A three-dimensional quantitative identification and evaluation system for human respiration based on flexible strain sensors

Abstract

A three-dimensional quantitative identification and evaluation system for human respiration based on flexible strain sensors belongs to the field of respiratory monitoring technology. The system includes: flexible strain sensors, respectively installed on fixed areas of the target user's chest, abdomen, and back, to collect resistance changes corresponding to skin deformation on the chest, abdomen, and back during respiration, monitoring skin surface deformation related to respiration; a data acquisition card, which converts the resistance change signals collected by the flexible strain sensors into recognizable digital respiratory signals; and a host computer containing a three-dimensional quantitative respiration identification software system. The three-dimensional quantitative respiration identification software system receives multi-channel respiratory signals transmitted from the data acquisition card and performs real-time preprocessing, baseline drift correction, respiratory cycle segmentation, three-dimensional respiratory motion feature extraction, three-dimensional respiratory pattern recognition, training effect evaluation, and result visualization on the multi-channel respiratory signals.

Description

Technical Field

[0001] This invention relates to the field of respiratory monitoring technology, specifically to a three-dimensional quantitative identification and evaluation system for human respiration based on a flexible strain sensor. Background Technology

[0002] Three-dimensional breathing can activate the parasympathetic nervous system and reduce sympathetic nerve excitability, which can improve posture, soothe nerves, and nourish internal organs to some extent. Currently, the criteria for judging three-dimensional breathing largely rely on the trainee's subjective judgment, making it difficult to grasp the training level and lacking objectivity. To verify the training effect, additional professional personnel are needed for evaluation, increasing costs for clinical patients and those training at home. Furthermore, current clinical evaluations of three-dimensional breathing largely rely on the experience of physicians and nurses or on related pulmonary function indicators. On the one hand, the lack of unified standards and quantitative conclusions in the experience of physicians and nurses makes it difficult to intuitively perceive the training effect; on the other hand, although relevant pulmonary function indicators are obtained through professional medical equipment, these indicators reflect the overall performance of lung function and cannot distinguish the effects achieved by three-dimensional breathing training, indicating a deficiency in the evaluation system.

[0003] Therefore, given the aforementioned shortcomings, how to provide an objective and quantitative three-dimensional respiratory recognition system and evaluation method with unified standards to assist trainees in quickly evaluating training results and providing targeted training suggestions, and to realize three-dimensional respiratory recognition and evaluation in multiple clinical and daily scenarios, has become an urgent problem to be solved. Summary of the Invention

[0004] The objective of this invention is to provide a three-dimensional quantitative identification system and evaluation method for human respiration based on a flexible strain sensor, in order to solve the problem that existing identification standards rely too much on the trainee's own or physician's experience, lack objective evaluation basis, and cannot effectively provide feedback on the results of three-dimensional respiration training, thereby failing to guide subsequent training.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A three-dimensional quantitative identification and evaluation system for human respiration based on a flexible strain sensor, the system comprising: a flexible strain sensor, a data acquisition card, and a host computer; The flexible strain sensor is an adhesive flexible strain sensor that can be directly attached to the surface of human skin. It is set in fixed areas on the chest, abdomen and back of the target user to collect the resistance change signal corresponding to the deformation of the skin surface of the chest, abdomen and back during the breathing process, and monitor the skin surface deformation related to breathing.

[0006] The data acquisition card converts the resistance change signal acquired by the flexible strain sensor into a recognizable digital breathing signal.

[0007] The host computer is equipped with a three-dimensional respiratory quantitative recognition software system. This software system receives multi-channel respiratory signals transmitted from the data acquisition card and performs real-time preprocessing, baseline drift correction, respiratory cycle segmentation, three-dimensional respiratory motion feature extraction, three-dimensional respiratory pattern recognition, training effect evaluation, and result visualization on the multi-channel respiratory signals.

[0008] Furthermore, the three-dimensional respiratory quantitative recognition software system includes a data access module, a signal preprocessing module, a baseline drift correction module, a respiratory cycle recognition module, a three-dimensional respiratory feature extraction module, a three-dimensional respiratory quantitative recognition module, a three-dimensional respiratory evaluation module, and a display and storage module.

[0009] The data access module is used to receive digital respiratory signals uploaded by the data acquisition card.

[0010] The signal preprocessing module is used to suppress outliers, reduce noise, and smooth the raw digital respiratory signal to obtain a stable respiratory waveform.

[0011] The baseline drift correction module is used to eliminate baseline drift caused by various factors during long-term monitoring of the flexible strain sensor.

[0012] The respiratory cycle recognition module is used to automatically identify the inspiratory start point, inspiratory peak, expiratory start point, expiratory trough, and complete respiratory cycle based on the baseline drift-corrected respiratory waveform.

[0013] The three-dimensional respiratory feature extraction module is used to extract the respiratory deformation features of the human body in three-dimensional space along the X, Y, and Z directions based on the resistance change signals of multiple flexible strain sensors on the chest, abdomen, and chest-abdomen junction of the target user, and further calculate the weight coefficients of the flexible strain sensors at each part in the X, Y, and Z directions.

[0014] The three-dimensional respiratory quantitative recognition module further normalizes the comprehensive respiratory deformation features in the X, Y, and Z directions to obtain the weight ratio of the sensors in each of the X, Y, and Z directions. Based on the weight ratio of the sensors in each of the X, Y, and Z directions, the three-dimensional respiratory state of the target user in the current respiratory cycle is determined.

[0015] The three-dimensional breathing evaluation module is used to comprehensively evaluate the three-dimensional breathing level of the target user within a monitoring cycle or training cycle based on the single-cycle three-dimensional breathing comprehensive score output by the three-dimensional breathing quantitative identification module.

[0016] The display and storage module is used to display the resistance changes of each sensor in real time and store the collected data; Furthermore, the 3D breathing feature extraction module: 1) The flexible strain sensor is set on the monitoring part of the human chest, abdomen and the chest-abdomen junction; according to the pasting direction and anatomical location of the flexible strain sensor, it corresponds to the three spatial dimensions of X, Y and Z in the human breathing process. The left and right lateral expansion direction of the human body is defined as the X direction, the front and back expansion direction of the human body is defined as the Y direction, and the longitudinal traction direction of the human head and feet is defined as the Z direction. 2) Within each respiratory cycle, the three-dimensional respiratory feature extraction module calculates the respiratory deformation characteristic values ​​of each flexible strain sensor. The respiratory deformation characteristic values ​​are the peak-to-valley difference, normalized resistance change, or normalized strain amplitude of the sensor signal within that respiratory cycle.

[0017] 3) According to the human body spatial direction corresponding to the flexible strain sensor, the comprehensive respiratory deformation characteristics in the X, Y and Z directions are calculated by weighting. The weighting coefficients corresponding to the respiratory deformation characteristic values ​​of each flexible strain sensor in the comprehensive respiratory deformation characteristics of each direction are determined according to the resistance change, strain amplitude and signal stability contribution of each monitoring part in the same direction.

[0018] Furthermore, the three-dimensional respiratory quantitative recognition module: Let the weighting ratio of human breathing in the X direction be . A The weighting ratio of breathing in the Y direction is B The weighting ratio of breathing in the Z direction is C ; Will A , B、C As a core quantitative indicator for determining three-dimensional respiratory patterns, among which, A This indicates the contribution of respiratory expansion in the X direction of the human body. B This indicates the contribution of respiratory expansion in the Y direction of the human body. C This represents the contribution of respiratory expansion in the human body in the Z direction.

[0019] Then, when the target user's actual measured three-dimensional respiratory weight ratio A , B、C Compared with standard ratio A 0、 B 0、 C When the deviation of 0 is within the preset allowable range, the system determines that the target user's current breathing cycle meets the requirements of three-dimensional coordinated breathing.

[0020] Furthermore, the actual measured three-dimensional respiratory weight ratio of the target user A , B、C Compared with standard ratio A 0、 B 0、 CA deviation of 0 is one evaluation criterion. In addition, a three-dimensional respiratory severity scoring system is established to comprehensively assess the three-dimensional respiratory status of the target user's current respiratory cycle. The three-dimensional respiratory severity scoring system includes three-dimensional proportional fit scoring, three-dimensional effective participation scoring, three-dimensional synergy scoring, respiratory rhythm stability scoring, and respiratory amplitude effectiveness scoring. The three-dimensional respiratory quantitative recognition module calculates the above score for each effective respiratory cycle and obtains a single-cycle three-dimensional respiratory comprehensive score. S k : (14) in, S k The three-dimensional respiratory comprehensive score for the k-th respiratory cycle is 100 points. S 1k The three-dimensional proportional fit is scored. S 2k Scoring of effective participation in three dimensions. S 3k For three-dimensional synergy scoring, S 4k Assess respiratory rhythm stability. S 5k The effectiveness score for respiratory amplitude is given.

[0021] Three-dimensional proportional fit rating: The actual measured A , B、C Compared with the preset standard ratio A 0、 B 0、 C Compare with 0 and assign scores based on the degree of deviation.

[0022] Three-dimensional effective participation scoring: This assesses whether the overall respiratory deformation characteristics in the X, Y, and Z directions reach the corresponding minimum effective thresholds, and assigns scores based on the degree to which the thresholds are reached and the number of directions. It evaluates whether effective respiratory deformation occurs in all three directions (X, Y, and Z), avoiding situations where the proportions are close to the standard value but the overall respiratory motion amplitude is too small.

[0023] Three-dimensional directional synergy score: used to evaluate the degree of temporal synchronization of respiratory movements in the X, Y, and Z directions.

[0024] Respiratory rhythm stability score: Used to evaluate whether the target user's breathing rhythm is stable over multiple consecutive respiratory cycles.

[0025] Breathing amplitude effectiveness score: used to evaluate whether the target user's breathing deformation amplitude is within the effective training range.

[0026] Furthermore, the calculation method for the three-dimensional synergy score is as follows: The peak times of the respiratory waveforms in the X, Y, and Z directions within the same respiratory cycle are extracted and denoted as follows: t X , t Y , t Z And calculate the maximum phase difference between directions: (15) Let the duration of the current respiratory cycle be... T Then the normalized phase difference is: (16) Different levels of scoring will be assigned according to the following criteria: when A value less than 0.10 indicates high synchronization; when 0.10 < <0.30 indicates basic synchronization; when 0.30 < <0.50 indicates mild asynchrony; when 0.50 < <0.90 indicates significant asynchrony; when A value greater than 0.90 indicates severe asynchrony.

[0027] Furthermore, the respiratory rhythm stability score is calculated as follows: Calculate the coefficient of variation of cycle duration for N consecutive effective respiratory cycles: (17) in, s T The standard deviation of the duration of a continuous respiratory cycle. This represents the average duration of a continuous respiratory cycle.

[0028] Calculate the coefficient of variation of deformation amplitude over N consecutive effective respiratory cycles: (18) in, s M The standard deviation of the overall deformation amplitude during a continuous respiratory cycle. This represents the average value of the comprehensive deformation amplitude during a continuous respiratory cycle.

[0029] Different levels of scoring will be assigned according to the following criteria: when CV T <10%, and CV M A percentage <15% indicates a highly stable rhythm; when CV T< 15%, and CV M < 20% indicates that the rhythm is basically stable; When CV T < 25%, or CV M < 30% indicates that the rhythm fluctuates slightly; When the cycle or amplitude fluctuates greatly, it indicates that the rhythm is significantly unstable; When an effective respiratory cycle cannot be continuously recognized, it indicates that there is no stable respiratory cycle; Furthermore, the calculation method of the respiratory amplitude effectiveness score is as follows: Calculate the three-dimensional comprehensive respiratory amplitude of the current respiratory cycle: (19) And compare it with the individual reference respiratory amplitude M 0 to obtain the normalized amplitude: (20) Among them, M 0 can be obtained through resting natural breathing, standard training breathing or individual calibration breathing.

[0030] Assign scores of different levels according to the following criteria: When 0.8 ≤ Q ≤ 1.2, it indicates that the amplitude is appropriate; When 0.6 ≤ Q < 0.8 or 1.2 < Q ≤ 1.5, it indicates that the amplitude is basically appropriate; When 0.4 ≤ Q < 0.6 or 1.5 < Q ≤ 1.8, it indicates that the amplitude is too small or too large; When Q < 0.4 or Q > 1.8, it indicates that the amplitude is significantly abnormal; When an effective respiratory deformation cannot be recognized, it indicates that there is no effective amplitude; Furthermore, the specific functions of the three-dimensional respiration evaluation module: Identify N effective respiratory cycles and calculate the three-dimensional respiration comprehensive score of each respiratory cycle respectively S k The system further calculates the average three-dimensional respiration score within this monitoring period: (21) Among them, is the average three-dimensional respiration score of the target user within the current monitoring period.

[0031] Then the three-dimensional respiration compliance rate: (22) Among them, N qThe number of breathing cycles required to achieve the preset target score. N This represents the total number of valid respiratory cycles identified within the current monitoring period.

[0032] The three-dimensional breathing level of the target user is comprehensively evaluated based on the average three-dimensional breathing score and the three-dimensional breathing compliance rate within a monitoring or training cycle.

[0033] Furthermore, the standard for comprehensively evaluating the three-dimensional breathing level of target users within a monitoring or training cycle is based on the average three-dimensional breathing score and the three-dimensional breathing target achievement rate: Based on the average three-dimensional respiratory score and three-dimensional respiratory target achievement rate K The target user's three-dimensional breathing level is divided into multiple levels: when When ≥90 and K≥85%, the evaluation result is highly coordinated three-dimensional breathing, and the three-dimensional breathing level is set as Level 1; When 80≤ When <90 and K≥70%, the evaluation result is good three-dimensional breathing, and the three-dimensional breathing level is set as level two; When 70≤ When <80 and K≥50%, the evaluation result is basic three-dimensional breathing, and the three-dimensional breathing level is set to level three; When 60≤ When K < 70 or K < 50%, the evaluation result is three-dimensional respiratory insufficiency, and the three-dimensional respiratory level is set to level four. when When the value is <60, the evaluation result is uncoordinated three-dimensional breathing or obvious compensatory breathing, and the three-dimensional breathing level is set to level five.

[0034] The beneficial effects of this invention are as follows: Utilizing flexible strain sensors to synchronously acquire respiratory deformation signals from the human chest, abdomen, and the chest-abdomen junction via multiple channels enables non-invasive and continuous monitoring of three-dimensional respiratory motion, avoiding errors caused by traditional manual observation and subjective judgment. Preprocessing the acquired respiratory signals through techniques such as outlier suppression, low-pass filtering, smoothing, and baseline drift correction improves the stability and accuracy of the respiratory waveform. Based on the sensor placement and orientation, the human respiratory deformation signals are mapped to the X, Y, and Z spatial directions, and the weight ratio of three-dimensional breathing in these three directions is calculated, thus transforming the three-dimensional breathing training effect into quantifiable and comparable numerical indicators. Furthermore, by combining indicators such as respiratory direction ratio, directional participation, directional coordination, respiratory rhythm stability, and respiratory amplitude effectiveness, a comprehensive three-dimensional breathing evaluation system is established. This system can determine whether the user has achieved a three-dimensional coordinated breathing state and identify problems such as chest-abdomen asynchrony, unidirectional compensation, or insufficient respiratory amplitude, providing a basis for subsequent individualized training guidance.

[0035] Therefore, this invention can improve the objectivity, accuracy and convenience of three-dimensional respiratory recognition and evaluation, and is applicable to scenarios such as clinical rehabilitation training, respiratory function training, home health management and exercise training. Attached Figure Description

[0036] Figure 1 This is a framework diagram of a three-dimensional respiratory quantitative recognition system. Detailed Implementation

[0037] A three-dimensional quantitative identification and evaluation system for human respiration based on a flexible strain sensor, comprising: Flexible strain sensor, data acquisition card, and host computer; The flexible strain sensor is an adhesive flexible strain sensor that can be directly attached to the surface of human skin. It is set in fixed areas on the chest, abdomen and back of the target user to collect the resistance change signal corresponding to the deformation of the skin surface of the chest, abdomen and back during the breathing process, and monitor the skin surface deformation related to breathing.

[0038] The data acquisition card converts the resistance change signal acquired by the flexible strain sensor into a recognizable digital breathing signal.

[0039] The host computer is equipped with a three-dimensional respiratory quantitative recognition software system. This software system receives multi-channel respiratory signals transmitted from the data acquisition card and performs real-time preprocessing, baseline drift correction, respiratory cycle segmentation, three-dimensional respiratory motion feature extraction, three-dimensional respiratory pattern recognition, training effect evaluation, and result visualization on the multi-channel respiratory signals.

[0040] The three-dimensional respiratory quantitative recognition software system includes a data access module, a signal preprocessing module, a baseline drift correction module, a respiratory cycle recognition module, a three-dimensional respiratory feature extraction module, a three-dimensional respiratory quantitative recognition module, a three-dimensional respiratory evaluation module, and a display and storage module.

[0041] The data access module is used to receive digital respiratory signals uploaded by the data acquisition card.

[0042] The data access module synchronously reads data from each channel according to a preset sampling frequency. In a preferred embodiment, the sampling frequency is 100 Hz. This preset sampling frequency is higher than the main frequency range of human respiratory signals, which can fully characterize the waveform changes during inhalation, exhalation, and respiratory transitions, and provide sufficient data redundancy for subsequent digital filtering and anomaly identification.

[0043] The signal preprocessing module is used to suppress outliers, reduce noise, and smooth the raw digitized respiratory signal to obtain a stable respiratory waveform. Specifically: (1) The signal preprocessing module first judges the amplitude change between adjacent sampling points. When the instantaneous change amplitude between adjacent sampling points exceeds a preset threshold, the sampling point is judged as an abnormal point. In a preferred embodiment, the preset threshold is 30%. Since human breathing is a low-frequency, continuous and smooth physiological movement, there are usually no large instantaneous jumps between adjacent sampling points. Therefore, an instantaneous change exceeding 30% can be considered to be caused by communication error, sensor contact jitter, mechanical disturbance or external impact.

[0044] (2) The signal preprocessing module uses a fourth-order Butterworth low-pass filter to filter the signal after anomaly correction. In a preferred embodiment, the cutoff frequency of the low-pass filter is 1.5 Hz. This cutoff frequency can retain the main waveform information under normal breathing and rapid breathing conditions, while effectively suppressing high-frequency jitter and measurement noise above the main breathing frequency range.

[0045] (3) The signal preprocessing module uses the exponential moving average algorithm to perform secondary smoothing on the filtered breathing waveform.

[0046] The baseline drift correction module is used to eliminate baseline drift caused by factors such as temperature changes, material creep, changes in skin contact state, and slow changes in body position during long-term monitoring of flexible strain sensors.

[0047] Specifically, the baseline drift correction module first identifies stable data segments with small fluctuations in the collected data and calculates the median of these stable data segments, using it as the baseline reference value for the current monitoring period. Using the median as the baseline reference value can reduce the impact of local spikes or short-term abnormal fluctuations on baseline estimation.

[0048] Subsequently, the baseline drift correction module performs low-order polynomial fitting on the signal data after the stationary phase to estimate the slowly varying drift trend in the signal. In a preferred embodiment, a fourth-order polynomial is used to fit the baseline drift trend.

[0049] The respiratory cycle recognition module is used to automatically identify the inspiratory start point, inspiratory peak, expiratory start point, expiratory trough, and complete respiratory cycle based on the baseline drift-corrected respiratory waveform.

[0050] Specifically, the respiratory cycle identification module detects extreme points in the corrected respiratory waveform and, in conjunction with the minimum respiratory cycle, minimum peak-to-valley difference, and waveform slope change conditions, eliminates spurious peaks and valleys caused by noise. The respiratory cycle identification module defines the time interval between two adjacent inspiratory initiations as a complete respiratory cycle, or the time interval between two adjacent expiratory valleys as a complete respiratory cycle.

[0051] For each respiratory cycle, the respiratory cycle recognition module calculates the inspiratory time, expiratory time, respiratory cycle duration, respiratory rate, peak and trough amplitude, upward slope, downward slope, and waveform stability index.

[0052] The three-dimensional respiratory feature extraction module is used to extract the respiratory deformation features of the human body in three-dimensional space along the X, Y, and Z directions based on the resistance change signals of multiple flexible strain sensors on the target user's chest, abdomen, and chest-abdomen junction. It further calculates the weighting coefficients of the flexible strain sensors at each location in the X, Y, and Z directions. In one embodiment, the flexible strain sensors are installed at 16 monitoring sites on the human body's chest, abdomen, and chest-abdomen junction. Specifically, these include: Seven flexible strain sensors are installed in the chest area, including a horizontal sensor on the front of the chest, a vertical sensor on the front of the chest, a horizontal sensor on the side of the chest, a vertical sensor on the side of the chest, a horizontal sensor on the chest and back, a vertical sensor on the chest and back spine, and a vertical sensor on the chest and back spine. Seven flexible strain sensors are installed in the abdominal area, including anterior horizontal sensor, anterior vertical sensor, lateral horizontal sensor, lateral vertical sensor, lateral horizontal sensor, lateral vertical sensor, lateral horizontal sensor, lateral vertical sensor, lateral vertical sensor, and lateral vertical sensor of the spine. Two flexible strain sensors are installed at the chest-abdomen junction: a left-side vertical sensor located on the left side of the junction between the ribs and the abdomen, and a right-side vertical sensor located on the right side of the junction between the ribs and the abdomen.

[0053] The aforementioned 16 flexible strain sensors correspond to the X, Y, and Z spatial dimensions of the human respiratory process, based on their adhesion direction and anatomical location. Among them, the horizontally attached sensors are mainly used to characterize the expansion deformation of the human torso in the lateral or anterior-posterior directions, while the vertically attached sensors are mainly used to characterize the longitudinal tensile deformation of the human torso.

[0054] Specifically, the lateral expansion direction of the human body is defined as the X direction, the anterior-posterior expansion direction of the human body is defined as the Y direction, and the longitudinal traction direction of the human head and feet is defined as the Z direction.

[0055] The X-direction includes a horizontally attached sensor on the chest. x 1. Sensors are attached horizontally to the front of the abdomen. x 2. Sensors attached horizontally to the chest and back x3 and the transverse sensor on the abdomen and back x 4; The Y-direction includes transverse sensors attached to the chest. y 1 and ventral transverse sensor y 2; Z-direction includes vertically attached sensors on the chest. z 1. Vertical sensor attached to the front of the abdomen z 2. Vertical sensor attached to chest and back z 3. Vertically attached sensors on the abdomen and back z 4. Vertical sensor attached to the chest side z 5. Vertically attached sensor on the abdomen z 6. Vertically attach the sensor at the chest-abdomen connection. z 7. Vertical sensor for chest and back spine z 8 and abdominal and back spine vertical adhesive sensors z 9.

[0056] The vertically attached sensor on the left side of the chest-abdomen connection and the vertically attached sensor on the right side of the chest-abdomen connection can be denoted as z. 7L and z 7R The three-dimensional respiratory feature extraction module obtains the comprehensive longitudinal feature z7 at the thoracic-abdomen junction through averaging, weighted averaging, or symmetry correction, i.e.: (1) or: (2) in, or L and or R The weighting coefficients are for the vertically attached sensors at the left and right chest-abdomen junctions, respectively, and satisfy the following: (3) Within each breathing cycle, the three-dimensional breathing feature extraction module calculates the breathing deformation characteristic values ​​of each flexible strain sensor. These breathing deformation characteristic values ​​can be the peak-to-valley difference of the sensor signal, the change in normalized resistance, or the normalized strain amplitude within that breathing cycle.

[0057] In a preferred embodiment, the peak-to-valley difference of the sensor signal within a single respiratory cycle is used as the respiratory deformation characteristic value of the corresponding monitoring site: (4) in, d i Let be the respiratory deformation characteristic value of the i-th flexible strain sensor during the current respiratory cycle. P i This represents the peak value within the current respiratory cycle. V i This represents the lowest value within the current respiratory cycle.

[0058] To eliminate the influence of differences in initial resistance, adhesion tightness, sensitivity, and individual skin condition among different flexible strain sensors, the 3D breathing feature extraction module further normalizes the breathing deformation feature values ​​of each sensor to obtain normalized feature values: (5) in, d i,ref This is the reference deformation value of the i-th sensor during baseline breathing, calibrated breathing, or historical stable breathing cycles.

[0059] After obtaining the normalized respiratory deformation characteristic values ​​of each monitoring site, the three-dimensional respiratory feature extraction module calculates the comprehensive respiratory deformation characteristics in the X, Y, and Z directions according to the human body spatial direction corresponding to the flexible strain sensor.

[0060] Among them, the comprehensive respiratory deformation characteristic X in the X direction can be represented as: (6) The comprehensive respiratory deformation characteristics Y in the Y direction can be represented as: (7) The comprehensive respiratory deformation characteristics Z in the Z-direction can be represented as: (8) in, α 1 to α 4 represents the weighting coefficient of each monitoring point in the X direction. β 1 to β 2 represents the weighting coefficient of each monitoring location in the Y direction. c 1 to c 9 represents the weighting coefficient of each monitoring location in the Z direction.

[0061] The weighting coefficients can be determined based on the resistance change, strain amplitude, and signal stability contribution of each monitoring point in the same direction. Specifically, in the same direction, the weighting coefficient of a certain monitoring point can be determined based on the proportion of the respiratory deformation characteristic value of that monitoring point in the sum of the respiratory deformation characteristic values ​​of all monitoring points in that direction.

[0062] The three-dimensional respiratory quantitative recognition module further normalizes the comprehensive respiratory deformation features in the X, Y, and Z directions to obtain the weight ratio of the sensors in each of the X, Y, and Z directions. Based on the weight ratio of the sensors in each of the X, Y, and Z directions, the three-dimensional respiratory state of the target user in the current respiratory cycle is determined.

[0063] Specifically, let the weighting ratio of human breathing in the X direction be . A The weighting ratio of breathing in the Y direction is BThe weighting ratio of breathing in the Z direction is C ,but: (9) (10) (11) in: (12) Therefore, the three-dimensional respiratory quantitative recognition module can obtain the respiratory weight ratio in three dimensions of the human body during the current respiratory cycle of the target user. A , B , C ).

[0064] In one implementation, the three-dimensional respiratory quantitative recognition module will ( A , B , C This serves as a core quantitative indicator for assessing three-dimensional respiratory patterns. Among them, A This indicates the contribution of respiratory expansion in the X direction of the human body. B This indicates the contribution of respiratory expansion in the Y direction of the human body. C This represents the contribution of respiratory expansion in the human body in the Z direction.

[0065] Then, when the target user's actual measured three-dimensional breathing weight ratio ( A , B , C ) and standard ratio ( A 0, B 0, C When the deviation of 0) is within the preset allowable range, the three-dimensional respiratory quantitative identification module determines that the target user's current respiratory cycle meets the requirements of three-dimensional coordinated breathing.

[0066] The three-dimensional respiratory quantitative identification module is used to determine the three-dimensional respiratory weight ratio ( A , B , C This determines the three-dimensional respiratory state of the target user during the current respiratory cycle.

[0067] In one implementation, the three-dimensional respiratory quantitative identification module pre-establishes a standard three-dimensional respiratory template, the standard three-dimensional respiratory template including X-direction weights. A 0, Y-direction weights B 0, Z-direction weight C 0, and satisfies: (13) To avoid misjudgments caused by judging the three-dimensional respiratory state solely based on a single ratio, the three-dimensional respiratory quantitative identification module further establishes a three-dimensional respiratory degree scoring system. This system includes at least three-dimensional proportional consistency scoring, three-dimensional effective participation scoring, three-dimensional synergy scoring, respiratory rhythm stability scoring, and respiratory amplitude effectiveness scoring.

[0068] The three-dimensional respiratory quantitative recognition module calculates the above score for each effective respiratory cycle and obtains a single-cycle three-dimensional respiratory comprehensive score. S k : (14) in, S k The three-dimensional respiratory comprehensive score is for the k-th respiratory cycle, with a maximum score of 100 points; S 1k The three-dimensional proportional fit is scored. S 2k Scoring of effective participation in three dimensions. S 3k For three-dimensional synergy scoring, S 4k Assess respiratory rhythm stability. S 5k The effectiveness score for respiratory amplitude is given.

[0069] The three-dimensional proportional fit score is used to evaluate the target user's current respiratory cycle. A , B , C Does the weighting ratio of breathing in the three directions approximate the standard three-dimensional breathing ratio?

[0070] In one implementation, the three-dimensional proportional fit score is out of 35 points. The three-dimensional respiratory quantitative recognition module will use the actually measured ( A , B , C ) and preset standard ratio ( A 0, B 0, C 0) Compare and assign scores based on the degree of deviation.

[0071] Scoring can be conducted according to the following criteria:

[0072] The three-dimensional effective participation score is used to evaluate whether effective respiratory deformation is generated in the X, Y, and Z directions, and to avoid situations where the proportion is close to the standard value but the overall respiratory motion amplitude is too small.

[0073] In one implementation, the maximum score for the three-dimensional effective participation is 15 points. The three-dimensional respiratory quantitative recognition module determines whether the comprehensive respiratory deformation characteristics in the X, Y, and Z directions reach the corresponding minimum effective threshold.

[0074] Scoring can be conducted according to the following criteria:

[0075] The three-dimensional synergy score is used to evaluate the degree of temporal synchronization of respiratory movements in the X, Y, and Z directions.

[0076] In an ideal three-dimensional breathing process, the thorax, abdomen, and the thoracoabdominal junction should expand and contract in a coordinated manner within the same respiratory cycle. If one direction is significantly ahead or behind, it indicates that the target user has problems such as asynchronous thoracoabdominal expansion, delayed lateral expansion, or longitudinal traction compensation.

[0077] In one implementation, the three-dimensional directional synergy score has a maximum score of 15 points. The three-dimensional respiratory quantitative recognition module extracts the peak times of the respiratory waveforms in the X, Y, and Z directions within the same respiratory cycle, and records them as follows: t X , t Y , t Z And calculate the maximum phase difference between directions: (15) Let the duration of the current respiratory cycle be... T Then the normalized phase difference is: (16) Scoring can be conducted according to the following criteria:

[0078] The respiratory rhythm stability score is used to evaluate whether a target user's breathing rhythm is stable across multiple consecutive respiratory cycles. Three-dimensional breathing not only requires reasonable directional proportions but also requires the breathing process to be continuous, stable, and repeatable.

[0079] In one implementation, the respiratory rhythm stability score has a maximum score of 20 points.

[0080] The three-dimensional respiratory quantitative identification module calculates the coefficient of variation of cycle duration for N consecutive effective respiratory cycles: (17) in, s T The standard deviation of the duration of a continuous respiratory cycle. This represents the average duration of a continuous respiratory cycle.

[0081] The three-dimensional respiratory quantitative identification module calculates the coefficient of variation of deformation amplitude over N consecutive effective respiratory cycles: (18) in, s M The standard deviation of the overall deformation amplitude during a continuous respiratory cycle. This represents the average value of the comprehensive deformation amplitude during a continuous respiratory cycle.

[0082] Scoring can be conducted according to the following criteria:

[0083] The breathing amplitude effectiveness score is used to evaluate whether the target user's breathing deformation amplitude is within the effective training range. If the breathing amplitude is too small, even if the three-dimensional proportions are good, it may indicate insufficient training; if the breathing amplitude is too large, it may indicate compensatory movements such as shrugging shoulders, puffing out the chest, collapsing the waist, or excessive exertion.

[0084] In one implementation, the respiratory amplitude effectiveness score has a maximum score of 15 points.

[0085] The three-dimensional respiratory quantitative identification module calculates the three-dimensional comprehensive respiratory amplitude of the current respiratory cycle: (19) And compare it with the individual baseline respiratory amplitude. M By comparing with 0, the normalized amplitude is obtained: (20) in, M 0 can be obtained through resting natural breathing, standard training breathing, or individual calibrated breathing.

[0086] Scoring can be conducted according to the following criteria:

[0087] The three-dimensional breathing evaluation module is used to comprehensively evaluate the three-dimensional breathing level of a target user within a monitoring cycle or training cycle based on the three-dimensional breathing comprehensive score output by the three-dimensional breathing quantitative identification module.

[0088] Within a monitoring cycle, the three-dimensional respiratory assessment module identifies N valid respiratory cycles and calculates a comprehensive three-dimensional respiratory score for each respiratory cycle. S k The three-dimensional respiratory quantitative identification module further calculates the average three-dimensional respiratory score within this monitoring period: (twenty one) in, The average three-dimensional respiratory score of the target user during the current monitoring period.

[0089] The three-dimensional respiratory target achievement rate is: (twenty two) in, N q The number of breathing cycles required to achieve the preset target score. N This represents the total number of valid respiratory cycles identified within the current monitoring period. In one implementation, the single-cycle three-dimensional respiratory comprehensive score... S k A score of ≥80 is considered a compliant three-dimensional respiratory cycle. This is based on the average three-dimensional respiratory score. and three-dimensional respiratory target achievement rate K The target user's three-dimensional breathing level is divided into multiple levels.

[0090] Level 1 indicates that the target user can stably and coordinately complete three-dimensional breathing; Level 2 indicates that the target user has basically mastered the three-dimensional breathing pattern, but there are still slight proportional deviations or rhythm fluctuations; Level 3 indicates that the target user can partially complete three-dimensional breathing, but the stability and coordination are insufficient; Level 4 indicates that the target user's three-dimensional breathing training is insufficient and needs to be corrected in a targeted manner; Level 5 indicates that the target user has not yet formed an effective three-dimensional breathing pattern, or there are obvious problems such as unidirectional compensation and chest-abdominal asynchrony.

[0091] The display and storage module is used to display the resistance changes of each sensor in real time and store the collected data; Through the above scoring system, this invention can not only determine whether the target user has reached a three-dimensional breathing state, but also quantify the degree of three-dimensional breathing from multiple perspectives such as the proportion of three-dimensional directions, the degree of participation in three-dimensional directions, the coordination between directions, the stability of breathing rhythm, and the effectiveness of breathing amplitude.

[0092] Compared with evaluation methods that rely solely on the subjective observation of medical staff or the trainee's self-perception, this invention can transform the effect of three-dimensional breathing training into calculable, displayable, and comparable quantitative results, and can further identify the specific deficiencies of the target user in three-dimensional breathing training, thereby providing a basis for subsequent individualized training.

[0093] Example 1: Implementation of a Three-Dimensional Breath Quantitative Identification and Evaluation System This embodiment provides a specific implementation of a three-dimensional quantitative identification and evaluation system for human respiration based on flexible strain sensors. The system includes flexible strain sensors, a data acquisition card, and a host computer. The flexible strain sensors employ an adhesive structure, allowing them to adhere to the surface of human skin to collect resistance changes caused by deformation of the skin surface in the chest, abdomen, and back during respiration. The data acquisition card converts the resistance change signals output by each flexible strain sensor into digital respiratory signals and transmits them to the host computer. The host computer contains a three-dimensional quantitative respiration identification software system, used for multi-channel respiratory signal access, preprocessing, baseline drift correction, respiratory cycle identification, three-dimensional respiratory feature extraction, three-dimensional quantitative respiration identification, three-dimensional respiratory evaluation, and result display and storage.

[0094] In this embodiment, flexible strain sensors are arranged in the chest region, abdominal region, and chest-abdomen junction of the human body. These sensors together form a multi-channel respiratory deformation acquisition structure. Based on the sensor's attachment direction and location, the system divides the deformation direction during human respiration into three directions: X, Y, and Z.

[0095] During monitoring, each sensor is first attached to the corresponding part of the target user, and the data acquisition card synchronously collects respiratory signals from each channel according to the preset sampling frequency.

[0096] For each effective respiratory cycle, the three-dimensional respiratory feature extraction module calculates the respiratory deformation characteristic values ​​of each channel flexible strain sensor. The respiratory deformation characteristic values ​​can be the peak-to-valley difference of the sensor signal, the change in normalized resistance, or the normalized strain amplitude within that respiratory cycle.

[0097] After obtaining the normalized respiratory deformation feature values ​​for each channel, the system calculates the comprehensive respiratory deformation features in the X, Y, and Z directions according to the spatial orientation corresponding to each sensor. Further, the three-dimensional respiratory quantitative recognition module normalizes the comprehensive respiratory deformation features in the X, Y, and Z directions to obtain the weight ratios A, B, and C of human respiration in these three directions. The system compares A, B, and C with preset standard three-dimensional respiratory ratios A0, B0, and C0 to determine whether the current respiratory cycle meets the requirements for three-dimensional coordinated breathing.

[0098] To avoid misjudgments caused by relying solely on directional proportions, the system further calculates a three-dimensional directional proportion consistency score, a three-dimensional directional effective participation score, a three-dimensional directional synergy score, a respiratory rhythm stability score, and a respiratory amplitude effectiveness score, and obtains a single-cycle three-dimensional respiratory comprehensive score based on these scores. The three-dimensional respiratory evaluation module calculates the average three-dimensional respiratory score and the three-dimensional respiratory target achievement rate based on the comprehensive score of multiple consecutive effective respiratory cycles, and outputs the target user's three-dimensional respiratory evaluation results according to preset level standards.

[0099] The display and storage module is used to display the respiratory waveforms of each channel, the respiratory weight ratios in the X, Y, and Z directions, the single-cycle three-dimensional respiratory comprehensive score, the average three-dimensional respiratory score, the three-dimensional respiratory compliance rate, and the three-dimensional respiratory level evaluation results in real time, and to store the collected data and evaluation results locally.

[0100] Through this embodiment, the system can convert the deformation state of various parts of the human body during three-dimensional respiration into quantitative indicators that can be collected, calculated, displayed, and stored, thereby achieving objective identification and evaluation of the three-dimensional respiration state.

[0101] Example 2: Implementation Case of Uncoordinated Breathing State Recognition This embodiment provides an implementation method for identifying uncoordinated or compensatory breathing states using the above-described system. This embodiment employs the same flexible strain sensor arrangement, data acquisition method, and signal processing method as Embodiment 1.

[0102] In this embodiment, the target user wears a flexible strain sensor for respiratory monitoring. The system simultaneously collects respiratory deformation signals from multiple monitoring sites, including the chest, abdomen, back, and the chest-abdomen junction. After data acquisition, the three-dimensional respiratory quantitative recognition software system in the host computer sequentially performs signal preprocessing, baseline drift correction, and respiratory cycle identification, and extracts respiratory deformation feature values ​​for each channel within each effective respiratory cycle.

[0103] When a target user exhibits significant expansion in a certain area, insufficient participation in other areas, asynchronous chest and abdominal movements, or unidirectional compensation during breathing, the system can identify this through comprehensive respiratory deformation features in the X, Y, and Z directions. Specifically, if the comprehensive respiratory deformation feature in a certain direction is significantly lower than the corresponding effective threshold, the system determines that the participation in that direction is insufficient; if the weight ratio of a certain direction is significantly higher than that of other directions, the system determines that there is a unidirectional dominant or compensatory breathing trend; if the peak times in the X, Y, and Z directions differ significantly, the system determines that the coordination in the three dimensions is insufficient; if the cycle duration or comprehensive deformation amplitude fluctuates significantly across multiple consecutive breathing cycles, the system determines that the respiratory rhythm stability is insufficient.

[0104] The system compares the weighted proportions A, B, and C of the current respiratory cycle in three dimensions with preset standard proportions A0, B0, and C0, and performs a comprehensive score based on the effective participation of the three dimensions, directional coordination, respiratory rhythm stability, and respiratory amplitude effectiveness. When the evaluation results show that the directional proportions deviate from the standard range, some directions are under-participated, the synchronicity between directions is poor, or the respiratory amplitude is not within the effective range, the system outputs the evaluation results of uncoordinated breathing, insufficient three-dimensional breathing, or compensatory breathing.

[0105] Through this embodiment, the system can not only determine whether the target user has achieved a three-dimensional coordinated breathing state, but also further identify the specific deficiencies and compensatory behaviors of the target user during the three-dimensional breathing training process, providing a basis for subsequent targeted training and individualized guidance.

Claims

1. A three-dimensional quantitative identification and evaluation system for human respiration based on a flexible strain sensor, characterized in that, The aforementioned three-dimensional human respiration quantitative identification and evaluation system includes: a flexible strain sensor, a data acquisition card, and a host computer; The flexible strain sensor is an adhesive flexible strain sensor that can be directly attached to the surface of human skin. It is set in fixed areas on the chest, abdomen and back of the target user to collect the resistance change signal corresponding to the deformation of the skin surface of the chest, abdomen and back during the breathing process, and monitor the skin surface deformation related to breathing. The data acquisition card converts the resistance change signal acquired by the flexible strain sensor into a recognizable digital breathing signal; The host computer is equipped with a three-dimensional respiratory quantitative recognition software system; the three-dimensional respiratory quantitative recognition software system is used to receive multi-channel respiratory signals transmitted by the data acquisition card, and to perform real-time preprocessing, baseline drift correction, respiratory cycle segmentation, three-dimensional respiratory motion feature extraction, three-dimensional respiratory pattern recognition, training effect evaluation, and result visualization display on the multi-channel respiratory signals.

2. The three-dimensional quantitative identification and evaluation system for human respiration based on a flexible strain sensor according to claim 1, characterized in that, The three-dimensional respiratory quantitative recognition software system includes a data access module, a signal preprocessing module, a baseline drift correction module, a respiratory cycle recognition module, a three-dimensional respiratory feature extraction module, a three-dimensional respiratory quantitative recognition module, a three-dimensional respiratory evaluation module, and a display and storage module. The data access module is used to receive digital respiratory signals uploaded by the data acquisition card; The signal preprocessing module is used to suppress outliers, reduce noise, and smooth the raw digital respiratory signal to obtain a stable respiratory waveform. The baseline drift correction module is used to eliminate baseline drift caused by various factors during long-term monitoring of the flexible strain sensor. The respiratory cycle recognition module is used to automatically identify the inspiratory start point, inspiratory peak, expiratory start point, expiratory trough, and complete respiratory cycle based on the baseline drift-corrected respiratory waveform. The three-dimensional respiratory feature extraction module is used to extract the respiratory deformation features of the human body in three-dimensional space along the X, Y, and Z directions based on the resistance change signals of multiple flexible strain sensors in the chest, abdomen, and chest-abdomen junction of the target user, and further calculate the weight coefficients of the flexible strain sensors in each part in the X, Y, and Z directions. The three-dimensional respiratory quantitative recognition module further normalizes the comprehensive respiratory deformation features in the X, Y, and Z directions to obtain the weight ratio of the sensors in the X, Y, and Z directions and in each direction; and determines the three-dimensional respiratory state of the target user in the current respiratory cycle based on the weight ratio of the sensors in the X, Y, and Z directions and in each direction. The three-dimensional breathing evaluation module is used to comprehensively evaluate the three-dimensional breathing level of the target user within a monitoring cycle or training cycle based on the single-cycle three-dimensional breathing comprehensive score output by the three-dimensional breathing quantitative identification module. The display and storage module is used to display the resistance changes of each sensor in real time and to store the collected data.

3. The three-dimensional quantitative identification and evaluation system for human respiration based on a flexible strain sensor according to claim 2, characterized in that, 3D respiratory feature extraction module: 1) The flexible strain sensor is disposed at the monitoring sites of the human chest, abdomen, and the connection between the chest and abdomen; according to the pasting direction of the flexible strain sensor and the anatomical site where it is located, corresponding to the three spatial dimensions X, Y, and Z during the human breathing process respectively, the left - right lateral expansion direction of the human body is defined as the X - direction, the front - back expansion direction of the human body is defined as the Y - direction, and the head - foot longitudinal stretching direction of the human body is defined as the Z - direction; 2) During each breathing cycle, the three - dimensional breathing feature extraction module calculates the breathing deformation feature values of each flexible strain sensor respectively; The breathing deformation feature value is the peak - valley difference of the sensor signal, the normalized resistance change amount, or the normalized strain amplitude within this breathing cycle; 3) According to the human body space direction corresponding to the flexible strain sensor, the comprehensive breathing deformation features in the X, Y, and Z directions are calculated respectively by a weighted method; The weight coefficients corresponding to the breathing deformation feature values of each flexible strain sensor in the comprehensive breathing deformation feature in each direction are determined according to the resistance change amount, strain amplitude, and signal stability contribution degree within the same direction of each monitoring site.

4. A three-dimensional quantitative identification and evaluation system for human respiration based on a flexible strain sensor according to claim 2, characterized in that, Three - dimensional breathing quantitative recognition module: Let the weighting ratio of human breathing in the X direction be . A The weighting ratio of breathing in the Y direction is B The weighting ratio of breathing in the Z direction is C ; Will A , B, C As a core quantitative indicator for judging three-dimensional respiratory patterns; among them; A This indicates the contribution of respiratory expansion in the X direction of the human body. B This indicates the contribution of respiratory expansion in the Y direction of the human body. C This indicates the contribution of respiratory expansion in the Z-direction of the human body; Then, when the target user's actual measured three-dimensional respiratory weight ratio A , B, C Compared with standard ratio A 0、 B 0、 C When the deviation of 0 is within the preset allowable range, the system determines that the target user's current breathing cycle meets the requirements of three-dimensional coordinated breathing.

5. A three-dimensional quantitative identification and evaluation system for human respiration based on a flexible strain sensor according to claim 4, characterized in that, The actual measured 3D respiratory weight ratio of the target user A , B, C Compared with standard ratio A 0、 B 0、 C A deviation of 0 is one evaluation criterion. In addition, a three-dimensional respiratory severity scoring system is established to comprehensively assess the three-dimensional respiratory status of the target user's current respiratory cycle. This system includes scores for three-dimensional proportional fit, effective participation, synergy, respiratory rhythm stability, and respiratory amplitude effectiveness. The three-dimensional respiratory quantitative recognition module calculates the above score for each effective respiratory cycle and obtains a single-cycle three-dimensional respiratory comprehensive score. S k : （14） in, S k The three-dimensional respiratory comprehensive score is for the k-th respiratory cycle, with a maximum score of 100 points; S 1k The three-dimensional proportional fit is scored. S 2k Scoring of effective participation in three dimensions. S 3k For three-dimensional synergy scoring, S 4k Assess respiratory rhythm stability. S 5k Scoring the effectiveness of respiratory amplitude; Three-dimensional proportional fit rating: The actual measured A , B, C Compared with the preset standard ratio A 0、 B 0、 C Compare with 0 and assign scores based on the degree of deviation; Three - dimensional direction effective participation degree scoring: Judge whether the comprehensive breathing deformation features in the X, Y, and Z directions reach the corresponding minimum effective thresholds respectively, and assign scores according to the degree of reaching the thresholds and the number of directions; used to evaluate whether effective breathing deformations occur in the X, Y, and Z directions, and avoid the situation where although the proportion is close to the standard value but the overall breathing motion amplitude is too small. Three - dimensional direction coordination scoring: Used to evaluate the synchronization degree of the breathing motions in the X, Y, and Z directions in time; Respiratory rhythm stability scoring: Used to evaluate whether the respiratory rhythm of the target user is stable in multiple consecutive breathing cycles; Respiratory amplitude effectiveness scoring: Used to evaluate whether the breathing deformation amplitude of the target user is within the effective training range.

6. A three-dimensional quantitative identification and evaluation system for human respiration based on a flexible strain sensor according to claim 5, characterized in that, The calculation method of the three - dimensional direction coordination scoring is: The peak times of the respiratory waveforms in the X, Y, and Z directions within the same respiratory cycle are extracted and denoted as follows: t X , t Y , t Z And calculate the maximum phase difference between directions: （15） Let the duration of the current respiratory cycle be... T Then the normalized phase difference is: （16） Assign scores of different levels according to the following criteria: when A value less than 0.10 indicates high synchronization; when 0.10 < <0.30 indicates basic synchronization; when 0.30 < <0.50 indicates mild asynchrony; when 0.50 < <0.90 indicates significant asynchrony; when A value greater than 0.90 indicates severe asynchrony.

7. A three-dimensional quantitative identification and evaluation system for human respiration based on a flexible strain sensor according to claim 5, characterized in that, The calculation method of the respiratory rhythm stability scoring is: Calculate the coefficient of variation of the cycle duration of N consecutive effective breathing cycles: （17） in, σ T The standard deviation of the duration of a continuous respiratory cycle. This represents the average duration of consecutive respiratory cycles; Calculate the coefficient of variation of the deformation amplitude of N consecutive effective breathing cycles: （18） in, σ M The standard deviation of the overall deformation amplitude during a continuous respiratory cycle. This represents the average value of the comprehensive deformation amplitude during a continuous respiratory cycle. Assign scores of different levels according to the following criteria: when CV T <10%, and CV M A percentage <15% indicates a highly stable rhythm; when CV T <15%, and CV M When the percentage is less than 20%, it indicates that the rhythm is basically stable; when CV T <25%, or CV M A percentage less than 30% indicates mild rhythm fluctuations; When the cycle or amplitude fluctuates greatly, it indicates that the rhythm is significantly unstable; When an effective breathing cycle cannot be continuously recognized, it indicates that there is no stable breathing cycle.

8. A three-dimensional quantitative identification and evaluation system for human respiration based on a flexible strain sensor according to claim 5, characterized in that, The calculation method of the respiratory amplitude effectiveness scoring is: Calculate the three - dimensional comprehensive respiratory amplitude of the current breathing cycle: （19） And compare it with the individual baseline respiratory amplitude. M By comparing with 0, the normalized amplitude is obtained: （20） in, M 0 is obtained through resting natural breathing, standard training breathing, or individual calibrated breathing; Assign scores of different levels according to the following criteria: When 0.8 ≤ Q ≤ 1.2, it indicates that the amplitude is appropriate; When 0.6 ≤ Q < 0.8 or 1.2 < Q ≤ 1.5, it indicates that the amplitude is basically appropriate; When 0.4 ≤ Q < 0.6 or 1.5 < Q ≤ 1.8, it indicates that the amplitude is too small or too large; When Q < 0.4 or Q > 1.8, it indicates that the amplitude is significantly abnormal; When an effective breathing deformation cannot be recognized, it indicates that there is no effective amplitude.

9. A three-dimensional quantitative identification and evaluation system for human respiration based on a flexible strain sensor according to claim 2, characterized in that, The specific functions of the three - dimensional breathing evaluation module: Identify N valid respiratory cycles and calculate the three-dimensional respiratory score for each respiratory cycle. S k The system further calculates the average three-dimensional respiratory score during this monitoring period: （21） in, The average three-dimensional respiratory score of the target user during the current monitoring period; Then the three - dimensional breathing pass rate: （22） in, N q The number of breathing cycles required to achieve the preset target score. N This represents the total number of valid respiratory cycles identified within the current monitoring period. The three-dimensional breathing level of the target user is comprehensively evaluated based on the average three-dimensional breathing score and the three-dimensional breathing compliance rate within a monitoring or training cycle.

10. A three-dimensional quantitative identification and evaluation system for human respiration based on a flexible strain sensor according to claim 9, characterized in that, A standard for comprehensively evaluating the three-dimensional respiratory function of a target user within a monitoring or training cycle, based on the average three-dimensional respiratory score and the three-dimensional respiratory target achievement rate: Based on the average three-dimensional respiratory score and three-dimensional respiratory target achievement rate K The target user's three-dimensional breathing level is divided into multiple levels: when When ≥90 and K≥85%, the evaluation result is highly coordinated three-dimensional breathing, and the three-dimensional breathing level is set as Level 1; When 80≤ When <90 and K≥70%, the evaluation result is good three-dimensional breathing, and the three-dimensional breathing level is set as level two; When 70≤ When <80 and K≥50%, the evaluation result is basic three-dimensional breathing, and the three-dimensional breathing level is set to level three; When 60≤ When K < 70 or K < 50%, the evaluation result is three-dimensional respiratory insufficiency, and the three-dimensional respiratory level is set to level four. when When the value is <60, the evaluation result is uncoordinated three-dimensional breathing or obvious compensatory breathing, and the three-dimensional breathing level is set to level five.

Images