elevated: WBC count exceeds 50,000 / μL;

[0076] The adjustment unit is used by the staff to adjust the rehabilitation training guidance unit based on the adjustment method;

[0077] The calculation formula of the adjustment method is as follows: Calculate the appropriate training intensity based on the patient's maximum heart rate or maximum oxygen uptake;

[0078] Target heart rate = (MHR - resting heart rate) × intensity percentage + resting heart rate

[0079] MHR is maximum heart rate, estimated by subtracting age, resting heart rate is the patient's heart rate at rest, and intensity percentage is the percentage set according to the training goal;

[0080] The training volume is calculated based on the patient's fitness level and stage of recovery, specifically:

[0081] TVI = training time × training intensity × training frequency

[0082] Training time refers to the duration of each training session, training intensity refers to the difficulty or intensity of the training, and training frequency refers to the number of times you train per week.

[0083] The rehabilitation effect evaluation module also includes a display unit, which is used to sort the data received by the data receiving unit through the bubble sorting algorithm in the biosensor module to facilitate doctors to observe the data.

[0084] The biosensor module monitors the patient in real time, and the treatment and rehabilitation effect evaluation module detects the value of the patient's rehabilitation training through the motion sensor module and compares it with the value detected by the biosensor module when the patient is resting to determine the patient's recovery status. During the rehabilitation training, the patient uses the emotional information analysis module to analyze the rehabilitation training effect and the pain value detected by the motion sensor module through the rehabilitation effect evaluation module. The doctor then adjusts the patient's medication according to the analysis results, and finally detects the absorption effect of the drug after the patient takes the drug through the metabolic state evaluation unit.

[0085] Compared with the prior art, the present invention has the following beneficial effects:

[0086] In the surgical patient rehabilitation assistance system based on the neural network model, the biosensor module is used to monitor the patient in real time. The treatment effect evaluation unit detects the value of the patient's rehabilitation training through the motion sensor module and compares it with the value detected by the biosensor module when the patient is resting to judge the patient's recovery status. During rehabilitation training, the patient uses the communication assistance unit to detect the rehabilitation training effect and the pain sensing unit to detect the patient's pain value and recovery status. The treatment effect evaluation unit then analyzes the results, and then the doctor adjusts the patient's medication according to the analysis results. The metabolic state evaluation unit detects the absorption effect of the drug after the patient takes the drug. At the same time, when the patient is undergoing rehabilitation training, the motion trajectory prediction unit predicts the patient's movement posture. When the movement posture is wrong, the alarm unit promptly alerts the staff. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 It is the overall module principle diagram of the present invention;

[0088] Figure 2 is a schematic diagram of a biosensor module of the present invention;

[0089] Figure 3 This is a schematic diagram of a motion sensor module of the present invention;

[0090] Figure 4 It is the principle diagram of the sentiment information analysis module of the present invention;

[0091] Figure 5 This is a schematic diagram of the rehabilitation effect evaluation module of the present invention.

[0092] The meaning of each number in the figure is:

[0093] 100, biosensor module; 110, physiological parameter detection unit; 120, metabolic state assessment unit; 130, infection prediction unit; 140, data cleaning unit;

[0094] 200, motion sensor module; 210, rehabilitation training guidance unit; 220, motion trajectory prediction unit; 230, pain sensing unit;

[0095] 300, emotional information analysis module; 310, emotion detection unit; 320, communication assistance unit; 330, alarm unit;

[0096] 400, rehabilitation effect evaluation module; 410, data receiving unit; 420, treatment effect evaluation unit; 430, adjustment unit; 440, display unit. DETAILED DESCRIPTION

[0097] The following will be combined with the accompanying drawings in the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0098] A surgical patient rehabilitation assistance system based on a neural network model, comprising a biosensor module 100, a motion sensor module 200, an emotional information analysis module 300 and a rehabilitation effect evaluation module 400;

[0099] The biosensor module 100 includes a physiological parameter detection unit 110, a metabolic state assessment unit 120, and an infection prediction unit 130;

[0100] The physiological parameter detection unit 110 is used to detect the physiological parameters of the patient and output the detected physiological parameters to the treatment and rehabilitation effect evaluation module 400 and the motion sensor module 200;

[0101] Physiological parameters include: heart rate, blood pressure, blood oxygen saturation;

[0102] Detect the patient's heart rate and observe the light absorption characteristics of the hemoglobin molecules in the blood. The detection is carried out specifically through a photoelectric sensor. The working principle is as follows:

[0103] Step 1: The photoelectric sensor emits infrared light and is composed of infrared LEDs;

[0104] Step 2: The light travels through the skin and is absorbed by the blood. The amount absorbed is proportional to the concentration of hemoglobin molecules in the blood and the blood flow.

[0105] Step 3: The electrical signal generated by the photoelectric sensor is inversely proportional to the light intensity, so as the blood flow changes, the electrical signal will also change;

[0106] The patient's blood pressure is measured by analyzing the pulse waveform to estimate the heart rate. The calculation method is as follows:

[0107] Heart rate = 60 / heart cycle time

[0108] Cardiac cycle time: calculated by measuring the time from the beginning of a rising peak in one cardiac cycle to the beginning of the next rising peak in the next cardiac cycle;

[0109] Heart rate: Heart rate is equal to the number of cardiac cycles in 1 minute;

[0110] The blood oxygen saturation of patients is measured by using the light absorption characteristics of hemoglobin. Hemoglobin has two forms: oxygenated hemoglobin and reduced hemoglobin. These two forms have different absorption rates for light of specific wavelengths. The calculation formula is as follows:

[0111]

[0112] SpO2 is the blood oxygen saturation, HbO2 is the concentration of oxygenated hemoglobin, and Hb is the concentration of reduced hemoglobin;

[0113] After the detection of the above detection value, the discrete information represented by the digital value is converted into a continuous voltage or current signal and then the detection value is output to the treatment and rehabilitation effect evaluation module 400, specifically a digital-to-analog conversion method:

[0114] In the digital-to-analog conversion method, each detected value has a corresponding weight, which determines the contribution to the output analog signal. The structure of the digital-to-analog conversion method is an R-2R ladder network, in which each node has two resistors, one with a resistance value of R and the other with a resistance value of 2R, allowing each bit to control a specific current value through its corresponding switch. The current value is proportional to the weight of the bit. The current or voltage of each bit is added according to its weight to produce a total analog output signal. The calculation formula is as follows:

[0115]

[0116] in:

[0117] V out is the output voltage of the DAC, V ref is the reference voltage, the digital input value is the digital signal input to the DAC, and the DAC resolution refers to the number of discrete levels that the DAC can represent, expressed in bits;

[0118] The metabolic state evaluation unit 120 is used to detect the degree of absorption of the drug in the patient's body; after the patient takes the drug, the changes in the heart rate and blood pressure detected by the physiological parameter detection unit 110 are observed and compared with those before taking the drug;

[0119] Blood pressure change rate:

[0120]

[0121] BP post is the blood pressure after drug administration, BP pre is the blood pressure before drug administration;

[0122] Heart rate changes:

[0123]

[0124] HR post is the heart rate after drug administration, HR pre is the heart rate before drug administration;

[0125] The infection prediction unit 130 is used to detect biological indicators of patients, so as to detect inflammation and infection at an early stage and adjust the treatment plan in time;

[0126] The formula for calculating white blood cells is as follows:

[0127] White blood cell count = (total white blood cell count / total blood volume) × 10^6

[0128] When inflammation occurs in a patient's body, the immune system is activated and inflammatory cells migrate to the site of inflammation, at which time changes occur in white blood cells.

[0129] The biosensor module 100 further includes a data cleaning unit 140;

[0130] The data cleaning unit 140 is used to assist the staff to screen out the error values ​​and abnormal values ​​in the detection results of the physiological parameter detection unit 110 through the bubble sort algorithm, so as to improve the overall quality of the data set;

[0131] The bubble sort algorithm compares adjacent elements in a data set, and if the current element is larger than the next element, they swap their positions, repeating the process until the elements in the data set are sorted;

[0132] Then the work checks whether the data is within a reasonable range. For example, a person’s age will not exceed 220 years old. If 300 years old appears in the data, it is wrong data.

[0133] The motion sensor module 200 includes a rehabilitation training guidance unit 210 and a motion trajectory prediction unit 220; the rehabilitation training guidance unit 210 is used for the physiological parameter detection unit 110 to output various values ​​of the patient to the inside of the rehabilitation training guidance unit 210 through the digital-to-analog conversion method in the physiological parameter detection unit 110, and then determine the appropriate training heart rate interval according to the patient's age, heart rate recovery ability and health status to customize the rehabilitation training plan for the patient, specifically a health algorithm, wherein the calculation formula is as follows:

[0134] Safe maximum heart rate = 220-age

[0135] Training heart rate lower limit = safe maximum heart rate × 0.5

[0136] Training heart rate upper limit = safe maximum heart rate × 0.85

[0137] At this time, the rehabilitation training plan customized for the patient should be between the patient's lower and upper training heart rate limits;

[0138] When the patient's heart rate during training exceeds the range of the training heart rate lower limit and the training heart rate upper limit, the heart rate defense can be output to the emotion information analysis module 300 through the digital-to-analog conversion method in the biosensor module 100;

[0139] The motion trajectory prediction unit 220 is used to simulate the patient's physical characteristics and motion rules through a physical simulation method, predict the patient's motion state and predict the patient's incorrect motion mode. The calculation formula of the physical simulation method is as follows:

[0140] Newton's second law: F = ma, where F is force, m is mass, and a is acceleration;

[0141] Dynamic equation: F = dp / dt, where F is force, p is momentum, and t is time;

[0142] When an incorrect movement pattern of the patient is predicted, the error signal is sent to the emotion information analysis module 300 through the digital-to-analog conversion method in the biosensor module 100.

[0143] The motion sensor module 200 also includes a pain sensing unit 230;

[0144] The pain sensing unit 230 is used to calculate the pain value suffered by the patient when the patient undergoes rehabilitation training through the rehabilitation training guidance unit 210 in combination with the numerical value output by the biosensor module 100;

[0145] Pain value = (heart rate change + blood pressure change + blood oxygen saturation change) / 3

[0146] At the same time, the patient's physiological indicators show abnormalities. The doctor asks the patient whether he feels pain and uses a pain assessment tool to determine the degree of pain.

[0147] The emotional information analysis module 300 includes an emotion detection unit 310, a communication assistance unit 320 and an alarm unit 330;

[0148] The emotion detection unit 310 is used to monitor the patient's psychological emotions in real time through the heart rate variability method, and then output the monitoring signal to the rehabilitation effect evaluation module 400;

[0149] The heart rate variability method collects the patient's physiological signals, which can reflect the patient's emotional state. For example, an increased heart rate may indicate anxiety or excitement, and changes in skin electrical response may be related to emotional arousal. The specific algorithm is as follows:

[0150]

[0151] x i is the time series data of consecutive heartbeat intervals, N is the total number of data points, and RMSSD is an indicator of HRV, which is related to emotional state;

[0152] The communication auxiliary unit 320 is used to monitor the patient's brain waves by establishing a brain wave model library, and then compare the frequency of the patient's brain waves with the data in the brain wave model library, so that the patient who has lost the ability to communicate can communicate. If the patient feels uncomfortable during training through the motion sensor module 200, the patient can output the discomfort signal to the alarm unit 330 in time through the communication auxiliary unit 320;

[0153] The alarm unit 330 is used to receive the alarm signal of the motion sensor module 200, and to send the control signal from one device to another device through wireless communication to the doctor through the remote reporting method. The transmission of the output signal of the remote reporting method needs to meet certain power and signal strength requirements;

[0154] The output power is calculated as follows:

[0155] Pout=Pin+Gt+Gr-Lf-Lp

[0156] Output power refers to the signal strength sent from the master device to the slave device, Pin is the input power, Gt is the transmit antenna gain, Gr is the receive antenna gain, Lf is the signal propagation loss, and Lp is the signal loss;

[0157] The signal strength is calculated as follows:

[0158] S=Pout-Lp-10·log10(d)

[0159] Pout is the output power, Lp is the signal propagation loss, and d is the distance.

[0160] The rehabilitation effect evaluation module 400 includes a data receiving unit 410, a treatment effect evaluation unit 420 and an adjustment unit 430;

[0161] The data receiving unit 410 is used to receive the data output by the biosensor module 100, the motion trajectory prediction unit 220 and the emotion detection unit 310, and to determine whether the patient has an infection;

[0162] After the data receiving unit 410 receives the value detected by the physiological parameter detection unit 110 and the actual number of white blood cells of the patient detected by the infection prediction unit 130;

[0163] During the treatment of the patient, the doctor determines the ideal value of the leukocytes in the patient according to the value detected by the physiological parameter detection unit 110. Then, the staff compares the actual number of leukocytes detected by the infection prediction unit 130 with the ideal value determined by the doctor according to the detection result of the physiological parameter detection unit 110. If the actual number of leukocytes is greater than the ideal value, it means that the patient is at risk of infection. After comparing the infection risk of the patient, the doctor outputs the patient's medication status to the inside of the adjustment unit 430 through the digital-to-analog conversion method in the physiological parameter detection unit 110.

[0164] The treatment effect evaluation unit 420 evaluates the patient's recovery degree through the data received by the data receiving unit 410, specifically, when the actual number of white blood cells in the patient is greater than the ideal value;

[0165] Normal white blood cell count range for adults: 4,000-10,000 / μL;

[0166] Mildly elevated: WBC count between 10,001-20,000 / μL;

[0167] Moderately elevated: WBC count between 20,001-50,000 / μL;

[0168] Severely elevated: WBC count exceeds 50,000 / μL;

[0169] The adjustment unit 430 is used for the staff to adjust the rehabilitation training guidance unit 210 based on the adjustment method;

[0170] The calculation formula of the adjustment method is as follows: Calculate the appropriate training intensity based on the patient's maximum heart rate or maximum oxygen uptake;

[0171] Target heart rate = (MHR - resting heart rate) × intensity percentage + resting heart rate

[0172] MHR is maximum heart rate, estimated as 220 minus age, resting heart rate is the patient's heart rate at rest, and intensity percentage is the percentage set according to the training goal;

[0173] The training volume is calculated based on the patient's fitness level and stage of recovery, specifically:

[0174] TVI = training time × training intensity × training frequency

[0175] Training time refers to the duration of each training session, training intensity refers to the difficulty or intensity of the training, and training frequency refers to the number of times you train per week.

[0176] The rehabilitation effect evaluation module 400 also includes a display unit 440, which is used to sort the data received by the data receiving unit 410 through the bubble sort algorithm in the biosensor module 100, so as to facilitate doctors to observe the data.

[0177] The biosensor module 100 monitors the patient in real time, and the treatment rehabilitation effect evaluation module 400 compares the value of the patient's rehabilitation training detected by the motion sensor module 200 with the value detected by the biosensor module 100 when the patient is resting to determine the patient's recovery status. During rehabilitation training, the patient uses the emotional information analysis module 300 to analyze the rehabilitation training effect and the pain value detected by the motion sensor module 200 through the rehabilitation effect evaluation module 400. The doctor then adjusts the patient's medication according to the analysis results, and finally detects the absorption effect of the drug after the patient takes the drug through the metabolic state evaluation unit 120.

[0178] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A surgical patient rehabilitation assistance system based on a neural network model, characterized by: The system comprises a biosensor module (100), a motion sensor module (200), an emotion information analysis module (300) and a rehabilitation effect evaluation module (400), wherein: The biosensor module (100) monitors the patient in real time. The therapeutic rehabilitation effect evaluation module (400) compares the value of the patient's rehabilitation training detected by the motion sensor module (200) with the value detected by the biosensor module (100) when the patient is resting, so as to judge the patient's recovery status. During the rehabilitation training, the patient uses the emotional information analysis module (300) to analyze the rehabilitation training effect and the pain value experienced by the patient detected by the motion sensor module (200) through the rehabilitation effect evaluation module (400). The doctor then adjusts the patient's medication according to the analysis results. Finally, the metabolic state evaluation unit (120) detects the absorption effect of the drug after the patient takes the drug.

2. The surgical patient rehabilitation assistance system based on the neural network model according to claim 1 is characterized in that: The biosensor module (100) comprises a physiological parameter detection unit (110), a metabolic state assessment unit (120) and an infection prediction unit (130); The physiological parameter detection unit (110) is used to detect the physiological parameters of the patient and output the detected physiological parameters to the treatment and rehabilitation effect evaluation module (400) and the motion sensor module (200); Physiological parameters include: heart rate, blood pressure, blood oxygen saturation; Detect the patient's heart rate and observe the light absorption characteristics of the hemoglobin molecules in the blood. The detection is carried out specifically through a photoelectric sensor. The working principle is as follows: Step 1: The photoelectric sensor emits infrared light and is composed of infrared LEDs; Step 2: The light travels through the skin and is absorbed by the blood. The amount absorbed is proportional to the concentration of hemoglobin molecules in the blood and the blood flow. Step 3: The electrical signal generated by the photoelectric sensor is inversely proportional to the light intensity, so as the blood flow changes, the electrical signal will also change; The patient's blood pressure is measured by analyzing the pulse waveform to estimate the heart rate. The calculation method is as follows: Heart rate = 60 / heart cycle time Cardiac cycle time: calculated by measuring the time from the beginning of one cardiac cycle (rising peak) to the beginning of the next cardiac cycle (next rising peak); Heart rate: Heart rate is equal to the number of cardiac cycles in 1 minute; The blood oxygen saturation of patients is measured by using the light absorption characteristics of hemoglobin. Hemoglobin has two forms: oxygenated hemoglobin and reduced hemoglobin. These two forms have different absorption rates for light of specific wavelengths. The calculation formula is as follows: SpO2 is the blood oxygen saturation, HbO2 is the concentration of oxygenated hemoglobin, and Hb is the concentration of reduced hemoglobin; After the detection of the above detection value, the discrete information represented by the digital value is converted into a continuous voltage or current signal and then the detection value is output to the treatment and rehabilitation effect evaluation module (400), specifically a digital-to-analog conversion method: In the digital-to-analog conversion method, each detected value has a corresponding weight, which determines the contribution to the output analog signal. The structure of the digital-to-analog conversion method is an R-2R ladder network, in which each node has two resistors, one with a resistance value of R and the other with a resistance value of 2R, allowing each bit to control a specific current value through its corresponding switch. The current value is proportional to the weight of the bit. The current or voltage of each bit is added according to its weight to produce a total analog output signal. The calculation formula is as follows: in: V out is the output voltage of the DAC, V ref is the reference voltage, the digital input value is the digital signal input to the DAC, and the DAC resolution refers to the number of discrete levels that the DAC can represent, expressed in bits; The metabolic state evaluation unit (120) is used to detect the degree of absorption of the drug in the patient's body; after the patient takes the drug, the changes in the heart rate and blood pressure detected by the physiological parameter detection unit (110) are observed and compared with those before taking the drug; Blood pressure change rate: BP post is the blood pressure after drug administration, BP pre is the blood pressure before drug administration; Heart rate changes: HR post is the heart rate after drug administration, HR pre is the heart rate before drug administration; The infection prediction unit (130) is used to detect biological indicators of patients, so as to detect inflammation and infection at an early stage and adjust the treatment plan in time; The calculation formula for white blood cells is as follows: White blood cell count = (total white blood cell count / total blood volume) × 10^6 When inflammation occurs in a patient's body, the immune system is activated and inflammatory cells migrate to the site of inflammation, at which time changes occur in white blood cells.

3. The surgical patient rehabilitation assistance system based on the neural network model according to claim 1 is characterized in that: The motion sensor module (200) comprises a rehabilitation training guidance unit (210) and a motion trajectory prediction unit (220); the rehabilitation training guidance unit (210) is used for the physiological parameter detection unit (110) to output various values ​​of the patient to the inside of the rehabilitation training guidance unit (210) through a digital-to-analog conversion method, and then determine an appropriate training heart rate interval according to the patient's age, heart rate recovery ability and health status to customize a rehabilitation training plan for the patient, specifically a health algorithm, wherein the calculation formula is as follows: Safe maximum heart rate = 220-age Training heart rate lower limit = safe maximum heart rate × 0.5 Training heart rate upper limit = safe maximum heart rate × 0.85 At this time, the rehabilitation training plan customized for the patient should be between the patient's lower and upper training heart rate limits; When the patient's heart rate during training exceeds the range of the training heart rate lower limit and the training heart rate upper limit, the heart rate change can be output to the emotion information analysis module (300) through the digital-to-analog conversion method in the biosensor module (100); The motion trajectory prediction unit (220) is used to simulate the patient's physical characteristics and motion laws through a physical simulation method, predict the patient's motion state and predict the patient's incorrect motion mode. The calculation formula of the physical simulation method is as follows: Newton's second law: F = ma, where F is force, m is mass, and a is acceleration; Dynamic equation: F = dp / dt, where F is force, p is momentum, and t is time; When an incorrect movement pattern of the patient is predicted, the error signal is sent to the emotion information analysis module (300) via the digital-to-analog conversion method in the biosensor module (100).

4. The surgical patient rehabilitation assistance system based on the neural network model according to claim 1 is characterized in that: The emotional information analysis module (300) comprises an emotion detection unit (310), a communication assistance unit (320) and an alarm unit (330); The emotion detection unit (310) is used to monitor the patient's psychological emotions in real time by using a heart rate variability method, and then output the monitoring signal to the rehabilitation effect evaluation module (400); The heart rate variability method collects the patient's physiological signals, which can reflect the patient's emotional state. For example, an increased heart rate may indicate anxiety or excitement, and changes in skin electrical response may be related to emotional arousal. The specific algorithm is as follows: x i is the time series data of consecutive heartbeat intervals, N is the total number of data points, and RMSSD is an indicator of HRV, which is related to emotional state; The communication auxiliary unit (320) is used to monitor the patient's brain waves by establishing a brain wave model library, and then compare the frequency of the patient's brain waves with the data in the brain wave model library, so that the patient who has lost the ability to communicate with language can communicate. If the patient feels uncomfortable during training through the motion sensor module (200), the patient can promptly output the discomfort signal to the alarm unit (330) through the communication auxiliary unit (320); The alarm unit (330) is used to receive the alarm signal of the motion sensor module (200), and to send a control signal from one device to another device via wireless communication to a remote reporting method to alarm a doctor, and the transmission of the remote reporting method output signal needs to meet certain power and signal strength requirements; The output power is calculated as follows: Pout=Pin+Gt+Gr-Lf-Lp Output power refers to the signal strength sent from the master device to the slave device, Pin is the input power, Gt is the transmit antenna gain, Gr is the receive antenna gain, Lf is the signal propagation loss, and Lp is the signal loss; The signal strength is calculated as follows: S=Pout-Lp-10·log10(d) Pout is the output power, Lp is the signal propagation loss, and d is the distance.

5. The surgical patient rehabilitation assistance system based on the neural network model according to claim 1 is characterized in that: The rehabilitation effect evaluation module (400) comprises a data receiving unit (410), a treatment effect evaluation unit (420) and an adjustment unit (430); The data receiving unit (410) is used to receive data output by the biosensor module (100), the motion trajectory prediction unit (220) and the emotion detection unit (310), and to determine whether the patient is infected; After the data receiving unit (410) receives the value detected by the physiological parameter detection unit (110) and the actual number of white blood cells of the patient detected by the infection prediction unit (130); During the treatment of the patient, the doctor determines the ideal value of the leukocytes in the patient based on the value detected by the physiological parameter detection unit (110). The staff then compares the actual number of leukocytes detected by the infection prediction unit (130) with the ideal value determined by the doctor based on the detection result of the physiological parameter detection unit (110). If the actual number of leukocytes is greater than the ideal value, it indicates that the patient is at risk of infection. After the doctor determines that the patient is at risk of infection, the doctor outputs the patient's medication status to the inside of the adjustment unit (430) through the digital-to-analog conversion method in the physiological parameter detection unit (110). The treatment effect evaluation unit (420) evaluates the patient's recovery degree through the data received by the data receiving unit (410), specifically, the evaluation is performed when the actual number of white blood cells in the patient is greater than the ideal value; The adjustment unit (430) is used by the staff to adjust the rehabilitation training guidance unit (210) based on the adjustment method; The calculation formula of the adjustment method is as follows: Calculate the appropriate training intensity based on the patient's maximum heart rate or maximum oxygen uptake; Target heart rate = (MHR - resting heart rate) × intensity percentage + resting heart rate MHR is maximum heart rate, estimated as 220 minus age, resting heart rate is the patient's heart rate at rest, and intensity percentage is the percentage set according to the training goal; The training volume is calculated based on the patient's fitness level and stage of recovery, specifically: TVI = training time × training intensity × training frequency Training time refers to the duration of each training session, training intensity refers to the difficulty or intensity of the training, and training frequency refers to the number of times you train per week.

6. The surgical patient rehabilitation assistance system based on the neural network model according to claim 1 is characterized in that: The biosensor module (100) further comprises a data cleaning unit (140); The data cleaning unit (140) is used to assist the staff in screening out the erroneous values ​​and abnormal values ​​in the detection results of the physiological parameter detection unit (110) by using a bubble sort algorithm, so as to improve the overall quality of the data set; The bubble sort algorithm compares adjacent elements in a data set, and if the current element is larger than the next element, they swap their positions, repeating the process until the elements in the data set are sorted; Then the work checks whether the data is within a reasonable range. For example, a person’s age will not exceed 220 years old. If 300 years old appears in the data, it is wrong data.

7. The surgical patient rehabilitation assistance system based on the neural network model according to claim 1 is characterized in that: The motion sensor module (200) further includes a pain sensing unit (230); The pain sensing unit (230) is used to calculate the pain value endured by the patient when the patient undergoes rehabilitation training through the rehabilitation training guidance unit (210) in combination with the numerical value output by the biosensor module (100); Pain value = (heart rate change + blood pressure change + blood oxygen saturation change) / 3 At the same time, the patient's physiological indicators show abnormalities. The doctor asks the patient whether he feels pain and uses a pain assessment tool to determine the degree of pain.

8. The surgical patient rehabilitation assistance system based on the neural network model according to claim 1 is characterized in that: The rehabilitation effect evaluation module (400) further includes a display unit (440), which is used to sort the data received by the data receiving unit (410) through a bubble sorting algorithm in the biosensor module (100) to facilitate doctors to observe the data.

Images