CN118001528A - Remote noninvasive respiratory parameter analysis and management system based on Internet - Google Patents

Abstract

The invention discloses an internet-based remote noninvasive respiratory parameter analysis management system, which comprises a parameter monitoring unit, a model design unit and a regulation and control alarm unit, wherein noninvasive respiratory parameters are acquired through the parameter monitoring unit, then a parameter analysis model is established through the model design unit for analysis and treatment, remote monitoring of noninvasive respiratory care of a patient is realized, parameter analysis is carried out from multiple angles of physical state, respiratory intention and respiratory intensity according to physiological characteristics and clinical manifestation of the patient, the current noninvasive respiratory state of the patient is comprehensively judged, the data comprehensiveness is ensured, the optimal personalized respiratory monitoring effect is realized, the suitability of the patient and breathing machine equipment is improved, and the abnormal respiratory fluctuation state is early warned, so that regulation and control management of respiratory equipment parameters and visual display of abnormality reasons are carried out in time, the auxiliary diagnosis decision of medical staff is facilitated, and the data analysis utilization rate is improved.

Description

Remote non-invasive respiratory parameter analysis and management system based on the Internet

Technical Field

The present invention relates to the field of non-invasive breathing technology, and in particular to a remote non-invasive breathing parameter analysis and management system based on the Internet.

Background technique

Non-invasive breathing is a treatment method that uses a mouth-nasal mask or a mask to deliver oxygen or air to the patient's lungs at a certain pressure to improve respiratory function and treat respiratory diseases. Compared with invasive breathing, non-invasive breathing does not require endotracheal intubation or tracheotomy, and uses a gentler way to achieve ventilation support, reducing the risk of infection and complications. Patients are more likely to tolerate non-invasive breathing and can stay awake, eat and communicate verbally under certain conditions. "Internet + non-invasive breathing" is suitable for application scenarios such as chronic respiratory diseases, sleep apnea, pulmonary function rehabilitation, elderly care and remote monitoring of home medical care;

Since the patient's current physical condition, breathing intensity and breathing intention can affect the accuracy of non-invasive breathing monitoring, however, it is difficult to conduct comprehensive analysis of multi-angle parameters in existing remote non-invasive breathing, which makes the ventilator equipment and the patient's personalization and adaptability insufficient, resulting in poor respiratory monitoring results. Under the conditions of remote home care, it may be difficult to detect abnormal breathing in time, affecting the decision-making diagnosis and emergency rescue of medical staff;

In view of the above technical defects, a solution is now proposed.

Summary of the invention

The purpose of the present invention is to solve the problem that the prior art remote non-invasive breathing is difficult to perform comprehensive analysis of multi-angle parameters, resulting in insufficient personalization and adaptability of ventilator equipment to patients, leading to poor respiratory monitoring effects and affecting the decision-making diagnosis and emergency rescue of medical staff. By collecting non-invasive respiratory parameters and analyzing and processing them, parameter analysis is performed from multiple angles such as physical state, respiratory intention and respiratory intensity according to the patient's physiological characteristics and clinical manifestations, ensuring the comprehensiveness of data to achieve the best personalized respiratory monitoring effect, improve the adaptability of patients to ventilator equipment, and preliminarily determine the cause of abnormal breathing through feedback analysis and perform visual display, thereby improving the utilization rate of data analysis and making the auxiliary decision-making highly targeted.

In order to achieve the above object, the present invention adopts the following technical solutions:

The remote non-invasive respiratory parameter analysis and management system based on the Internet includes a parameter monitoring unit, a model design unit and a control alarm unit, wherein the parameter monitoring unit, the model design unit and the control alarm unit are signal-connected;

The parameter monitoring unit is used to collect non-invasive breathing parameters, wherein the non-invasive breathing parameters include body state data, breathing intention data and breathing intensity data;

The model design unit is used to establish a parameter analysis model to analyze and process the non-invasive breathing parameters: by analyzing and processing the body state data, breathing intention data and breathing intensity data, the body state, breathing intention and breathing intensity of the patient are evaluated; then the current non-invasive breathing state of the patient is comprehensively determined, and the corresponding parameter control signal is generated; then the non-invasive breathing state is dynamically monitored, so as to warn of the abnormal breathing fluctuation state, preliminarily determine the abnormal cause, and generate the corresponding alarm prompt signal;

The control alarm unit is used to receive parameter control signals and alarm prompt signals, and perform corresponding processing operations and visual displays.

Furthermore, the process of collecting non-invasive respiratory parameters is as follows:

The physical status data includes the blood oxygen saturation, blood oxygen partial pressure and carbon dioxide partial pressure of arterial blood and venous blood; the blood oxygen saturation of arterial blood is marked as SaO2, and the blood oxygen saturation of venous blood is marked as SvO2; the arterial blood oxygen partial pressure value is marked as PaO2, and the venous blood oxygen partial pressure value is marked as PvO2; the arterial blood carbon dioxide partial pressure value is marked as PaCO2, and the venous blood carbon dioxide partial pressure value is marked as PvCO2;

The breathing intention data includes sound feature information and motion feature information. The sound feature information includes the patient's sound decibel value and sound frequency value. The motion feature information includes the patient's motion amplitude value and motion frequency value.

The respiratory intensity data includes the respiratory frequency, airflow velocity and tidal volume in the breathing mask; the respiratory frequency per unit time is marked as Fh, the airflow velocity per unit time is marked as Vq, and the tidal volume per unit time is marked as Lc;

The parameter collection cycle is set to regularly collect non-invasive respiratory parameters and send them to the model design unit for parameter analysis and processing.

Furthermore, the specific construction process of the parameter analysis model is as follows:

The parameter analysis model includes a physical state assessment sub-model, a breathing intention assessment sub-model, a breathing intensity assessment sub-model and a comprehensive assessment sub-model;

Sa: First, the preliminary analysis of non-invasive respiratory parameters is performed in sequence through the body state assessment sub-model, the breathing intention assessment sub-model, and the breathing intensity assessment sub-model:

Sa-1: The physical state assessment sub-model generates physical state assessment coefficients by analyzing and processing physical state data, thereby assessing the patient's physical state;

Sa-2: The breathing intention assessment submodel generates a breathing intention assessment coefficient by analyzing and processing the breathing intention data, thereby assessing the patient's breathing intention;

Sa-3: The respiratory intensity assessment sub-model generates a respiratory intensity assessment coefficient by analyzing and processing the respiratory intensity data, thereby assessing the patient's respiratory intensity;

Sb: Comprehensive analysis is then performed through the comprehensive evaluation sub-model:

Sb-1: The comprehensive assessment sub-model generates a non-invasive respiratory comprehensive assessment index by combining the physical state assessment coefficient, the respiratory intention assessment coefficient and the respiratory intensity assessment coefficient;

Sb-2: Determine the patient's current non-invasive breathing status and generate corresponding parameter control signals;

Sb-3: Dynamic monitoring of the comprehensive evaluation index of non-invasive breathing is then performed to warn of abnormal fluctuations in non-invasive breathing, preliminarily determine the cause of the abnormality, and generate corresponding alarm prompt signals.

Furthermore, the specific process of analyzing and processing the body status data is as follows:

Sa-101: First, integrate the arterial blood oxygen saturation SaO2, venous blood oxygen saturation SvO2, arterial blood oxygen partial pressure PaO2, venous blood oxygen partial pressure PvO2, arterial blood carbon dioxide partial pressure PaCO2 and venous blood carbon dioxide partial pressure PvCO2 into an indicator set of body status data;

Sa-102: Preprocessing of various indicators of physical status data:

Mark any indicator of the physical state data as a, mark the data value of indicator a as Ja, and set the standard interval of indicator a as Qa[q1, q2];

By combining the data value Ja with the standard interval Qa, the reference value Va of the indicator a is obtained;

Sa-103: Integrate various indicators of the pre-processed body state data into an indicator reference set of the body state data, which includes an arterial blood oxygen saturation reference value VSa, a venous blood oxygen saturation reference value VSv, an arterial blood oxygen partial pressure reference value VPa1, a venous blood oxygen partial pressure reference value VPv1, an arterial blood carbon dioxide partial pressure reference value VPa2, and a venous blood carbon dioxide partial pressure reference value VPv2;

Sa-104: The specific process of generating the body state evaluation coefficient Xst through the index reference set of the body state data is as follows: first, by assigning corresponding weight coefficients to the reference values of arterial blood and venous blood under the same index category, respectively, so as to generate the influence coefficient of the index category, and then comprehensively processing the influence coefficient to generate the body state evaluation coefficient Xst;

Sa-105: Then set the corresponding comparison interval for the physical condition evaluation coefficient Xst to determine the degree of the patient's physical condition.

Furthermore, the specific process of analyzing and processing the breathing intention data is as follows:

Sa-201: First, the patient's voice decibel value, voice frequency value, movement amplitude value and movement frequency value are integrated into an index set of breathing intention data;

Sa-202: Preprocessing of various indicators of breathing intention data:

Mark any index of the breathing intention data as m, mark the dynamic curve of index m as Sm, preset the simulated standard curve of index m as s0, and compare the curve Sm with the simulated standard curve s0:

Extract n0 points from the curve Sm and the simulated standard curve s0 at the same time, mark any point of the curve Sm as p (Xp, Yp), mark the point on the simulated standard curve s0 at the same time as point p as q (Xq, Yq), and then calculate the vertical coordinate difference of n0 groups of corresponding points, so as to obtain the curve difference coefficient Cm, preset the threshold Um of the curve difference coefficient Cm, and judge the state of the indicator m by comparing the threshold value;

Sa-203: Integrate the various indicators of the preprocessed breathing intention data into an indicator reference set of the breathing intention data, the indicator reference set of the breathing intention data includes the sound decibel value difference coefficient Ceb and the sound frequency value difference coefficient Cfb, the movement amplitude value difference coefficient Dwg and the movement frequency value difference coefficient Drg;

Sa-204: The specific process of generating the breathing intention evaluation coefficient Xyt through the indicator reference set of breathing intention data is as follows:

Sa-204-1: Analyze various indicators of sound feature information, and generate a comprehensive sound feature evaluation coefficient Xc by combining the sound decibel value difference coefficient Ceb and the sound frequency value difference coefficient Cfb;

Sa-204-2: Analyze various indicators of motion feature information, and generate the motion feature evaluation coefficient Xd by combining the motion amplitude value difference coefficient Dwg and the motion frequency value difference coefficient Drg;

Sa-204-3: The sound feature evaluation coefficient Xc and the action feature evaluation coefficient Xd are combined to generate the breathing intention action coefficient Xyz;

A standard interval Qy of the respiratory intention action coefficient Xyz is set, and the respiratory intention evaluation coefficient Xyt is obtained by combining the respiratory intention action coefficient Xyz with the standard interval Qy. The current patient's respiratory intention state is determined by interval comparison.

Furthermore, the specific process of analyzing and processing the respiratory intensity data is as follows:

By combining the respiratory frequency Fh, airflow velocity Vq and tidal volume Lc, the respiratory intensity effect coefficient Xqz is generated;

Then set the standard interval Qd of the respiratory intensity action coefficient Xqz, and obtain the respiratory intensity evaluation coefficient Xqd by combining the respiratory intensity action coefficient Xqz with the standard interval Qd;

Then, the current breathing state of the patient is determined by interval comparison.

Furthermore, the specific process of comprehensively determining the patient's current non-invasive breathing status is as follows:

The comprehensive evaluation sub-model generates a non-invasive respiratory comprehensive evaluation index WC by combining the physical state evaluation coefficient Xst, the respiratory intention evaluation coefficient Xyt and the respiratory intensity evaluation coefficient Xqd;

Set the risk interval of the non-invasive respiratory comprehensive assessment index WC, determine the risk level of the patient's current non-invasive respiratory status through interval comparison, and generate parameter control signals of corresponding levels.

Furthermore, the specific process of dynamically monitoring the non-invasive respiratory state is as follows:

Construct a dynamic curve of the non-invasive respiratory comprehensive evaluation index WC, preset the fluctuation range of the non-invasive respiratory comprehensive evaluation index, and when the non-invasive respiratory comprehensive evaluation index WC exceeds the preset fluctuation range, the patient's non-invasive breathing is judged to be abnormal, and an early warning is issued for the abnormal fluctuation state of non-invasive breathing;

Then, through feedback analysis of the body state assessment coefficient Xst, breathing intention assessment coefficient Xyt and breathing intensity assessment coefficient Xqd, the cause of the abnormality can be preliminarily determined, and the corresponding alarm prompt signal can be generated for visual display.

In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

1. The present invention collects non-invasive respiratory parameters through a parameter monitoring unit, and then establishes a parameter analysis model through a model design unit for analysis and processing, thereby evaluating the patient's physical state, respiratory intention and respiratory intensity, and then comprehensively determining the patient's current non-invasive respiratory state and performing dynamic monitoring, generating parameter control signals and alarm prompt signals, and giving early warnings for abnormal respiratory fluctuations, thereby timely regulating and managing respiratory equipment parameters, and preliminarily determining the cause of the abnormality for visual display, which is convenient for medical staff to assist in diagnosis and decision-making;

2. The present invention realizes remote monitoring of non-invasive respiratory care for patients. According to the physiological characteristics and clinical manifestations of patients, parameter analysis is performed from multiple angles such as physical state, respiratory intention and respiratory intensity to ensure the comprehensiveness of data, so as to achieve the best personalized respiratory monitoring effect, improve the compatibility of patients and ventilator equipment, and preliminarily determine the cause of abnormal breathing through feedback analysis and display it visually, thereby improving the utilization rate of data analysis and assisting in decision-making with strong pertinence.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 shows a schematic diagram of an overall module of the present invention;

FIG2 shows a schematic diagram of the overall process of the present invention;

FIG3 shows a schematic diagram of modules of the parameter analysis model of the present invention.

Detailed ways

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

Embodiment 1:

As shown in FIG1-3, the Internet-based remote non-invasive respiratory parameter analysis and management system includes a parameter monitoring unit, a model design unit, and a control alarm unit, wherein the parameter monitoring unit, the model design unit, and the control alarm unit are signal-connected;

The working steps are as follows:

S1: The parameter monitoring unit is used to collect non-invasive breathing parameters, wherein the non-invasive breathing parameters include body state data, breathing intention data and breathing intensity data;

The process of collecting non-invasive respiratory parameters is as follows:

The body status data includes the blood oxygen saturation, blood oxygen partial pressure and carbon dioxide partial pressure of arterial blood and venous blood; the body status data is collected by the oximeter, and the blood oxygen saturation of arterial blood is marked as SaO2, and the blood oxygen saturation of venous blood is marked as SvO2; the arterial blood oxygen partial pressure value is marked as PaO2, and the venous blood oxygen partial pressure value is marked as PvO2; the arterial blood carbon dioxide partial pressure value is marked as PaCO2, and the venous blood carbon dioxide partial pressure value is marked as PvCO2;

The breathing intention data includes sound feature information and motion feature information. The sound feature information includes the patient's voice decibel value and sound frequency value, and the motion feature information includes the patient's motion amplitude value and motion frequency value. The breathing intention data is collected and acquired through sound sensors and video monitoring equipment.

The respiratory intensity data includes the respiratory frequency, airflow velocity and tidal volume in the respiratory mask; the respiratory intensity data is collected through a gas flow sensor and a pressure sensor; the respiratory frequency per unit time is marked as Fh, the airflow velocity per unit time is marked as Vq, and the tidal volume per unit time is marked as Lc;

The parameter collection cycle is set to regularly collect non-invasive respiratory parameters and send them to the model design unit for parameter analysis and processing; it should be noted that the above parameter information is collected and remotely transmitted based on the Internet, and the system users are synchronously measured and compared as a whole through big data technology, and the standard range of each parameter in the program is preset in combination with professional medical knowledge;

S2: The model design unit is used to establish a parameter analysis model to analyze and process non-invasive respiratory parameters. The specific construction process of the parameter analysis model is as follows:

The parameter analysis model includes a physical state assessment sub-model, a breathing intention assessment sub-model, a breathing intensity assessment sub-model and a comprehensive assessment sub-model;

Sa: First, the preliminary analysis of non-invasive respiratory parameters is performed in sequence through the body state assessment sub-model, the breathing intention assessment sub-model, and the breathing intensity assessment sub-model:

Sa-1: The physical state assessment sub-model generates physical state assessment coefficients by analyzing and processing physical state data, thereby assessing the patient's physical state;

The specific process of analyzing and processing body status data is as follows:

Sa-101: First, integrate the arterial blood oxygen saturation SaO2, venous blood oxygen saturation SvO2, arterial blood oxygen partial pressure PaO2, venous blood oxygen partial pressure PvO2, arterial blood carbon dioxide partial pressure PaCO2 and venous blood carbon dioxide partial pressure PvCO2 into an indicator set of body status data;

Sa-102: Preprocessing of various indicators of physical status data:

Mark any indicator of the physical state data as a, mark the data value of indicator a as Ja, and set the standard interval of indicator a as Qa[q1, q2];

By combining the data value Ja with the standard interval Qa, the reference value Va of indicator a is obtained:

Va＝(Ja-q1)*(q2-Ja)+1

When the data value Ja of indicator a is within the standard interval Qa, the reference value Va is greater than 1, and the indicator data is judged to be normal; conversely, when the data value Ja of indicator a is higher or lower than the standard interval Qa, the reference value Va is less than 1, and the indicator data is judged to be abnormal;

Sa-103: Integrate various indicators of the pre-processed body state data into an indicator reference set of the body state data, which includes an arterial blood oxygen saturation reference value VSa, a venous blood oxygen saturation reference value VSv, an arterial blood oxygen partial pressure reference value VPa1, a venous blood oxygen partial pressure reference value VPv1, an arterial blood carbon dioxide partial pressure reference value VPa2, and a venous blood carbon dioxide partial pressure reference value VPv2;

Sa-104: The specific process of generating the body state evaluation coefficient Xst through the index reference set of the body state data is as follows: first, the reference values of arterial blood and venous blood under the same index category are assigned corresponding weight coefficients respectively, thereby generating the influence coefficient of the index category, and then the influence coefficient is comprehensively processed to generate the body state evaluation coefficient Xst. The specific process is as follows:

Sa-104-1: Generate the blood oxygen saturation influence coefficient X1 by combining the arterial blood oxygen saturation reference value VSa with the venous blood oxygen saturation reference value VSv: X1 = α1*VSa + α2*VSv;

Among them, α1 and α2 are weight coefficients of the arterial blood oxygen saturation reference value VSa and the venous blood oxygen saturation reference value VSv, respectively, and α1 and α2 are both greater than 0; the weight coefficients are preset and obtained through a large amount of data measurement. When the arterial blood oxygen saturation reference value VSa and the venous blood oxygen saturation reference value VSv are higher, the blood oxygen saturation influence coefficient X1 is higher, indicating that the patient's blood oxygen saturation state is better, thereby evaluating the patient's pulmonary ventilation and metabolic status.

Sa-104-2: Generate the blood oxygen partial pressure influence coefficient X2 by combining the arterial blood oxygen partial pressure reference value VPa1 and the venous blood oxygen partial pressure reference value VPv1: X2 = α3*VPa1+α4*VPv1;

Among them, α3 and α4 are weight coefficients of the reference value of arterial oxygen partial pressure VPa1 and the reference value of venous oxygen partial pressure VPv1, respectively, and both α3 and α4 are greater than 0; when the reference value of arterial oxygen partial pressure VPa1 and the reference value of venous oxygen partial pressure VPv1 are higher, the blood oxygen partial pressure influence coefficient X2 is higher, indicating that the patient's blood oxygen partial pressure state is better, thereby evaluating the patient's pulmonary ventilation and metabolism state better;

Sa-104-3: Generate the influence coefficient of blood carbon dioxide partial pressure X3 by combining the reference value of arterial blood carbon dioxide partial pressure VPa2 and the reference value of venous blood carbon dioxide partial pressure VPv2: X3 = α5*VPa2+α6*VPv2;

Among them, α5 and α6 are weight coefficients of the reference value VPa2 of the partial pressure of carbon dioxide in arterial blood and the reference value VPv2 of the partial pressure of carbon dioxide in venous blood, respectively, and both α5 and α6 are greater than 0; when the reference value VPa2 of the partial pressure of carbon dioxide in arterial blood and the reference value VPv2 of the partial pressure of carbon dioxide in venous blood are higher, the influence coefficient X3 of the partial pressure of carbon dioxide in blood is higher, indicating that the patient's partial pressure of carbon dioxide in blood is better, thereby evaluating the patient's pulmonary ventilation and metabolism status better;

Sa-104-4: The body condition assessment coefficient Xst is generated by combining the blood oxygen saturation influence coefficient X1, the blood oxygen partial pressure influence coefficient X2 and the blood carbon dioxide partial pressure influence coefficient X3:

Xst＝μ1*X1+μ2*X2+μ3*X3

Among them, μ1, μ2 and μ3 are weight factors of blood oxygen saturation influence coefficient X1, blood oxygen partial pressure influence coefficient X2 and blood carbon dioxide partial pressure influence coefficient X3, respectively, and μ1, μ2 and μ3 are all greater than 0; when the blood oxygen saturation influence coefficient X1, blood oxygen partial pressure influence coefficient X2 and blood carbon dioxide partial pressure influence coefficient X3 are higher, the physical state assessment coefficient Xst is higher, the comprehensive judgment and assessment of the patient's pulmonary ventilation and metabolism status is better, indicating that the patient's physical state is better;

Sa-105: Then set a corresponding comparison interval for the physical state evaluation coefficient Xst, so as to determine the degree of the patient's physical state. For example, a first-level interval H1, a second-level interval H2 and a third-level interval H3 are set. When the physical state evaluation coefficient Xst is in the first-level interval H1, the degree of the patient's physical state is determined to be excellent; when the physical state evaluation coefficient Xst is in the second-level interval H2, the degree of the patient's physical state is determined to be average; when the physical state evaluation coefficient Xst is in the third-level interval H3, the degree of the patient's physical state is determined to be poor;

From the above, the physical state assessment sub-model monitors the blood oxygen saturation and carbon dioxide level, assesses the patient's lung ventilation and metabolism status, generates a physical state assessment coefficient, and thus assesses the patient's physical state;

Sa-2: The breathing intention assessment sub-model generates a breathing intention assessment coefficient by analyzing and processing the breathing intention data, thereby assessing the patient's breathing intention;

The specific process of analyzing and processing breathing intention data is as follows:

Sa-201: First, the patient's voice decibel value, voice frequency value, movement amplitude value and movement frequency value are integrated into an index set of breathing intention data;

Sa-202: Preprocessing of various indicators of breathing intention data:

Mark any index of the breathing intention data as m, mark the dynamic curve of index m as Sm, preset the simulated standard curve of index m as s0, and compare the curve Sm with the simulated standard curve s0:

Extract n0 points from the curve Sm and the simulated standard curve s0 at the same time, mark any point of the curve Sm as p (Xp, Yp), mark the point on the simulated standard curve s0 at the same time as point p as q (Xq, Yq), and then calculate the vertical coordinate difference of n0 groups of corresponding points to obtain the curve difference coefficient Cm:

A threshold value Um of the curve difference coefficient Cm is preset. When the curve difference coefficient Cm exceeds the threshold value Um, the indicator m is judged to be in an abnormal state, indicating that the breathing state is not stable, and the breathing intention is higher at this time; otherwise, the indicator m is judged to be in a normal state, indicating that the breathing state is stable, and the breathing intention is normal at this time;

Sa-203: Integrate various indicators of the preprocessed breathing intention data into an indicator reference set of the breathing intention data, wherein the indicator reference set of the breathing intention data includes a sound decibel value difference coefficient Ceb, a sound frequency value difference coefficient Cfb, a motion amplitude value difference coefficient Dwg, and a motion frequency value difference coefficient Drg;

Sa-204: The specific process of generating the breathing intention evaluation coefficient Xyt through the index reference set of the breathing intention data is as follows: firstly, data analysis is performed on each index of the sound feature information and the motion feature information to obtain the sound feature evaluation coefficient Xc and the motion feature evaluation coefficient Xd, and then the evaluation coefficients are comprehensively processed to generate the breathing intention evaluation coefficient Xyt. The specific process is as follows:

Sa-204-1: The process of analyzing each indicator of sound characteristic information is as follows:

Acquire sound feature information from the patient's chest, respiratory tract, and throat, mark any respiratory sound position as b, mark the sound decibel value at position b as Eb, and the sound frequency value as Fb, first fit and construct a dynamic change graph of the sound feature information, and identify and mark the patient's abnormal sound fragments through big data analysis and sound signal comparison technology;

Among them, the abnormal sound clips of the patient are identified and marked by recording them with a recording device, and the sound signals are analyzed by an existing sound spectrum analyzer to obtain the sound decibel value Eb and the sound frequency value Fb;

The sound decibel value Eb and the sound frequency value Fb are integrated as the indicators of the sound characteristics, and the curve change diagram of each indicator of the sound characteristics is constructed. By preprocessing the sound decibel value Eb and the sound frequency value Fb and performing curve analysis, the difference coefficients of the sound decibel value Eb and the sound frequency value Fb are obtained in turn, and the sound decibel value difference coefficient is marked as Ceb, and the sound frequency value difference coefficient is marked as Cfb;

The sound characteristic evaluation coefficient Xc is generated by combining the sound decibel value difference coefficient Ceb and the sound frequency value difference coefficient Cfb: Xc = λ1*Ceb+λ2*Cfb;

Among them, λ1 and λ2 are weight coefficients of the sound decibel value difference coefficient Ceb and the sound frequency value difference coefficient Cfb, respectively, and both λ1 and λ2 are greater than 0; when the sound decibel value difference coefficient Ceb and the sound frequency value difference coefficient Cfb are higher, the sound feature evaluation coefficient Xc is higher, indicating that the patient's breathing intention is higher;

Sa-204-2: The process of analyzing each indicator of motion feature information is as follows:

The sound feature information of the chest, respiratory tract and limbs is collected while the patient is sleeping. Any movement position is marked as g, the movement amplitude value of position g is marked as Wg, and the movement frequency value is marked as Rg. The dynamic change graph of the movement feature information is first fitted and constructed, and the abnormal movement fragments of the patient are identified and marked through big data analysis and movement signal comparison technology.

Among them, identifying and marking abnormal action clips of patients is to use existing visual technology, use cameras or depth cameras and other equipment to analyze and process human body images or depth images, extract human body key points or bone information, thereby realizing the detection and tracking of limb movements to obtain the action amplitude value Wg and action frequency value Rg;

The action amplitude value Wg and the action frequency value Rg are integrated as indicators of the action characteristics, and the curve change diagram of each indicator of the action characteristics is constructed. By preprocessing and curve analysis of the action amplitude value Wg and the action frequency value Rg, the difference coefficients of the action amplitude value Wg and the action frequency value Rg are obtained in turn, and the difference coefficients of the action amplitude value are marked as Dwg, and the difference coefficients of the action frequency value are marked as Drg;

By combining the action amplitude value difference coefficient Dwg and the action frequency value difference coefficient Drg, the action feature evaluation coefficient Xd is generated: Xd = λ3*Dwg+λ4*Drg;

Among them, λ3 and λ4 are weight coefficients of the action amplitude value difference coefficient Dwg and the action frequency value difference coefficient Drg, respectively, and both λ3 and λ4 are greater than 0; when the action amplitude value difference coefficient Dwg and the action frequency value difference coefficient Drg are higher, the action characteristic evaluation coefficient Xd is higher, indicating that the patient's breathing intention is higher;

Sa-204-3: Then, the sound feature evaluation coefficient Xc and the action feature evaluation coefficient Xd are combined to generate the breathing intention effect coefficient Xyz: Xyz = υ1*Xc+υ2*Xd;

Among them, υ1 and υ2 are weight factors of the sound feature evaluation coefficient Xc and the motion feature evaluation coefficient Xd, respectively, and both υ1 and υ2 are greater than 0; when the sound feature evaluation coefficient Xc and the motion feature evaluation coefficient Xd are higher, the respiratory intention effect coefficient Xyz is higher, and the comprehensive judgment and evaluation of the patient's respiratory intention is stronger, indicating that the current patient's non-invasive respiratory state is more unstable and the patient's respiratory demand is higher;

A standard interval Qy of the breathing intention effect coefficient Xyz is set, and the breathing intention effect coefficient Xyz is combined with the standard interval Qy, and the interval range of the preset standard interval Qy is [y1, y2], and the breathing intention evaluation coefficient Xyt is obtained: Xyt = (Xyz-y1)*(y2-Xyz);

Determine the patient's current breathing intention state by comparing the intervals:

When the respiratory intention action coefficient Xyz is in the standard interval Qy, the respiratory intention evaluation coefficient Xyt is greater than 0, and the patient's respiratory intention is judged to be in a normal and stable state; conversely, when the respiratory intention action coefficient Xyz is higher or lower than the standard interval Qy, the respiratory intention evaluation coefficient Xyt is less than 0, wherein: when the respiratory intention action coefficient Xyz is lower than the standard interval Qy, the patient's respiratory intention is judged to be in a weak state; when the respiratory intention action coefficient Xyz is higher than the standard interval Qy, the patient's respiratory intention is judged to be in a strong state; both weak and strong respiratory intentions can indicate that the patient's current non-invasive breathing situation is critical and the state is dangerous;

From the above, the breathing intention assessment sub-model generates a breathing intention assessment coefficient by analyzing the sound features and action features, thereby assessing the patient's breathing intention;

Sa-3: The respiratory intensity assessment sub-model generates a respiratory intensity assessment coefficient by analyzing and processing the respiratory intensity data, thereby assessing the patient's respiratory intensity;

The specific process of analyzing and processing respiratory intensity data is as follows:

By combining the respiratory frequency Fh, airflow velocity Vq and tidal volume Lc, the respiratory intensity effect coefficient Xqz is generated: Xqz = ε1*Fh+ε2*Vq+ε3*Lc;

Among them, ε1, ε2 and ε3 are weight factors of respiratory frequency Fh, airflow velocity Vq and tidal volume Lc, respectively, and ε1, ε2 and ε3 are all greater than 0; when the respiratory frequency Fh, airflow velocity Vq and tidal volume Lc are higher, it means that the patient's breathing intensity is higher;

The standard interval Qd of the respiratory intensity action coefficient Xqz is set. By combining the respiratory intensity action coefficient Xqz with the standard interval Qd, the interval range of the preset standard interval Qd is [d1, d2], and the respiratory intensity evaluation coefficient Xqd is obtained: Xqd = (Xqz-d1)*(d2-Xqz);

Determine the patient's current respiratory status by comparing the intervals:

When the respiratory intensity action coefficient Xqz is in the standard interval Qd, the respiratory intensity assessment coefficient Xqd is greater than 0, and it is determined that the patient's respiratory intensity is in a stable state. The higher the respiratory intensity action coefficient Xqz is, the better the patient's non-invasive respiratory state is. On the contrary, when the respiratory intensity action coefficient Xqz is higher or lower than the standard interval Qd, the respiratory intensity assessment coefficient Xqd is less than 0, wherein: when the respiratory intensity action coefficient Xqz is lower than the standard interval Qd, it is determined that the patient's respiratory intensity is in a weak state; when the respiratory intensity action coefficient Xqz is higher than the standard interval Qd, it is determined that the patient's respiratory intensity is in a rapid state.

From the above, the respiratory intensity assessment sub-model evaluates whether the patient's respiratory intensity is stable by combining the respiratory frequency, gas flow rate and tidal volume in the respiratory mask, generates a respiratory intensity assessment coefficient, and thus evaluates whether the patient's respiratory intensity state is stable;

Sb: Comprehensive analysis is then performed through the comprehensive evaluation sub-model:

Sb-1: The comprehensive assessment sub-model generates a non-invasive respiratory comprehensive assessment index by combining the physical state assessment coefficient, the respiratory intention assessment coefficient and the respiratory intensity assessment coefficient;

The specific process of comprehensively determining the patient's current non-invasive breathing status is as follows:

The comprehensive evaluation sub-model generates a non-invasive respiratory comprehensive evaluation index WC by combining the body state evaluation coefficient Xst, the respiratory intention evaluation coefficient Xyt, and the respiratory intensity evaluation coefficient Xqd:

WC＝ ^Xstω1 + ^Xytω2 + ^Xqdω3

Among them, ω1, ω2 and ω3 are weight factors of the physical state assessment coefficient Xst, the respiratory intention assessment coefficient Xyt and the respiratory intensity assessment coefficient Xqd, respectively, and ω1, ω2 and ω3 are all greater than 0; when the physical state assessment coefficient Xst, the respiratory intention assessment coefficient Xyt and the respiratory intensity assessment coefficient Xqd are higher, the non-invasive respiratory comprehensive assessment index WC is higher, indicating that the patient's non-invasive respiratory state is more stable and the physical signs are better;

Sb-2: Determine the patient's current non-invasive breathing status and generate corresponding parameter control signals;

Set the risk interval of the non-invasive respiratory comprehensive assessment index WC, determine the risk level of the patient's current non-invasive respiratory status through interval comparison, and generate a parameter control signal of the corresponding level;

Sb-3: Dynamic monitoring of the comprehensive evaluation index of non-invasive breathing is then performed to warn of abnormal fluctuations in non-invasive breathing, preliminarily determine the cause of the abnormality, and generate corresponding alarm prompt signals;

The specific process of dynamic monitoring of non-invasive respiratory status is:

By dynamically monitoring the non-invasive respiratory comprehensive evaluation index WC, the fluctuation range of the non-invasive respiratory comprehensive evaluation index is preset. When the non-invasive respiratory comprehensive evaluation index WC exceeds the preset fluctuation range, the patient's non-invasive breathing is judged to be abnormal, and an early warning is issued for the abnormal fluctuation state of non-invasive breathing;

Then, through feedback analysis of the body state evaluation coefficient Xst, the breathing intention evaluation coefficient Xyt and the breathing intensity evaluation coefficient Xqd, the cause of the abnormality can be preliminarily determined, and the corresponding alarm prompt signal can be generated for visual display:

When the comparison determines that the patient's physical condition is poor, a first alarm prompt signal is generated; when the comparison determines that the patient's breathing intention is weak or strong, a second alarm prompt signal is generated; when the comparison determines that the patient's breathing intensity is weak or rapid, a third alarm prompt signal is generated;

Integrate the first alarm prompt signal, the second alarm prompt signal and the third alarm prompt signal into an alarm prompt signal group, and output the group to the control alarm unit;

S3: The control alarm unit is used to receive the parameter control signal and the alarm prompt signal, and perform corresponding processing operations and visual displays;

By receiving parameter control signals, the parameters of the ventilator equipment are controlled within a preset range, such as pressurizing the oxygen supply, adjusting the inspiratory pressure (IPAP) to adjust the lung expansion and air inflow when the patient inhales, and maintaining the patency of the expiratory airway with the expiratory pressure (EPAP) to avoid collapse at the end of exhalation, as well as adjusting the temperature and humidity in the respiratory gas to ensure patient comfort and airway moisturizing.

By receiving the alarm prompt signal, the corresponding abnormal reason text is visualized and the alarm sound is timely alarmed, wherein, when the first alarm prompt signal is received, the text "the patient's physical condition is poor" is edited and displayed; when the second alarm prompt signal is received, the text "the patient's breathing intention is abnormal" is edited and displayed; when the third alarm prompt signal is received, the text "the patient's breathing intensity is abnormal" is edited and displayed; thereby prompting medical staff to initiate corresponding emergency rescue according to the corresponding preliminary abnormal reason;

It should be noted that if a patient experiences sudden respiratory abnormalities, in addition to the above-mentioned treatment operations, non-invasive respiratory treatment should be stopped immediately under the premise of safety, and a preliminary assessment of the cause of the respiratory abnormality should be conducted in a timely manner for medical staff to refer to for diagnosis. At the same time, medical staff should take appropriate first aid measures and switch to invasive ventilators in a timely manner when necessary to ensure that the patient's breathing is unobstructed and adequate oxygen supply.

In summary, the present invention collects non-invasive respiratory parameters through a parameter monitoring unit, and then establishes a parameter analysis model through a model design unit for analysis and processing, thereby evaluating the patient's physical state, respiratory intention and respiratory intensity, and then comprehensively determining the patient's current non-invasive respiratory state and performing dynamic monitoring, generating parameter control signals and alarm prompt signals, and giving early warnings for abnormal respiratory fluctuations, thereby timely regulating and managing respiratory equipment parameters, and preliminarily determining the cause of the abnormality for visual display, which is convenient for medical staff to assist in diagnosis and decision-making;

The present invention realizes remote monitoring of non-invasive respiratory care for patients. According to the physiological characteristics and clinical manifestations of patients, parameter analysis is performed from multiple angles such as physical state, respiratory intention and respiratory intensity to ensure the comprehensiveness of data, so as to achieve the best personalized respiratory monitoring effect and improve the compatibility of patients and ventilator equipment. The cause of abnormal breathing is preliminarily determined through feedback analysis and visualized, which improves the utilization rate of data analysis and has strong targeted auxiliary decision-making.

The size of the interval and threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by technical personnel in this field for each group of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.

The above formulas are all dimensionless and numerical calculations. The formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims (8)

1. Remote noninvasive breathing parameter analysis and management system based on Internet is characterized in that: the system comprises a parameter monitoring unit, a model design unit and a regulation alarm unit, wherein the parameter monitoring unit, the model design unit and the regulation alarm unit are connected through signals;

The parameter monitoring unit is used for acquiring noninvasive breathing parameters, wherein the noninvasive breathing parameters comprise physical state data, breathing intention data and breathing intensity data;

The model design unit is used for establishing a parameter analysis model to analyze and process noninvasive respiratory parameters: assessing the physical state, respiration intent, and respiration intensity of the patient by analyzing and processing the physical state data, respiration intent data, and respiration intensity data; further comprehensively judging the current noninvasive respiratory state of the patient and generating corresponding parameter regulation signals; dynamically monitoring the noninvasive respiratory state, thereby carrying out early warning on the respiratory abnormality fluctuation state, preliminarily judging the abnormality cause and generating a corresponding alarm prompt signal;

the regulation and control alarm unit is used for receiving the parameter regulation and control signal and the alarm prompt signal, and carrying out corresponding processing operation and visual display.

2. The internet-based remote noninvasive respiratory parameter analysis and management system of claim 1, wherein: the collection process of the noninvasive respiratory parameters is as follows:

The physical state data comprise blood oxygen saturation, blood oxygen partial pressure and carbon dioxide partial pressure of arterial blood and venous blood; marking the blood oxygen saturation of arterial blood as SaO2 and the blood oxygen saturation of venous blood as SvO2; marking the arterial blood oxygen partial pressure value as PaO2 and the venous blood oxygen partial pressure value as PvO2; labeling the arterial blood carbon dioxide partial pressure value as PaCO2, and labeling the venous blood carbon dioxide partial pressure value as PvCO2;

The breathing intention data comprises sound characteristic information and action characteristic information, the sound characteristic information comprises sound decibel values and sound frequency values of a patient, and the action characteristic information comprises action amplitude values and action frequency values of the patient;

The respiratory intensity data includes respiratory rate, airflow rate, and tidal volume within the respiratory mask; the breathing frequency in unit time is marked as Fh, the airflow speed in unit time is marked as Vq, and the tidal volume in unit time is marked as Lc;

Setting a parameter acquisition period to acquire noninvasive respiratory parameters regularly, and sending the noninvasive respiratory parameters to a model design unit for parameter analysis and processing.

3. The internet-based remote noninvasive respiratory parameter analysis and management system of claim 2, wherein: the specific construction process of the parameter analysis model comprises the following steps:

The parameter analysis model comprises a physical state evaluation sub-model, a respiration intention evaluation sub-model, a respiration intensity evaluation sub-model and a comprehensive evaluation sub-model;

Sa: firstly, sequentially performing preliminary analysis of noninvasive respiratory parameters through a physical state evaluation sub-model, a respiratory intention evaluation sub-model and a respiratory strength evaluation sub-model:

sa-1: the physical state evaluation sub-model generates physical state evaluation coefficients by analyzing and processing physical state data, thereby evaluating the physical state of the patient;

sa-2: the respiratory intention evaluation sub-model generates respiratory intention evaluation coefficients by analyzing and processing respiratory intention data, thereby evaluating respiratory intention of a patient;

sa-3: the breath intensity evaluation sub-model generates a breath intensity evaluation coefficient by analyzing and processing breath intensity data, thereby evaluating the breath intensity of the patient;

sb: and then carrying out comprehensive analysis by the comprehensive evaluation submodel:

sb-1: the comprehensive evaluation sub-model generates a noninvasive respiratory comprehensive evaluation index by combining the physical state evaluation coefficient, the respiratory intention evaluation coefficient and the respiratory intensity evaluation coefficient;

sb-2: judging the current noninvasive respiratory state of the patient, and generating corresponding parameter regulation signals;

sb-3: and then dynamically monitoring the comprehensive evaluation index of the noninvasive breath, pre-warning the abnormal fluctuation state of the noninvasive breath, and primarily judging the cause of the abnormality to generate a corresponding alarm prompt signal.

4. The internet-based remote noninvasive respiratory parameter analysis and management system of claim 3, wherein: the specific process of analyzing and processing the physical state data is as follows:

Sa-101: firstly integrating the blood oxygen saturation SaO2 of arterial blood, the blood oxygen saturation SvO2 of venous blood, the arterial blood oxygen partial pressure value PaO2, the venous blood oxygen partial pressure value PvO2, the arterial blood carbon dioxide partial pressure value PaCO2 and the venous blood carbon dioxide partial pressure value PvCO2 into an index number set of physical state data;

Sa-102: preprocessing each index of the physical state data:

marking any index of the physical state data as a, marking the data value of the index a as Ja, and setting a standard interval Qa [ q1, q2] of the index a;

The data value Ja is combined with the standard interval Qa to obtain a reference value Va of the index a;

Sa-103: integrating each index of the preprocessed body state data into an index reference set of the body state data, wherein the index reference set of the body state data comprises an arterial blood oxygen saturation reference value VSa, a venous blood oxygen saturation reference value VSv, an arterial blood oxygen partial pressure reference value VPa1, a venous blood oxygen partial pressure reference value VPv1, an arterial blood carbon dioxide partial pressure reference value VPa2 and a venous blood carbon dioxide partial pressure reference value VPv2;

Sa-104: the specific process of generating the physical state evaluation coefficient Xst through the index reference set of the physical state data is as follows: firstly, respectively giving corresponding weight coefficients to the reference values of arterial blood and venous blood under the same index category, so as to generate an influence coefficient of the index category, and then comprehensively processing the influence coefficient, so as to generate a physical state evaluation coefficient Xst;

Sa-105: and setting a corresponding comparison interval for the physical state evaluation coefficient Xst so as to judge the physical state degree of the patient.

5. The internet-based remote noninvasive respiratory parameter analysis and management system of claim 4, wherein: the specific process of analyzing and processing the respiratory intention data is as follows:

sa-201: firstly integrating a sound decibel value, a sound frequency value, a motion amplitude value and a motion frequency value of a patient into an index number set of breathing intention data;

Sa-202: preprocessing each index of respiratory intention data:

marking any index of breathing intention data as m, marking a dynamic curve of the index m as Sm, presetting a simulation standard curve of the index m as s0, and comparing the curve Sm with the simulation standard curve s 0:

Extracting n0 points from the curve Sm and the simulation standard curve s0 according to the same time, marking any point of the curve Sm as p (Xp, yp), marking the point of the simulation standard curve s0 at the same time as the point p as q (Xq, yq), further measuring and calculating the vertical coordinate difference value of n0 groups of corresponding points, thereby obtaining a curve difference coefficient Cm, presetting a threshold Um of the curve difference coefficient Cm, and comparing the state of a judgment index m through the threshold;

Sa-203: integrating each index of the preprocessed breath intention data into an index reference set of the breath intention data, wherein the index reference set of the breath intention data comprises a sound decibel value difference coefficient Ceb, a sound audio frequency value difference coefficient Cfb, an action amplitude value difference coefficient Dwg and an action frequency value difference coefficient Drg;

sa-204: the specific process of generating the respiratory intent assessment coefficients Xyt from the index reference set of respiratory intent data is as follows:

Sa-204-1: analyzing each index of the sound characteristic information, and comprehensively generating a sound characteristic evaluation coefficient Xc by combining the sound decibel value difference coefficient Ceb and the sound audio frequency value difference coefficient Cfb;

Sa-204-2: analyzing each index of the motion characteristic information, and comprehensively generating a motion characteristic evaluation coefficient Xd by combining the motion amplitude value difference coefficient Dwg and the motion frequency value difference coefficient Drg;

sa-204-3: then, the sound characteristic evaluation coefficient Xc and the action characteristic evaluation coefficient Xd are combined to generate a respiration intention action coefficient Xyz;

Setting a standard interval Qy of the respiration intention action coefficient Xyz, combining the respiration intention action coefficient Xyz with the standard interval Qy to obtain a respiration intention evaluation coefficient Xyt, and judging the respiration intention state of the current patient through interval comparison.

6. The internet-based remote noninvasive respiratory parameter analysis and management system of claim 5, wherein: the specific process of analyzing and processing the breath intensity data is as follows:

Generating a breath intensity action coefficient Xqz by combining the breath frequency Fh, the airflow speed Vq and the tidal volume Lc;

and setting a standard interval Qd of the breath intensity action coefficient Xqz, combining the breath intensity action coefficient Xqz with the standard interval Qd to obtain a breath intensity evaluation coefficient Xqd, and judging the breath state of the current patient through interval comparison.

7. The internet-based remote noninvasive respiratory parameter analysis and management system of claim 6, wherein: the specific process for comprehensively judging the current noninvasive respiratory state of the patient comprises the following steps:

The comprehensive evaluation submodel generates a noninvasive respiratory comprehensive evaluation index WC by combining the physical state evaluation coefficient Xst, the respiratory intention evaluation coefficient Xyt and the respiratory intensity evaluation coefficient Xqd;

Setting a risk interval of a noninvasive respiration comprehensive evaluation index WC, comparing and judging the risk degree of the current noninvasive respiration state of the patient through the interval, and generating a parameter regulation and control signal of a corresponding level.

8. The internet-based remote noninvasive respiratory parameter analysis and management system of claim 7, wherein: the specific process of dynamically monitoring the noninvasive respiratory state is as follows:

constructing a dynamic curve of a noninvasive respiration comprehensive evaluation index WC, presetting a fluctuation interval of the noninvasive respiration comprehensive evaluation index, judging that the noninvasive respiration of a patient is abnormal when the noninvasive respiration comprehensive evaluation index WC exceeds the preset fluctuation interval, and pre-warning the abnormal fluctuation state of the noninvasive respiration;

And then, the physical state evaluation coefficient Xst, the respiration intention evaluation coefficient Xyt and the respiration intensity evaluation coefficient Xqd are analyzed through feedback, so that the abnormal cause is primarily judged, and a corresponding alarm prompt signal is generated for visual display.