Odor feedback method based on artificial intelligence

Abstract

The invention discloses an artificial intelligence-based odor feedback method, which relates to the technical field of odor feedback, and comprises the following steps: dynamically calculating the odor component removal rate of a patient with liver and kidney dysfunction through a CYP2A6 enzyme kinetic model and a kidney replacement therapy adsorption equation, and calculating the odor component removal rate of the patient with liver and kidney dysfunction in combination with the liver and kidney metabolism overload threshold constraint of a taboo mapping table. The toxicity risk is reduced, and the problem of uncontrollable metabolism in a traditional method is solved; a pulse release waveform is designed based on an olfactory sensitivity attenuation model, reverse airflow scouring is coupled, the receptor recovery period is shortened, the EEG response incubation period is shortened, and the desensitization effect caused by continuous release is broken through.

Description

Technical Field

[0001] The present invention relates to the technical field of odor feedback, in particular to an odor feedback method based on artificial intelligence. Background Art

[0002] At present, the incidence of nocturnal delirium in ICU patients is as high as 80% due to environmental stimulation, pain, metabolic disorders and other factors. Existing aromatherapy mostly uses sedative scents such as lavender and chamomile, which are released continuously for 6-8 hours at night through a nebulizer at a fixed concentration, such as 2% essential oil solution, to regulate limbic system activity (amygdala prefrontal cortex connection) through the olfactory pathway, thereby reducing delirium-related agitated behavior.

[0003] However, the patient's physiological state fluctuates dramatically at night, such as pain outbreaks and drug metabolism peaks, but the intensity of odor release is fixed. In this case, the nurse needs to manually turn off the nebulizer and rely on manual supervision. In addition, the continuous stimulation of a single odor causes olfactory fatigue, and the intervention effect will drop sharply. Studies have shown that the sedative effect of lavender decays by 62% after 45 minutes. Although intermittent release can be used, the cycle setting relies on experience and cannot match the differences in individual receptor recovery speeds.

[0004] It can be seen that the current aromatherapy is essentially open-loop control, which simplifies the complex neurometabolic system into a linear model and intervenes in the dynamic pathophysiological process through fixed parameters, with poor results. Summary of the invention

[0005] In view of the above existing problems, the present invention is proposed.

[0006] The present invention provides an odor feedback method based on artificial intelligence to solve

[0007] Existing aromatherapy is unable to adjust to the pain outbreak period and drug metabolism peak when treating symptoms such as nocturnal delirium; and the continuous stimulation of a single smell causes olfactory fatigue.

[0008] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0009] The embodiment of the present invention provides an odor feedback method based on artificial intelligence, which includes:

[0010] Acquire a real-time physiological data set of a target patient, wherein the data set at least includes biochemical indicators of liver and kidney function, respiratory rate time series data, and bispectral index;

[0011] Constructing a knowledge graph containing metabolic pathways of odor molecules, wherein the knowledge graph relates the patient's individualized metabolic capacity to the clearance rate of odor components;

[0012] Generate a dynamic release strategy based on the real-time physiological dataset and the knowledge graph, where the strategy includes pulsed release waveform parameters and an odor component taboo mapping table;

[0013] Execute the dynamic release strategy through the atomization control module, and the execution process includes adjusting the atomized particle size distribution according to the real-time nasal mucosa humidity feedback.

[0014] As a preferred solution of the artificial intelligence-based odor feedback method described in the present invention, wherein: the acquisition device of the real-time physiological dataset includes:

[0015] An embedded biosensor array, which is attached to the outer wall of the nasal cavity and measures the change rate of mucosal impedance;

[0016] A non-contact millimeter-wave radar, which is installed 20-30 cm directly above the hospital bed and detects millimeter-level displacements of chest cavity fluctuations.

[0017] As a preferred solution of the artificial intelligence-based odor feedback method described in the present invention, wherein: the construction method of the knowledge graph includes:

[0018] Establish an odor molecule metabolic enzyme correspondence library, which contains the decomposition rate parameter of CYP2A6 enzyme for linalool;

[0019] Mark the metabolic compensation path of patients with abnormal liver and kidney functions, and the path includes the molecular adsorption rate correction factor during renal replacement therapy;

[0020] Integrate real-time hemodynamic data and calculate the dynamic estimation value of the vascular permeability of odor components.

[0021] As a preferred solution of the artificial intelligence-based odor feedback method described in the present invention, wherein: in the knowledge graph, metabolic enzyme relationship modeling is performed to construct an enzyme-substrate kinetic model, and the linalool decomposition rate equation is defined as:

[0022]

[0023] where, v c represents the catalytic rate of CYP2A6 enzyme for linalool, with the unit of μmol / min, and the subscript c specifically refers to CYP2A6 enzyme, k cat is the catalytic constant for the decomposition of the enzyme-substrate complex into products, with the unit of s -1 , [E] is the concentration of free CYP2A6 enzyme, with the unit of nM, [S] is the concentration of linalool substrate, with the unit of mM, K m is the Michaelis constant, representing the affinity between the enzyme and the substrate, with the unit of mM, and β is the influence factor of the competitive inhibitor on the enzyme activity, and β = 0.01·e 0.5·ΔGIt is calculated that [I] is the concentration of the competitive inhibitor in plasma, in μM, and ΔG is the change in the binding free energy between the inhibitor and the enzyme active site, in kcal / mol;

[0024] Dynamically calculate the vascular permeability and establish the adsorption kinetic equation for renal replacement therapy. The formula is:

[0025]

[0026] where C eff represents the effective concentration removed by adsorption during renal replacement therapy, in mg / L, and C blood represents the real-time concentration of the molecule to be removed in the blood, in mg / L, k ad represents the binding rate constant of the adsorbent material to the target molecule, in min-1, t represents the duration of a single renal replacement therapy, in min, and η cr is a dynamic correction factor based on the creatinine clearance rate, calculated as η cr = 0.8 + 0.02·(CrCl - 30), which takes effect when CrCl ≥ 30 mL / min;

[0027] Dynamically calculate the vascular permeability. The calculation formula is:

[0028]

[0029] where C tis represents the concentration of the odor molecule in the target tissue, in μg / g, and P v represents the dynamic vascular permeability coefficient, in cm / s, and is updated in real time as P v = 0.1·CO·(1 + 0.05·ΔRR). A represents the effective capillary exchange area, in cm 2 , calculated based on the ultrasonic blood flow imaging data, ΔC is the concentration gradient between the blood and the tissue, in μg / mL / cm, and τ met represents the local tissue metabolism time constant, in min, measured by microdialysis technology, CO represents the real-time cardiac output, in L / min, and ΔRR represents the deviation of the current respiratory rate from the baseline value, in breaths / min.

[0030] As a preferred embodiment of the odor feedback method based on artificial intelligence according to the present invention, wherein: in the dynamic release strategy, an olfactory sensitivity attenuation model is established, and the association model between the drug concentration and the olfactory threshold is established as:

[0031]

[0032] Among them, S(t) represents the olfactory sensitivity at time t, with the unit of AU, calibrated by the electrophysiological signal of the olfactory bulb, S0 is the baseline sensitivity, and k decay is the neural adaptation attenuation coefficient, with the unit of min -1 , which is corrected to k through the bispectral index of electroencephalogram decay = 6·BIS + 3, and C drug represents the concentration of sedative drugs in plasma, with the unit of μg / mL. The pharmacokinetic curve is read from the electronic medical record system. λ represents the olfactory receptor recovery rate constant, with the unit of min-1, which is negatively correlated with age. The update formula is: λ = 0.02·(1 - 0.005·Age);

[0033] To predict the risk of respiratory depression, an LSTM-ATT prediction model is constructed, which is expressed as:

[0034] h t = LSTM(x t , h t-1 ; W h , b h ),

[0035]

[0036] Among them, h t represents the hidden state at time t, with a dimension of 128. W h represents the weight matrix (256×128), and b h is the bias term. ΔRR t represents the time series change of respiratory rate, with the unit of times / min 2 , e t is the attention matching degree, v is the attention weight vector, 64×1, W a is the attention matrix 192×64, α t is the time attention weight, and R risk is the risk probability of respiratory depression in the next 10 - 15 minutes, ranging from 0 to 1, and σ is the Sigmoid function.

[0037] As a preferred solution of the odor feedback method based on artificial intelligence described in the present invention, among them: the specific steps for generating the dynamic release strategy include:

[0038] Extract the historical sedative drug usage records from the electronic medical record system and establish a drug half-life olfactory sensitivity attenuation model;

[0039] Based on the LSTM network, extract the features of the respiratory rate time series data and predict the risk level of respiratory depression within the next 10 - 15 minutes;

[0040] When the risk level of respiratory depression is detected to exceed the threshold, the odor component replacement mechanism is automatically triggered, and the release formula is switched to an essential oil combination without monoterpenoid compounds;

[0041] The triggering conditions of the odor component replacement mechanism further include:

[0042] When the real-time blood oxygen saturation decrease rate ΔSpO2 / Δt ≤ -0.5% / min, the current release pulse is immediately interrupted;

[0043] At the beginning of the expiratory phase of the first respiratory cycle after interruption, low-flow pure oxygen carrier gas flushing is started, and the flushing time t = 3*RR -1 , where RR is the respiratory rate, in breaths per minute.

[0044] As a preferred solution of the odor feedback method based on artificial intelligence according to the present invention, wherein: the pulse release waveform parameters include:

[0045] The pulse width interval is set to [T min , T max , where T min ≥10 s and T max ≤60 s;

[0046] The interval time dynamic adjustment coefficient α = 1 + 0.2 * current BIS value - target BIS value, where α ∈ [0.8, 1.5];

[0047] The concentration peak constraint condition is C peak ≤0.03*CrCl + 0.5, where CrCl is the creatinine clearance rate, in mL / min.

[0048] As a preferred solution of the odor feedback method based on artificial intelligence according to the present invention, wherein: in the dynamic release strategy, pulse parameter dynamic optimization is performed, and the pulse parameter generation equation under the constraint condition is established as:

[0049]

[0050] C peak = min(0.03·CrCl + 0.5, 0.8·S(t)·e -0.1·(BIS-40) ),

[0051] where, T width represents the pulse width, C peak represents the concentration peak, T base is the reference pulse width, preset to 30 s, clip is a truncation function that constrains the value within the interval [T min , T max , and S(t) is the current olfactory sensitivity;

[0052] Perform real-time update of the tabu mapping and construct the molecular interaction constraint conditions as follows:

[0053]

[0054] Among them, is the set of tabu molecular pairs,

[0055] is the odor molecule m i and m j The second-order derivative of the concentration interaction in the tissue, and ξ is the threshold of liver and kidney metabolism overload, with the unit of (μg / g) 2 , calculated according to the knowledge graph as ξ = 2.5·(1 - e 0.1·BIS ).

[0056] As a preferred solution of the odor feedback method based on artificial intelligence described in the present invention, wherein: the execution process of the atomization control module further includes:

[0057] Start a reverse air flow during the adjacent pulse interval, and the flow rate V of the reverse air flow is V = 0.5Q + 2, where Q is the total amount of odor molecule release in the previous pulse stage, with the unit of μL;

[0058] According to the environmental temperature and humidity sensor data, dynamically adjust the oscillation frequency f of the atomization sheet to f = 200 + 10(T - 25) + 5*(H - 60), where T is the temperature, with the unit of °C, and H is the humidity, with the unit of %RH.

[0059] As a preferred solution of the odor feedback method based on artificial intelligence described in the present invention, wherein: the method further includes establishing an olfactory-visual cross-modal compensation mechanism:

[0060] Capture the coordinates of the patient's fixation area through an eye tracking device, and activate the release of menthol components when the fixation points fall on the alarm device area for 5 consecutive times or more;

[0061] When the eye movement frequency > 8Hz is detected during the rapid eye movement sleep stage, switch the release base carrier to a tetradecane microcapsule suspension with phase change cooling characteristics.

[0062] The beneficial effects of the present invention are as follows: By using the CYP2A6 enzyme kinetic model and the adsorption equation for renal replacement therapy, the present invention dynamically calculates the clearance rate of odor components in patients with liver and kidney dysfunction. Combining with the threshold constraints of liver and kidney metabolism overload in the taboo mapping table, the toxicity risk is reduced, and the problem of uncontrollable metabolism in traditional methods is solved. Based on the olfactory sensitivity decay model, a pulse release waveform is designed, and coupled with reverse airflow flushing, the receptor recovery period is shortened, and the EEG response latency is shortened, breaking through the desensitization effect caused by continuous release. Through the atomization frequency temperature and humidity compensation formula and the phase change microcapsule carrier, the atomization particle size fluctuation is compressed from ±2.1 μm to ±0.3 μm under extreme conditions, and the deposition efficiency is increased from 52% to 89%, overcoming the influence of environmental parameter drift. Through multi-modal data fusion and dynamic constraint models, the millisecond-level spatio-temporal synchronization of odor intervention and the real-time mapping of the metabolic safety boundary are achieved, systematically solving the static control defects in the background technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0064] Figure 1 It is a schematic flow chart of an odor feedback method based on artificial intelligence in Embodiment 1. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0065] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following will describe the specific embodiments of the present invention in detail with reference to the drawings of the specification.

[0066] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0067] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation of the present invention. The "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selectively exclusive embodiment from other embodiments.

[0068] Embodiment 1, refer to Figure 1 This embodiment provides an odor feedback method based on artificial intelligence, including the following steps:

[0069] Step S1, obtain the real-time physiological data set of the target patient, where the data set at least includes liver and kidney function biochemical indicators, respiratory rate time series data, and bispectral index of electroencephalogram;

[0070] The acquisition device for the real-time physiological data set includes:

[0071] An embedded biosensor array, which is attached to the outer wall of the nasal cavity and measures the change rate of mucosal impedance;

[0072] A non-contact millimeter-wave radar, which is installed 20 - 30 cm directly above the hospital bed and detects the millimeter-level displacement of chest cavity undulation;

[0073] Step S2, construct a knowledge graph containing the metabolic pathways of odor molecules, where the knowledge graph associates the individual metabolic capacity of the patient with the odor component clearance rate;

[0074] The construction method of the knowledge graph includes:

[0075] Establish a correspondence library of odor molecule metabolic enzymes, where the library contains the decomposition rate parameter of CYP2A6 enzyme for linalool;

[0076] Annotate the metabolic compensation pathways of patients with abnormal liver and kidney functions, where the pathways include the molecular adsorption rate correction factor during renal replacement therapy;

[0077] Integrate real-time hemodynamic data and calculate the dynamic estimation value of the vascular permeability rate of odor components;

[0078] In the knowledge graph, perform metabolic enzyme relationship modeling, construct an enzyme-substrate kinetic model, and define the linalool decomposition rate equation as:

[0079]

[0080] where, v c represents the catalytic rate of CYP2A6 enzyme for linalool, with the unit of μmol / min, the subscript c specifically refers to CYP2A6 enzyme, k cat is the catalytic constant for the decomposition of the enzyme-substrate complex into products, with the unit of s -1 , [E] is the concentration of free CYP2A6 enzyme, with the unit of nM, [S] is the concentration of linalool substrate, with the unit of mM, K m is the Michaelis constant, representing the affinity between the enzyme and the substrate, with the unit of mM, β is the influence factor of the competitive inhibitor on the enzyme activity, calculated by β = 0.01·e 0.5·ΔG , [I] is the concentration of the competitive inhibitor in plasma, with the unit of μM, and ΔG is the change in the binding free energy between the inhibitor and the enzyme active site, with the unit of kcal / mol;

[0081] Dynamically calculate the vascular permeability and establish the adsorption kinetic equation for renal replacement therapy. The formula is as follows:

[0082]

[0083] Among them, C eff represents the effective concentration of adsorbed clearance during renal replacement therapy, with the unit of mg / L. C blood represents the real-time concentration of the molecule to be cleared in the blood, with the unit of mg / L. k ad represents the binding rate constant of the adsorbent material to the target molecule, with the unit of min -1 , t represents the duration of a single renal replacement therapy, with the unit of min. η cr is the dynamic correction factor based on the creatinine clearance rate, calculated as η cr = 0.8 + 0.02·(CrCl - 30), which takes effect when CrCl ≥ 30 mL / min;

[0084] Dynamically calculate the vascular permeability. The calculation formula is as follows:

[0085]

[0086] Among them, C tis represents the concentration of odor molecules in the target tissue, with the unit of μg / g. P v represents the dynamic vascular permeability coefficient, with the unit of cm / s, and is updated in real time as P v = 0.1·CO·(1 + 0.05·ΔRR). A represents the effective capillary exchange area, with the unit of cm 2 , calculated based on the ultrasonic blood flow imaging data. Δc is the concentration gradient between the blood and the tissue, with the unit of μg / mL / cm. τ met represents the local tissue metabolism time constant, with the unit of min, measured by microdialysis technology. CO represents the real-time cardiac output, with the unit of L / min. ΔRR represents the deviation of the current respiratory rate from the baseline value, with the unit of times / min;

[0087] In the above-mentioned dynamic release strategy, perform olfactory sensitivity attenuation modeling and establish the association model between drug concentration and olfactory threshold as follows:

[0088]

[0089] Among them, S(t) represents the olfactory sensitivity at time t, with the unit of AU, calibrated by the electrophysiological signal of the olfactory bulb. S0 is the baseline sensitivity. k decay is the neural adaptation attenuation coefficient, with the unit of min -1 , corrected by the bispectral index of electroencephalogram to k decay = 6·BIS + 3. C drugIt represents the concentration of sedative drugs in plasma, with the unit of μg / mL. The pharmacokinetic curve is read from the electronic medical record system. λ represents the olfactory receptor recovery rate constant, with the unit of min -1 , which is negatively correlated with age. The updated formula is: λ = 0.02·(1 - 0.005·Age);

[0090] Perform respiratory depression risk prediction and construct an LSTM-ATT prediction model, expressed as:

[0091] h t = LSTM(x t ,h t-1 ; W h ,b n ),

[0092]

[0093] where h t represents the hidden state at time t, with a dimension of 128. W h represents the weight matrix (256×128), b h is the bias term, ΔRR t represents the time series change in respiratory rate, with the unit of breaths / min 2 , e t is the attention matching degree, v is the attention weight vector, 64×1, W a is the attention matrix 192×64, α t is the time attention weight, R risk is the respiratory depression risk probability for the next 10 - 15 minutes, ranging from 0 to 1, and σ is the Sigmoid function;

[0094] Step S3: Generate a dynamic release strategy based on the real-time physiological data set and the knowledge graph. The strategy includes pulse release waveform parameters and an odor component taboo mapping table;

[0095] The specific steps for generating the dynamic release strategy include:

[0096] Extract historical sedative drug usage records from the electronic medical record system and establish a drug half-life olfactory sensitivity decay model;

[0097] Extract features from the respiratory rate time series data based on the LSTM network and predict the respiratory depression risk level for the next 10 - 15 minutes;

[0098] When it is detected that the respiratory depression risk level exceeds the threshold, automatically trigger the odor component replacement mechanism and switch the release formula to an essential oil combination without monoterpene compounds;

[0099] The triggering conditions of the odor component replacement mechanism further include:

[0100] When the real-time blood oxygen saturation decline rate ΔSpO2 / Δt ≤ -0.5% / min, immediately interrupt the current released pulse;

[0101] At the beginning of the expiratory phase of the first respiratory cycle after interruption, start low-flow pure oxygen carrier gas flushing, and the flushing time t = 3*RR -1 , where RR is the respiratory rate, in breaths per minute;

[0102] The pulse-type release waveform parameters include:

[0103] The pulse width interval is set to [T min , T max , where T min ≥ 10 seconds and T max ≤ 60 seconds;

[0104] The dynamic adjustment coefficient of the interval time α = 1 + 0.2 * current BIS value - target BIS value, where α ∈ [0.8, 1.5];

[0105] The concentration peak constraint condition is C peak ≤ 0.03 * CrCl + 0.5, where CrCl is the creatinine clearance rate, in mL / min;

[0106] In the dynamic release strategy, perform dynamic optimization of the pulse parameters, and establish the pulse parameter generation equation under the constraint conditions as:

[0107]

[0108] C peak = min(0.03·CrCl + 0.5, 0.8·S(t)·e -0.1·(BIS-40) ),

[0109] where, T width represents the pulse width, C peak represents the concentration peak, T base is the reference pulse width, preset to 30 seconds, clip is the truncation function that constrains the value within the interval [T min , T max , and S(t) is the current olfactory sensitivity;

[0110] Perform real-time update of the tabu mapping, and construct the molecular interaction constraint condition as:

[0111]

[0112] where, is the set of tabu molecule pairs,

[0113] is the odor molecule mi with m j The second-order derivative of the concentration interaction in the tissue, where ξ is the threshold value of liver and kidney metabolic overload, with the unit of (μg / g) 2 , calculated according to the knowledge graph as ξ = 2.5·(1 - e 0.1·BIS );

[0114] Specifically, this strategy achieves precise intervention through four-level dynamic constraints; the attenuation model couples pharmacokinetic integration with neural adaptive attenuation to solve the lag problem of traditional static threshold models;

[0115] The LSTM-ATT model uses the change in respiratory frequency as the context feature of the attention mechanism, increasing the risk prediction accuracy by 23%; the pulse parameter generation introduces dual constraint conditions, incorporating the antagonistic effect of CrCl and BIS into the calculation of the concentration peak to avoid liver and kidney metabolic overload; the taboo mapping captures the molecular synergy effect through the second-order derivative, and its dynamic threshold mechanism can predict metabolic conflicts 15 minutes in advance;

[0116] Step S4, execute the dynamic release strategy through the atomization control module, and the execution process includes adjusting the atomization particle size distribution according to the real-time nasal mucosa humidity feedback;

[0117] The execution process of the atomization control module further includes:

[0118] Start a reverse air flow during the adjacent pulse interval, where the reverse air flow velocity V = 0.5Q + 2, and Q is the total amount of odor molecule release in the previous pulse stage, with the unit of μL;

[0119] Dynamically adjust the oscillation frequency of the atomization sheet f = 200 + 10(T - 25) + 5*(H - 60) according to the environmental temperature and humidity sensor data, where T is the temperature, with the unit of °C, and H is the humidity, with the unit of %RH;

[0120] The method further includes establishing an olfactory-visual cross-modal compensation mechanism:

[0121] Capture the coordinates of the patient's fixation area through an eye movement tracking device, and activate the release of menthol components when the fixation points fall on the alarm device area for 5 consecutive times or more;

[0122] When the eye movement frequency > 8Hz is detected during rapid eye movement sleep, switch the release base carrier to a tetradecane microcapsule suspension with phase change cooling characteristics.

[0123] Example 2, this example provides a temperature and humidity compensation mechanism for the oscillation frequency of the atomization sheet in an artificial intelligence-based odor feedback method. Aiming at the problem that the atomization particle size distribution drifts due to environmental temperature and humidity fluctuations and the nasal deposition efficiency drops by more than 40%, it includes:

[0124] Configured Hardware:

[0125] Piezoelectric atomization sheet: The material is PZT-5H (piezoelectric constant d 33 = 650×10 -12 C / N, frequency temperature coefficient β T = -0.02% / °C);

[0126] Environmental sensor: Adopts SHT35 type temperature and humidity sensor (detection accuracy ±0.2°C / ±2%RH).

[0127] Compensation logic: Calculate the oscillation frequency of the atomization sheet, f = 200 + 10×(T - 25) + 5×(H - 60),

[0128] Temperature compensation term 10×(T - 25):

[0129] Experimentally measured that for the piezoelectric sheet at a reference temperature of 25°C, when the temperature increases by 1°C, the resonance frequency shifts by 0.3% (verified by an impedance analyzer).

[0130] Humidity compensation term 5×(H - 60):

[0131] According to ASTM E104-02 standard, when the humidity increases by 10%RH, the surface tension of the solution decreases by 1.2 mN / m, resulting in an increase in the mass median diameter (MMAD) by 1.8 μm.

[0132] Implementation steps:

[0133] Calibration stage: Under the conditions of 25°C / 60%RH, adjust the driving voltage to make the resonance frequency of the atomization sheet 200 Hz;

[0134] Real-time adjustment: When the ambient temperature rises to 28°C and the humidity is 75%, calculate the target frequency:

[0135] f = 200 + 10×(28 - 25) + 5×(75 - 60) = 305 Hz;

[0136] Adjust the amplitude of the driving voltage through a PID controller to make the actual frequency error ≤ ±2 Hz.

[0137] Conduct experimental verification, as shown in Table 1;

[0138] Table 1:

[0139]

[0140] Example 3, this example provides a temperature-sensitive microcapsule carrier system in an odor feedback method based on artificial intelligence. Aiming at the problem that the nasal mucosa temperature rises by 0.5 - 1.2°C during the rapid eye movement sleep period (REM), resulting in the imbalance of the volatility of conventional essential oils, it includes:

[0141] Microcapsule preparation parameters:

[0142] Core material composition: molar ratio of tetradecane to hexadecane is 7:3 (eutectic point 34 ± 0.5 °C);

[0143] Wall material structure: gelatin - gum arabic composite film (crosslinking degree 85%, thickness 2.8 ± 0.2 μm);

[0144] Particle size distribution: D10 = 10 μm, D50 = 28 μm, D90 = 50 μm (measured by laser particle size analyzer).

[0145] Carrier switching control:

[0146] Trigger condition: eye movement frequency > 8 Hz for 10 seconds (detected by Tobii Pro Nano eye tracker);

[0147] Comparison of carrier parameters, see Table 2;

[0148] Table 2:

[0149]

[0150] Verification data of phase change characteristics:

[0151] Differential scanning calorimetry (DSC):

[0152] Melting onset temperature: 33.7 °C;

[0153] Melting enthalpy value: 218 J / g (testing instrument: TA Q2000, heating rate 5 °C / min).

[0154] Thermal response test: when the nasal mucosa temperature rises from 33.5 °C to 34.2 °C, the microcapsule release rate increases by 180%.

[0155] Example 4, this example provides an exhalation - phase precise flushing mechanism in an odor feedback method based on artificial intelligence. Aiming at the problem that the traditional flushing strategy is out of phase with the respiratory cycle phase, resulting in an increase in alveolar dead space, it includes:

[0156] Time calculation model: flushing time, t = 3 × RR -1 ,

[0157] Physiological basis: adult anatomic dead space volume: 150 mL;

[0158] Flushing air flow rate: 500 mL / s (according to ISO 80601 - 2 - 55 standard);

[0159] Theoretical flushing time: 150 mL / 500 mL / s = 0.3 s, covering 1 exhalation cycle (when RR = 20 breaths / min, the cycle is 3 s).

[0160] Implementation steps:

[0161] Dynamic monitoring: Detect the starting point of the expiratory phase through a respiratory flow sensor (triggered when the peak expiratory flow rate drops to -200 mL / s);

[0162] Interruption condition: When the rate of decrease in blood oxygen saturation ΔSpO2 / Δt ≤ -0.5% / min, immediately stop the odor release;

[0163] Flushing execution: Start a 500 mL / s pure oxygen flow during the expiratory phase, with a duration of t = 3 / RRmin (when RR = 18 breaths / min, t = 10 seconds).

[0164] Lung function compensation and correction:

[0165] For patients with chronic obstructive pulmonary disease (COPD), add a correction factor t adj :

[0166] t adj = t × (1 + 0.05 × FEV1%), where FEV1% is the percentage of the forced expiratory volume in the first second to the predicted value.

[0167] Clinical effect comparison, see Table 3;

[0168] Table 3:

[0169]

[0170] All experimental data of Examples 2 - 4 are based on tests in an ISO / IEC 17025 - certified laboratory and comply with the clinical verification specifications for medical devices.

[0171] In summary, it can be seen that the present invention dynamically calculates the odor component clearance rate of patients with liver and kidney dysfunction through the CYP2A6 enzyme kinetic model and the renal replacement therapy adsorption equation, and combines the liver and kidney metabolism overload threshold constraints of the taboo mapping table to reduce the toxicity risk and solve the problem of uncontrollable metabolism in traditional methods; designs a pulse release waveform based on the olfactory sensitivity attenuation model, couples reverse airflow flushing, shortens the receptor recovery period and the EEG response latency, and breaks through the desensitization effect caused by continuous release; compresses the atomization particle size fluctuation from ±2.1 μm to ±0.3 μm and increases the deposition efficiency from 52% to 89% under extreme conditions through the atomization frequency temperature and humidity compensation formula and the phase - change microcapsule carrier, overcoming the influence of environmental parameter drift; realizes millisecond - level spatio - temporal synchronization of odor intervention and real - time mapping of the metabolic safety boundary through multi - modal data fusion and dynamic constraint models, and systematically solves the static control defects in the background technology.

[0172] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. An artificial intelligence-based odor feedback method, characterized in that: Including, Obtain the real-time physiological data set of the target patient, where the data set at least includes liver and kidney function biochemical indicators, respiratory rate time-series data, and bispectral index; Construct a knowledge graph including the metabolic pathway of odor molecules, which associates the individual metabolic capacity of the patient with the odor component clearance rate; Generate a dynamic release strategy based on the real-time physiological data set and the knowledge graph, where the strategy includes pulse release waveform parameters and an odor component taboo mapping table; Execute the dynamic release strategy through the atomization control module, and the execution process includes adjusting the atomization particle size distribution according to the real-time nasal mucosa humidity feedback.

2. The odor feedback method based on artificial intelligence according to claim 1, wherein The acquisition device of the real-time physiological data set includes: An embedded biosensor array, which is attached to the outer wall of the nasal cavity and measures the change rate of mucosal impedance; A non-contact millimeter-wave radar, which is installed 20-30 cm directly above the hospital bed and detects millimeter-level displacement of chest fluctuations.

3. The odor feedback method based on artificial intelligence according to claim 1, characterized in that The construction method of the knowledge graph includes: Establish a correspondence library of odor molecule metabolic enzymes, where the library includes the decomposition rate parameters of CYP2A6 enzyme for linalool; Label the metabolic compensation path of patients with abnormal liver and kidney function, where the path includes the molecular adsorption rate correction factor during renal replacement therapy; Integrate real-time hemodynamic data and calculate the dynamic estimation of the vascular permeability of odor components.

4. The odor feedback method based on artificial intelligence according to claim 1, characterized in that In the knowledge graph, perform metabolic enzyme relationship modeling, construct an enzyme-substrate kinetic model, and define the linalool decomposition rate equation as: Among them, v c represents the catalytic rate of CYP2A6 enzyme on linalool, with the unit of μmol / min. The subscript c specifically refers to the CYP2A6 enzyme, and k cat is the catalytic constant for the decomposition of the enzyme-substrate complex into products, with the unit of s -1 , [E] is the concentration of free CYP2A6 enzyme, with the unit of nM, [S] is the concentration of linalool substrate, with the unit of mM, and K m is the Michaelis constant, representing the affinity between the enzyme and the substrate, with the unit of mM. β is the influence factor of the competitive inhibitor on the enzyme activity, which is calculated by β = 0.01·e 0.5·ΔG and [I] is the concentration of the competitive inhibitor in plasma, with the unit of μM, and ΔG is the change in the binding free energy between the inhibitor and the enzyme active site, with the unit of kcal / mol; Dynamically calculate the vascular permeability, establish the adsorption kinetic equation of renal replacement therapy, and the formula is: Among them, C eff represents the effective concentration removed by adsorption during renal replacement therapy, with the unit of mg / L, C blood represents the real-time concentration of the molecule to be removed in the blood, with the unit of mg / L, k ad represents the binding rate constant of the adsorbent material to the target molecule, with the unit of min-1, t represents the duration of a single renal replacement therapy, with the unit of min, η cr is a dynamic correction factor based on creatinine clearance rate, calculated as η cr = 0.8 + 0.02·(CrCl - 30), which takes effect when CrCl ≥ 30 mL / min; Dynamically calculate the vascular permeability, and the calculation formula is: Among them, C tis represents the concentration of odor molecules in the target tissue, with the unit of μg / g, and P v represents the dynamic vascular permeability coefficient, with the unit of cm / s, and is updated in real time according to P v = 0.1·CO·(1 + 0.05·ΔRR). A represents the effective capillary exchange area, with the unit of cm 2 , which is calculated based on ultrasonic blood flow imaging data. ΔC is the concentration gradient between blood and tissue, with the unit of μg / mL / cm, and τ met represents the local tissue metabolic time constant, with the unit of min, which is measured by microdialysis technology. CO represents the real-time cardiac output, with the unit of L / min, and ΔRR represents the deviation of the current respiratory rate from the baseline value, with the unit of times / min.

5. The odor feedback method based on artificial intelligence according to claim 1, characterized in that, In the dynamic release strategy, perform olfactory sensitivity attenuation modeling, and establish a correlation model between drug concentration and olfactory threshold as: Among them, S(t) represents the olfactory sensitivity at time t, with the unit of AU, calibrated by the electrophysiological signal of the olfactory bulb. S0 is the baseline sensitivity, and k decay is the neural adaptation attenuation coefficient, with the unit of min -1 , and is corrected to k through the bispectral index of electroencephalogram decay = 6·BIS + 3, where C drug represents the concentration of sedative drugs in plasma, with the unit of μg / mL. The pharmacokinetic curve is read from the electronic medical record system. λ represents the olfactory receptor recovery rate constant, with the unit of min -1 , which is negatively correlated with age. The update formula is: λ = 0.02·(1 - 0.005·Age); Perform respiratory depression risk prediction, and construct an LSTM-ATT prediction model, which is expressed as: h t = LSTM(x t , h t-1 ; W h , b h ), Among them, h t represents the hidden state at time t, with a dimension of 128, W h represents the weight matrix (256×128), b h is the bias term, ΔRR t represents the time series change in respiratory rate, with the unit of breaths / min 2 , e t is the attention matching degree, v is the attention weight vector, 64×1, W a is the attention matrix 192×64, α t is the time attention weight, R risk is the risk probability of respiratory depression in the next 10 - 15 minutes, ranging from 0 to 1, and σ is the Sigmoid function.

6. The odor feedback method based on artificial intelligence according to claim 1, characterized in that The specific steps for generating the dynamic release strategy include: Extract the historical sedative drug use records from the electronic medical record system and establish a drug half-life olfactory sensitivity attenuation model; Based on the LSTM network, extract features from the respiratory rate time-series data and predict the respiratory depression risk level within the next 10-15 minutes; When it is detected that the respiratory depression risk level exceeds the threshold, automatically trigger the odor component replacement mechanism and switch the release formula to an essential oil combination without monoterpene compounds; The triggering conditions of the odor component replacement mechanism further include: When the real-time blood oxygen saturation decline rate ΔSpO2 / Δt ≤ -0.5% / min, immediately interrupt the current release pulse; At the beginning of the expiratory phase of the first respiratory cycle after interruption, low-flow pure oxygen carrier gas flushing is initiated, and the flushing time t = 3 * RR -1 , where RR is the respiratory rate, in breaths per minute.

7. The odor feedback method based on artificial intelligence according to claim 1, characterized in that, The pulse release waveform parameters include: The pulse width interval is set to [T min , T max , where T min ≥ 10 seconds and T max ≤ 60 seconds; The interval time dynamic adjustment coefficient α = 1 + 0.2 * current BIS value - target BIS value, where α ∈ [0.8, 1.5]; The peak concentration constraint is C peak ≤ 0.03 * CrCl + 0.5, where CrCl is the creatinine clearance rate in units of mL / min.

8. The odor feedback method based on artificial intelligence according to claim 1, characterized in that, In the dynamic release strategy, perform dynamic optimization of pulse parameters, and establish a pulse parameter generation equation under constraint conditions as: C peak = min(0.03·CrCl + 0.5, 0.8·S(t)·e -0.1·(BIS-40) ) Among them, T width represents the pulse width, C peak represents the peak concentration, T base is the reference pulse width, preset to 30 seconds, clip is a truncation function that constrains the value within the interval of [T min , T max , and S(t) is the current olfactory sensitivity; Perform real-time update of taboo mapping and construct molecular interaction constraint conditions as: Among them, is a set of taboo molecule pairs, is the odor molecule m i and m j is the second-order derivative of the concentration interaction in the tissue, and ξ is the liver and kidney metabolic overload threshold, with the unit of (μg / g) 2 , calculated according to the knowledge graph as ξ = 2.5·(1 - e 0.1·BIS ).

9. The odor feedback method based on artificial intelligence according to claim 1, characterized in that The execution process of the atomization control module further includes: Start a reverse air flow during the adjacent pulse interval, and the reverse air flow velocity V = 0.5Q + 2, where Q is the total amount of odor molecule release in the previous pulse stage, in units of μL; According to the environmental temperature and humidity sensor data, dynamically adjust the oscillation frequency of the atomizing sheet \(f = 200+10(T - 25)+5\times(H - 60)\), where \(T\) is the temperature in \(^{\circ}C\) and \(H\) is the humidity in \(\%RH\).

10. The odor feedback method based on artificial intelligence according to claim 1, characterized in that, The odor feedback of the method further includes establishing an olfactory-visual cross-modal compensation mechanism: Capture the coordinates of the patient's fixation area through an eye movement tracking device. When the fixation points fall on the alarm device area for 5 consecutive times or more, activate the release of menthol components; When the eye movement frequency > 8 Hz is detected during the rapid eye movement sleep period, switch the release base carrier to a tetradecane microcapsule suspension with phase change cooling characteristics.

Images