Intelligent moxibustion treatment method and system based on multi-modal interaction

Abstract

The invention relates to the technical field of treatment safety early warning, and relates to an intelligent moxibustion treatment method and system based on multi-modal interaction, and the method comprises the steps: obtaining a PPG signal of a patient and an infrared thermogram comprising all moxibustion regions when moxibustion is carried out in a plurality of moxibustion regions of the patient; calculating a thermal load coefficient of each moxibustion area; calculating a heart rate variation coefficient for the PPG signal; calculating a tolerance coefficient according to the heart rate variation coefficient and the thermal load coefficients of all the moxibustion areas; and determining whether to give an alarm or not according to the tolerance coefficient, the heart rate variation coefficient and the thermal load coefficients of all the moxibustion areas. According to the method, the physical thermal load is comprehensively and objectively evaluated, and the scalding risk caused by subjectively judging the skin temperature and the tolerance condition of the patient is avoided.

Description

Technical Field

[0001] This invention generally relates to the field of treatment safety early warning technology. More specifically, this invention relates to an intelligent moxibustion treatment method and system based on multimodal interaction. Background Technology

[0002] Moxibustion is a traditional Chinese medicine therapy originating from ancient China. It uses moxa wool or sticks made from mugwort (Artemisia argyi) to stimulate acupoints on the body's surface through the heat generated by burning. In traditional Chinese medicine theory, this therapy is believed to have the effects of warming the meridians and dispelling cold, promoting qi and blood circulation, and supporting yang qi. It is often used to relieve various cold-related and deficiency-related symptoms (such as joint pain, dysmenorrhea due to uterine cold, and spleen and stomach deficiency) as well as for daily disease prevention and health maintenance. Due to its relatively simple operation, wide range of applications, and unique therapeutic effects, moxibustion therapy occupies an important position in modern rehabilitation therapy and disease prevention.

[0003] In the clinical application of moxibustion, whether it is traditional moxibustion with moxa sticks or moxa cones, or modern moxibustion devices, the therapeutic effect and safety highly depend on the precise control of the "moxibustion dosage." The "moxibustion dosage" includes not only the intensity of the heat and the duration of the treatment, but also the distance between the moxibustion site and the heat source. Insufficient dosage results in insufficient heat, making it difficult to achieve the desired therapeutic effect; while excessive dosage can easily lead to skin burns or even tissue damage, causing adverse consequences.

[0004] In current moxibustion treatments, safety control largely relies on the experience of medical staff. They typically determine the appropriate heat level by observing the patient's condition (such as skin reaction) or inquiring about their subjective feelings, and then control the distance and duration of moxibustion accordingly. This method is highly subjective and poses safety risks. For example, when a patient's sensory perception is impaired (e.g., elderly people), they are asleep, or they fail to report discomfort for other reasons, medical staff may be unable to accurately judge the actual skin temperature and the patient's true tolerance, thus increasing the risk of burns. Summary of the Invention

[0005] To address the technical problem that medical personnel may be unable to accurately determine the actual skin temperature and the patient's true tolerance, thereby increasing the risk of burns, the present invention provides solutions in the following aspects.

[0006] In a first aspect, an intelligent moxibustion therapy method based on multimodal interaction includes: acquiring the patient's PPG signal and an infrared thermogram including all moxibustion areas while performing moxibustion on multiple moxibustion areas of the patient; calculating the heat load coefficient of each moxibustion area, wherein, at any given time, the first... i The heat load coefficient of the moxibustion area is related to the first iThe grayscale value of all pixels in the moxibustion area is directly proportional to the maximum grayscale value. i Represents the index; calculate the heart rate variability coefficient for the PPG signal, where the first... j Heart rate variability at time t and the first j Time to the j - n It is related to the difference between all adjacent peaks in the PPG signal at time t. j Represents an index. n It is a positive integer of a preset size. j - n >0; Calculate the tolerance coefficient based on the heart rate variability coefficient and the heat load coefficient of all moxibustion areas, wherein at any given time, the tolerance coefficient is inversely proportional to the heart rate variability coefficient and directly proportional to the heat load coefficient of all moxibustion areas; Determine whether to trigger an alarm based on the tolerance coefficient, the heart rate variability coefficient, and the heat load coefficient of all moxibustion areas.

[0007] Preferably, calculate the first time step at the current moment. i The formula for the heat load coefficient of a moxibustion area is: ; At the current moment, I i For the first i The heat load coefficient of each moxibustion area L max For the first i The maximum grayscale value of all pixels in the moxibustion area. L s,i For the first i The first moxibustion area s grayscale value of each pixel. N i For the first i The number of all pixels in each moxibustion area L T The threshold for acute burns is set at a preset size. L safe The preset size is the safety threshold for low-temperature burns, and L T > L safe , α The first weighting coefficient is of a preset size. β The second weighting coefficient is of a preset size, and α + β =1, s Represents an index.

[0008] Preferably, the first weighting coefficient α The value is 0.3, the second weighting coefficientβ The value is 0.7.

[0009] Preferably, calculate the first j The formula for the coefficient of variation of heart rate at any given time is: ; Among them, in the first j Time to the j - n In the PPG signal at time Δ P m,j For the first m The first peak and the second m +1 time difference between wave peaks, Δ P m+1,j For the first m +1 peak and the first m + the time difference between the two peaks k For the number of all peaks, m Represents an index.

[0010] Preferably, the formula for calculating the tolerance coefficient at the current moment is: ; in, V The tolerance coefficient at the current moment, h The coefficient of variation of heart rate at the current moment. I i For the current moment, the first i The heat load coefficient of each moxibustion area k The scaling factor is the preset size. ω i For the first i The weighting coefficients corresponding to each moxibustion area norm () is the standard normalization function, and exp() is the exponential function with the natural constant e as the base.

[0011] Preferably, an intelligent moxibustion treatment method based on multimodal interaction further includes calculating the first... i The weighting coefficient corresponding to each moxibustion area ω i The calculation formula is as follows: ,in N i For the first i The number of all pixels in the moxibustion area.

[0012] Preferably, the proportionality coefficient k The value is 2.

[0013] Preferably, an intelligent moxibustion treatment method based on multimodal interaction further includes: calculating the first... jBefore the heart rate variability coefficient at time 1, for the 1st j Time to the j - n The PPG signal is bandpass filtered at any given time.

[0014] Preferably, determining whether to trigger an alarm based on the tolerance coefficient, the heart rate variability coefficient, and the heat load coefficient of all moxibustion areas includes: triggering an alarm in response to the tolerance coefficient being greater than a preset tolerance threshold; triggering an alarm in response to the heart rate variability coefficient being less than a preset heart rate variability threshold; and triggering an alarm in response to the heat load coefficient of at least one moxibustion area being greater than a preset heat load threshold.

[0015] In a second aspect, an intelligent moxibustion therapy system based on multimodal interaction includes a processor and a memory, the memory storing a computer program, and the processor executing the computer program to implement an intelligent moxibustion therapy method based on multimodal interaction as described in any of the above-described inventions.

[0016] The beneficial effects of this invention are as follows: This invention calculates the heat load coefficient based on infrared thermography. This coefficient objectively combines the maximum temperature of each area of ​​the patient receiving moxibustion treatment (to assess acute burns) and the temperature of the portion exceeding the safe threshold (to assess low-temperature burns). This provides a comprehensive and objective assessment of the physical heat load, avoiding the risk of burns caused by subjective judgment of the patient's skin temperature and tolerance.

[0017] This invention utilizes PPG signals to calculate the coefficient of heart rate variability (CFR). The CFR is used to assess a patient's physiological stress response to pain or discomfort, thus solving the problem that subjective judgment cannot reveal the patient's true physiological state. Attached Figure Description

[0018] The above and other objects, features, and advantages of exemplary embodiments of the present invention will become readily apparent upon reading the following detailed description with reference to the accompanying drawings. In the drawings, several embodiments of the invention are illustrated by way of example and not limitation, and like or corresponding reference numerals denote like or corresponding parts, wherein: Figure 1 This is a schematic flowchart illustrating the steps of an intelligent moxibustion treatment method based on multimodal interaction according to an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the structure of an intelligent moxibustion treatment system based on multimodal interaction according to this embodiment. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0021] Figure 1 This is a schematic flowchart illustrating the steps of an intelligent moxibustion treatment method based on multimodal interaction according to an embodiment of the present invention.

[0022] like Figure 1 As shown, an intelligent moxibustion treatment method based on multimodal interaction includes steps S1 to S5.

[0023] Step S1: When performing moxibustion on multiple moxibustion areas of the patient, obtain the patient's PPG signal and infrared thermograms including all moxibustion areas.

[0024] It's important to note that infrared thermal images are generated by detecting the thermal radiation of an object using infrared light, and are used to display the temperature distribution on the object's surface. Infrared thermal images typically use different brightness levels to represent different temperatures. Generally, brighter pixels indicate higher temperatures, and darker pixels indicate lower temperatures.

[0025] In one embodiment, medical staff mark the boundaries of each moxibustion area for the patient. After obtaining an infrared thermogram including all moxibustion areas, an image edge detection algorithm is used to obtain the boundaries of each moxibustion area, and then the pixels within the boundaries of each moxibustion area are marked as the pixels of the corresponding moxibustion area.

[0026] Step S2: Calculate the heat load coefficient of each moxibustion area.

[0027] Wherein, at any given moment, the first i The heat load coefficient of the moxibustion area is related to the first i The grayscale value of all pixels in the moxibustion area is directly proportional to the maximum grayscale value. i Represents an index.

[0028] In one embodiment, the calculation of the current time's... i The formula for the heat load coefficient of a moxibustion area is: ; At the current moment, I i For the first i The heat load coefficient of each moxibustion area Lmax For the first i The maximum grayscale value of all pixels in the moxibustion area. L s,i For the first i The first moxibustion area s grayscale value of each pixel. N i For the first i The number of all pixels in each moxibustion area L T The threshold for acute burns is set at a preset size. L safe The preset size is the safety threshold for low-temperature burns, and L T > L safe , α The first weighting coefficient is of a preset size. β The second weighting coefficient is of a preset size, and α + β =1, s Represents the index. The first weight coefficient. α The value is 0.3, the second weighting coefficient β The value is 0.7.

[0029] In one embodiment, if the grayscale value of a pixel in the infrared thermogram is equal to the acute burn threshold... L T If the grayscale value of a pixel in the infrared thermogram is equal to the low-temperature burn safety threshold, then the patient's skin temperature corresponding to that pixel is 50°C. L safe If the temperature of the patient's skin corresponding to that pixel is 44℃, then the temperature of the skin at that pixel is 44℃.

[0030] It should be noted that the maximum grayscale value of all pixels in the moxibustion area exceeds the critical threshold for acute burns, i.e., the formula... The higher the value, the greater the local high temperature caused by excessive moxibustion heat or too close proximity of the moxa stick to the skin, potentially posing a risk of acute burns to the patient. The higher the temperature of the moxibustion area and the larger the area with the high temperature, the greater the potential risk. (Formula) The higher the value, the greater the risk of low-temperature burns if the patient's large area exceeds the preset low-temperature burn temperature for an extended period.

[0031] Based on this, the first i The heat load coefficient of the moxibustion area at the current moment I i The larger the patient's... i The higher the local or overall temperature of the moxibustion area at the current moment, the greater the probability of the patient suffering burns.

[0032] Step S3: Calculate the heart rate variability coefficient for the PPG signal.

[0033] Among them, the first j Heart rate variability at time t and the first j Time to the j - n It is related to the difference between all adjacent peaks in the PPG signal at time t. j Represents an index. n It is a positive integer of a preset size. j - n >0.

[0034] In one embodiment, the calculation of the first j The formula for the coefficient of variation of heart rate at any given time is: ; Among them, in the first j Time to the j - n In the PPG signal at time Δ P m,j For the first m The first peak and the second m +1 time difference between wave peaks, Δ P m+1,j For the first m +1 peak and the first m + the time difference between the two peaks k For the number of all peaks, m Represents an index.

[0035] It's important to note that PPG signaling is a non-invasive method for recording vascular pulsation by measuring the attenuation of light reflected back from blood vessels and other tissues on the skin surface. The principle behind PPG signaling is that an LED light source emits light that penetrates the skin and is absorbed by the blood. Changes in blood flow with the heartbeat cause variations in light absorption, thus generating a signal. This technology is simple, low-cost, and widely used in health monitoring and physiological data acquisition. The peaks in the PPG signal correspond to the peak values ​​of blood pumped by the heart with each contraction. Therefore, the time difference between the m-th and (m+1)-th peaks is the m-th cardiac cycle. (Formula) This indicates the difference in duration between two consecutive cardiac cycles. When a patient is in pain, anxiety, or stress (such as discomfort caused by excessive heat from moxibustion), their sympathetic nervous system becomes dominant, and their heartbeat becomes rigid and rhythmic, like a metronome. In this state, the patient's heart rate variability is low, and their physiological stability is low.

[0036] In one embodiment, the invention further includes: calculating the first...j Before the heart rate variability coefficient at time 1, for the 1st j Time to the j - n The PPG signal is bandpass filtered at any given time.

[0037] It should be noted that bandpass filters work by setting two cutoff frequencies: a low cutoff frequency and a high cutoff frequency. Only signals between these two frequencies can pass through, while signals below and above these frequencies are attenuated.

[0038] PG signals are highly susceptible to two main types of noise during acquisition. The first type is low-frequency noise, primarily originating from the patient's breathing and slight limb movements. These activities cause a slow drift in the PPG signal's baseline (known as "baseline drift"), resulting in constantly changing peak heights. This can easily lead to algorithms missing peaks or misinterpreting respiratory waves as heartbeats. The high-pass portion of a bandpass filter (filtering out low frequencies) effectively removes this slow baseline drift. The second type is high-frequency noise, mainly from power frequency interference from moxibustion devices and other electronic instruments, as well as electronic noise from the sensor itself or weak electromyography (EMG) signals. This high-frequency noise can make the smooth PPG waveform "rough" or produce "sharp spikes," easily causing algorithms to misinterpret these noise peaks as "peaks," thus calculating incorrect, extremely short cardiac cycles. The low-pass portion of a bandpass filter (filtering out high frequencies) effectively removes these interferences. Therefore, by setting a reasonable passband (e.g., 0.5Hz to 5Hz, covering the heart rate range of most people), bandpass filtering ensures that the data used to calculate the first... j The coefficient of variation of heart rate at any given time is a signal that reflects the true heart rhythm.

[0039] Step S4: Calculate the tolerance coefficient based on the heart rate variability coefficient and the heat load coefficient of all moxibustion areas.

[0040] At any given time, the tolerance coefficient is inversely proportional to the heart rate variability coefficient, and the tolerance coefficient is directly proportional to the heat load coefficient of all moxibustion areas.

[0041] In one embodiment, the formula for calculating the tolerance coefficient at the current moment is: ; in, V The tolerance coefficient at the current moment, h The coefficient of variation of heart rate at the current moment. I i For the current moment, the first i The heat load coefficient of each moxibustion area k The scaling factor is the preset size. ω iFor the first i The weighting coefficients corresponding to each moxibustion area norm () is the standard normalization function, and exp() is an exponential function with the natural constant e as its base. The proportionality coefficient... k The value is 2.

[0042] It should be noted that the formula... The higher the value, the greater the physical risk to the patient, i.e., the greater the risk of burns. (Formula) This indicates a higher likelihood of a subjective physiological stress response from the patient. Based on this, the tolerance coefficient at the current moment... V The larger the value, the greater the risk for the patient undergoing moxibustion treatment.

[0043] In one embodiment, the calculation of the first i The weighting coefficient corresponding to each moxibustion area ω i The calculation formula is as follows: ,in N i For the first i The number of all pixels in the moxibustion area.

[0044] It should be noted that, since large-area heat exposure places a greater systemic burden on the body's thermoregulation system, this invention uses the number of pixels included in the moxibustion area to represent its size, and calculates the percentage of this area to the total moxibustion area. A weighted sum is then performed on the percentages corresponding to each moxibustion area. This ensures that a larger moxibustion area carries a higher weight in assessing the total heat load, i.e., calculating the heat load coefficient.

[0045] Step S5: Determine whether to trigger an alarm based on the tolerance coefficient, the heart rate variability coefficient, and the heat load coefficient of all moxibustion areas.

[0046] In one embodiment, determining whether to trigger an alarm based on the tolerance coefficient, the heart rate variability coefficient, and the heat load coefficient of all moxibustion areas includes: triggering an alarm in response to the tolerance coefficient being greater than a preset tolerance threshold; triggering an alarm in response to the heart rate variability coefficient being less than a preset heart rate variability threshold; and triggering an alarm in response to the heat load coefficient of at least one moxibustion area being greater than a preset heat load threshold.

[0047] It should be noted that the alarm is determined by the heat load coefficient of all moxibustion areas. No matter how good the patient's subjective feeling is (i.e., when the heart rate variability coefficient is large), an alarm must be triggered immediately if the objective temperature risk of any area of ​​skin exceeds the physical safety threshold.

[0048] Determining whether to trigger an alarm using the heart rate variability coefficient is to prevent patients from experiencing excessive stress. This ensures that regardless of how low the objective temperature is (i.e., when the heat load coefficient of all moxibustion areas is low), if the patient's physiological indicators (the heart rate variability coefficient being too low) indicate that they are in a state of tension, pain, or severe discomfort, an alarm must be triggered immediately.

[0049] To avoid systemic collapse of the patient's body due to the continuous accumulation of heat and physiological burden, even when the local temperature is not excessive and the physiological stress is not extreme, the total physical heat load (formula) is still within acceptable limits. (value) and physiological regulatory burden (formula) When the values ​​of ( ) are all in a critical state, an alarm needs to be triggered to avoid stress or other accidents to the patient during subsequent treatment due to high temperature.

[0050] The purpose of issuing an alarm is to notify medical staff to examine the patient's body and check the distance between the moxa wool or moxa stick and the patient's body to prevent the patient from being harmed.

[0051] In summary, this invention provides an intelligent moxibustion treatment method based on multimodal interaction. This method calculates and assesses the heat load coefficient for acute and low-temperature burn risks using infrared thermography, and simultaneously calculates the heart rate variability coefficient using PPG signals to objectively assess the patient's pain or stress state. This method combines these two objective indicators to calculate a comprehensive tolerance coefficient, and triggers an alarm when any one of the heat load, heart rate variability coefficient, or tolerance coefficient exceeds a threshold, rather than relying on the patient's subjective feedback, thus reducing the patient's burn risk. (Some patients, such as the elderly or diabetic patients, may have dulled skin sensation and be unable to accurately perceive temperature changes; or patients may fall asleep during treatment due to comfort and be unable to report discomfort in a timely manner.) This overcomes the unreliability of relying on subjective patient feedback, achieving dual objective monitoring of physical burn risk and physiological stress state, ensuring the safety of moxibustion treatment.

[0052] Figure 2 This is a schematic diagram illustrating the structure of an intelligent moxibustion treatment system based on multimodal interaction according to this embodiment.

[0053] This invention also provides an intelligent moxibustion treatment system based on multimodal interaction. For example... Figure 2 As shown, the system includes a processor and a memory, the memory storing computer program instructions, which, when executed by the processor, implement an intelligent moxibustion treatment method based on multimodal interaction according to the first aspect of the present invention.

[0054] The system also includes other components well known to those skilled in the art, such as communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

[0055] In this invention, the aforementioned memory can be any tangible medium containing or storing a program that can be used or combined with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store desired information and can be accessed by an application, module, or both. Any such computer storage medium can be part of a device or accessible to or connected to a device. Any application or module described in this invention can be implemented using computer-readable / executable instructions that can be stored or otherwise maintained by such a computer-readable medium.

[0056] In the description of this specification, "multiple" or "several" means at least two, such as two, three or more, unless otherwise explicitly specified.

[0057] While this specification has shown and described numerous embodiments of the invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in the practice of this invention.

Claims

1. A multi-modal interaction based intelligent moxibustion therapy method, characterized in that, The method comprises: obtaining a PPG signal of a patient and an infrared thermal image of all moxibustion areas of the patient when moxibustion is performed on the moxibustion areas of the patient; Calculate the heat load coefficient of each moxibustion area, where, at any given time, the first... i The heat load coefficient of the moxibustion area is related to the first i The grayscale value of all pixels in the moxibustion area is directly proportional to the maximum grayscale value. i Represents an index; calculating a heart rate variability coefficient for the PPG signal, wherein the heart rate variability coefficient at a first time instant is related to the difference between all adjacent peaks in the PPG signal from a second time instant to a third time instant, j j j - n j representing an index, n is a positive integer of a predetermined size, j - n > 0;​​​ calculating a tolerance coefficient according to the heart rate variability coefficient and the heat load coefficients of all moxibustion areas, wherein the tolerance coefficient is inversely proportional to the heart rate variability coefficient and is proportional to the heat load coefficients of all moxibustion areas at any time; determining whether to alarm according to the tolerance coefficient, the heart rate variability coefficient and the heat load coefficients of all moxibustion areas. 2.The intelligent moxibustion treatment method based on multi-modal interaction according to claim 1, characterized in that, The formula for calculating the heat load coefficient of the moxibustion area at the current time is: i ​ ； At the current moment, I i For the first i The heat load coefficient of each moxibustion area L max For the first i The maximum grayscale value of all pixels in the moxibustion area. L s,i For the first i The first moxibustion area s grayscale value of each pixel. N i For the first i The number of all pixels in each moxibustion area L T The threshold for acute burns is set at a preset size. L safe The preset size is the safety threshold for low-temperature burns, and L T > L safe , α The first weighting coefficient is of a preset size. β The second weighting coefficient is of a preset size, and α + β =1, s Represents an index. 3.The intelligent moxibustion treatment method based on multi-modal interaction of claim 2, wherein, the first weight coefficient α has a value of 0.3, and the second weight coefficient β has a value of 0.

7. 4.The intelligent moxibustion treatment method based on multi-modal interaction of claim 1, wherein, The formula for calculating the coefficient of variation of heart rate at the moment t is: j t = 1, 2, 3,..., n ； Among them, in the first j Time to the j - n In the PPG signal at time Δ P m,j For the first m The first peak and the second m +1 time difference between wave peaks, Δ P m+1,j For the first m +1 peak and the first m + the time difference between the two peaks k For the number of all peaks, m Represents an index.

5. The intelligent moxibustion treatment method based on multi-modal interaction according to claim 1, characterized in that, The formula for calculating the tolerance coefficient at the current time is: ； in, V The tolerance coefficient at the current moment, h The coefficient of variation of heart rate at the current moment. I i For the current moment, the first i The heat load coefficient of each moxibustion area k The scaling factor is the preset size. ω i For the first i The weighting coefficients corresponding to each moxibustion area norm () is the standard normalization function, and exp() is the exponential function with the natural constant e as the base.

6. The intelligent moxibustion treatment method based on multi-modal interaction according to claim 5, characterized in that, Also include calculating the weight coefficient corresponding to the i moxibustion area ω i The calculation formula is: Wherein N i The number of all pixel points in the i moxibustion area 7. The intelligent moxibustion treatment method based on multi-modal interaction according to claim 5, characterized in that, the proportionality coefficient k has a value of 2. 8.The intelligent moxibustion treatment method based on multi-modal interaction of claim 1, wherein, The method further comprises: Before calculating the coefficient of variation of the heart rate at the time t j , the PPG signal is band-pass filtered between the times t j and t j - n . 9.The intelligent moxibustion treatment method based on multi-modal interaction of claim 1, wherein, determining whether to alarm according to the tolerance coefficient, the heart rate variability coefficient and the heat load coefficients of all moxibustion areas comprises: performing alarm in response to the tolerance coefficient being greater than a preset tolerance threshold value; performing alarm in response to the heart rate variability coefficient being less than a preset heart rate variability threshold value; performing alarm in response to the heat load coefficient of at least one moxibustion area being greater than a preset heat load threshold value.

10. An intelligent moxibustion therapy system based on multi-modal interaction, comprising a processor and a memory, characterized in that, The memory stores a computer program, and the processor executes the computer program to implement the method according to any one of claims 1-9.

Images