CN117503078A

CN117503078A - System and method for detecting wound stress response

Info

Publication number: CN117503078A
Application number: CN202311484566.1A
Authority: CN
Inventors: 左云霞; 王其锋; 赵雨意
Original assignee: West China Hospital of Sichuan University
Current assignee: West China Hospital of Sichuan University
Priority date: 2023-11-09
Filing date: 2023-11-09
Publication date: 2024-02-06

Abstract

The present description provides a system and method for detecting a wound stress response, the system comprising: a sample detection module configured to determine an abundance of a biomarker in a sample of the subject, wherein the biomarker comprises a Volatile Organic Compound (VOC) exhaled by the subject; and a processor configured to predict wound stress response information of the subject based on the abundance of the biomarker.

Description

System and method for detecting wound stress response

Technical Field

The present specification relates to the field of stress response detection, and in particular to a system and method for detecting a traumatic stress response.

Background

Diagnosis and treatment of pain relies on monitoring of nociceptive stimuli. For example, the extent of nociceptive stimulation during anesthesia surgery is associated with a variety of post-operative adverse complications. Therefore, by monitoring the nociceptive stimulus stress response during the anesthesia operation, the anesthesiologist can be assisted to know the traumatic stress response level of the patient in real time, and guide the anesthesiologist to perform anti-inflammatory, anti-stress and analgesic related operations, so as to reduce the occurrence rate of postoperative complications (such as postoperative cognitive dysfunction, postoperative delirium, postoperative infection, etc.).

For this reason, various monitoring techniques have been developed to monitor the extent of nociceptive stimulation during surgery. For example, the patient's traumatic stress response level is obtained by monitoring hemodynamic parameters (such as blood pressure and heart rate) or autonomic nervous system changes, and brain electrical changes. However, these monitoring means tend to be relatively complex to operate and are not acceptable to clinicians and patients.

It is therefore desirable to provide a method and system that is capable of more conveniently and sensitively detecting a wound stress response.

Disclosure of Invention

One or more embodiments of the present specification provide a system for detecting a wound stress response, the system comprising: a sample detection module configured to determine an abundance of a biomarker in a sample of a subject, wherein the biomarker comprises a Volatile Organic Compound (VOC) exhaled by the subject; and a processor configured to predict traumatic stress response information for the subject based on the abundance of the biomarker.

One of the embodiments of the present specification provides a method of detecting a wound stress response implemented on a computing device having at least one processor and at least one storage device, the method comprising: determining the abundance of a biomarker in a sample of a subject, wherein the biomarker comprises a Volatile Organic Compound (VOC) exhaled by the subject; based on the abundance of the biomarker, wound stress response information of the subject is predicted.

One or more embodiments of the present specification provide a non-transitory computer-readable medium comprising at least one set of instructions, wherein the at least one set of instructions, when executed by at least one processor of a computing device, cause the computing device to perform a method comprising: determining the abundance of a biomarker in a sample of a subject, wherein the biomarker comprises a Volatile Organic Compound (VOC) exhaled by the subject; based on the abundance of the biomarker, wound stress response information of the subject is predicted.

Drawings

FIG. 1 is a schematic illustration of an application scenario of a wound stress detection system according to some embodiments of the present disclosure;

FIG. 2 is an exemplary schematic diagram of predicting traumatic stress response information of a subject, according to some embodiments of the present description;

FIG. 3 is an exemplary schematic diagram of a stress model shown in accordance with some embodiments of the present description;

FIG. 4 is an exemplary schematic diagram of a stress model shown in accordance with some embodiments of the present description;

fig. 5 is a graph of ROC curves for prediction of wound stress response from VOC markers according to some embodiments of the present description.

Detailed Description

The drawings that are used in the description of the embodiments will be briefly described below. The drawings do not represent all embodiments.

As used herein, a "system," "apparatus," "unit," and/or "module" is a means for distinguishing between different components, elements, parts, portions, or assemblies at different levels. Other words may be substituted for the words by other expressions if the words achieve the same purpose.

As used in this specification and the claims, the terms "a," "an," "the," and/or "the" are not specific to a singular, but may include a plurality, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.

In the embodiments of the present disclosure, when operations performed by the steps are described, unless otherwise specified, the order of the steps may be changed, the steps may be omitted, and other steps may be included in the operation.

Fig. 1 is a schematic diagram of an application scenario 100 of a wound stress detection system according to some embodiments of the present disclosure.

As shown in fig. 1, an application scenario 100 of a wound stress detection system may include a subject 110, a network 120, a processing device 130, a monitoring system 140, a storage device 150, a terminal 160, and a doctor 170.

The wound stress detection system may be used to monitor a subject's wound stress response. For example, in the clinical operation process, the anesthetized subject cannot express various responses of the subject in the operation, the wound stress response detection system can monitor various parameters of the subject through the monitoring system 140, the processing device 130 analyzes the various parameters, and the analysis result is displayed through the terminal 160, so as to help the doctor judge the physiological condition and the physiological response of the subject.

Subject 110 refers to a subject in need of traumatic stress response monitoring. For example, the subject may include a patient in need of nociceptive stimulation monitoring during an anesthesia procedure. For another example, the subject may include subjects for socioeconomic studies, biological and neuroscience studies, clinical psychological and mental health studies, and stress training.

Network 120 may include any suitable network providing information and/or data capable of facilitating the exchange of information. In some embodiments, information and/or data may be exchanged between one or more components of application scenario 100 via network 120. Network 120 may include a Local Area Network (LAN), wide Area Network (WAN), wired network, wireless network, etc., or any combination thereof. In some embodiments, one or more components in the application scenario 100 (e.g., the processing device 130, the monitoring system 140, the storage device 150, and the terminal 160) may send information and/or data to another component in the application scenario 100 via the network 120. In some embodiments, the network 120 may communicate the characteristic parameter characteristics of the subject monitored by the monitoring system 140 to the storage device 150. In some embodiments, the network 120 may communicate the baseline characteristics of the subject, surgical information, etc. entered into the terminal 160 by the physician 170 to the processing device 130.

Processing device 130 may process data and/or information obtained from other devices or various components of the system to execute program instructions based on such data, information, and/or processing results to perform one or more functions described herein. In some embodiments, processing device 130 may process the retrieval of data and/or information from monitoring system 140, storage device 150. In some embodiments, the processing device 130 may be a single server or a group of servers. In some embodiments, processing device 130 includes a processor 220. The processing device 130 may be local, remote. The processing device 130 may be implemented on a cloud platform.

In some embodiments, the processor in the processing device 130 may predict the subject's traumatic stress response information based on the abundance of the biomarker.

The monitoring system 140 may be used to monitor various status parameters of the subject. For example, the monitoring system 140 may monitor at least one of pulse oximetry, an electrocardiogram, a heart rate, non-invasive arterial blood pressure, end-tidal carbon dioxide partial pressure, and anesthesia depth monitoring indicators of the subject. In some embodiments, the monitoring system 140 can also be used to detect the abundance of VOCs in a subject.

Storage device 150 may be used to store data, instructions, and/or any other information. In some embodiments, storage device 150 may store data and/or information obtained from, for example, subject 110, processing device 130, and the like. For example, the storage device 150 may store instructional information, subject's vital parameter characteristics, baseline characteristics, and the abundance of VOCs, among others. In some embodiments, the storage device 150 may store data and/or instructions that the processing device 130 uses to perform or use to accomplish the exemplary methods described in this specification. In some embodiments, the storage device 150 may include mass storage, removable storage, volatile read-write memory, read-only memory (ROM), and the like, or any combination thereof.

Terminal 160 refers to one or more terminal devices or software used by a user. Wherein the user may be a person associated with a wound stress detection system. In some embodiments, the user using terminal 160 may be one or more doctors. In some embodiments, terminal 160 can be one or any combination of a tablet computer, laptop computer, desktop computer, and the like, other devices having input and/or output capabilities. In some embodiments, the physician 170 may enter the baseline characteristics of the subject 110 via the terminal 160.

Doctor 170 refers to a person who provides medical services to a subject. For example, the doctor may be a person who anesthetizes and perioperatively manages the subject. In some embodiments, doctor 170 may control monitoring system 140 to monitor the vital parameter characteristics of subject 110 by inputting instructions through terminal 160. In some embodiments, doctor 170 may also consider adjusting a control regimen for the wound stress based on the wound stress response information of subject 110 displayed by terminal 160.

It should be noted that the application scenario is provided for illustrative purposes only and is not intended to limit the scope of the present description. Many modifications and variations will be apparent to those of ordinary skill in the art in light of the present description. For example, the application scenario may also include a sample detection module. As another example, application scenarios may be implemented on other devices to implement similar or different functionality. However, variations and modifications do not depart from the scope of the present description.

Fig. 2 is an exemplary schematic diagram 200 of predicting traumatic stress response information of a subject, according to some embodiments of the present description. As shown in fig. 2, in some embodiments, a system for detecting a wound stress response may include a sample detection module 210 and a processor 220.

In some embodiments, the sample detection module is configured to determine an abundance of a biomarker in a sample of the subject, wherein the biomarker comprises a Volatile Organic Compound (VOC) exhaled by the subject.

A sample detection module refers to a module capable of acquiring biomarker related data. For example, the sample detection module may comprise a detection device. In some embodiments, the abundance of a biomarker in a sample of a subject can be detected with a ultraviolet light ionization time-of-flight mass spectrometry (UVI-TOF MS) device. For another example, the sample detection module may acquire biomarker-related data collected by an external device to determine the abundance of the biomarker in the sample of the subject.

The subject sample refers to a sample of gas exhaled by the subject.

Biomarkers refer to indicators that can be objectively measured and evaluated, reflect physiological or pathological processes in a subject, and produce biological effects on exposure or therapeutic intervention. The biomarker may be a molecule, cell, tissue, or physiological parameter within an organism. In some embodiments, the biomarker of wound stress response comprises Volatile Organic Compounds (VOCs) exhaled by the subject.

Volatile Organic Compounds (VOCs) are gaseous organic molecules with high volatility, and human VOCs are derived from many endogenous biochemical processes of the human body, including lipid oxidation, carbohydrate and fatty acid metabolism, and the like. The gas phase metabolites and breakdown products from these processes are transported through the circulatory system and rapidly expelled from the body through the lungs. Since cellular metabolic processes are altered by disease, the change in VOCs can serve as biomarkers for specific pathophysiological conditions.

In some embodiments, VOCs as biomarkers are significantly correlated with wound stress related indicators. Significant correlation refers to a statistically significant relationship between VOC and a wound stress related index, and P <0.05 is considered to be significant. For verification data on the correlation of wound stress related indicators with different VOCs, see the description below.

The significant correlations include significant positive correlations and significant negative correlations. For example, VOCs that are significantly correlated in terms of VOC abundance and wound stress related indicators over a first time and a second time may be determined as biomarkers. In some embodiments, the processor may determine VOCs as biomarkers that are significantly correlated with the VOC abundance and the wound stress related index at a first time and a second time, and that are also significantly correlated at a second time and a third time. Wherein the first time, the second time, and the third time may be different time periods associated with the procedure. In some embodiments, the first time may be a period of time before the surgical procedure begins, during which the subject has not yet developed substantial trauma, e.g., has not yet skived; the second time may be a period of time during which the subject has had substantial trauma, e.g., after skin incision; the third time may refer to a post-operative period of time, e.g., after a skin incision suture.

Biomarker abundance refers to the proportion of a biomarker in the volatile organic compounds in the exhaled breath of a subject.

In some embodiments, biomarker abundance may be determined based on mass spectrometry, spectroscopy, or other methods. For more details on determining abundance see below and the description related thereto.

In some embodiments, the processor is configured to predict the subject's traumatic stress response information based on the abundance of the biomarker.

Wound stress response information refers to information that reflects the response of a subject after suffering from a physiological or psychological wound. The traumatic stress response information may include various forms, such as whether the subject is producing a traumatic stress response, the extent of the traumatic stress response, and the like.

In some embodiments, the processor may determine the subject's traumatic stress response information by querying a first preset table based on the abundance of the biomarker. Wherein the first preset table may be constructed based on a history experience of the doctor.

In some embodiments, the processor may predict the subject's traumatic stress response information based on the abundance of the biomarker through a stress model. See the description of fig. 3 and 4 below for an embodiment of a stress model.

In some embodiments, the wound stress response information may be a degree of wound stress response of the subject. For example, the stress model may output a degree of traumatic stress response in the subject as grade 3. The wound stress response information may be presented in various ways. In some embodiments, the wound stress response information predicted by the stress model may be further processed by the processor and displayed on the terminal as a reminder for the healthcare-related personnel. For example, the processor may determine that the degree of the traumatic stress response is greater than a stress threshold, and the degree of the traumatic stress response displayed at the terminal may be displayed in a red flashing manner to alert the doctor to timely make the anti-stress treatment. For more details on stress models, see fig. 3, fig. 4 and their associated descriptions.

In some embodiments of the present description, the information of the wound stress response of the subject is determined by the volatile organic compounds exhaled by the subject, so that the possible wound stress response of the subject can be predicted more quickly and conveniently, and the anesthesiologist can be helped to learn the condition of the patient in time and take measures to slow down the wound stress response of the patient in the clinical operation process.

In some embodiments, the sample detection module is further configured to: acquiring a signal response value of an ion fragment corresponding to at least one volatile organic compound; based on the signal response value, determining the abundance of at least one of the biomarkers.

An ionic fragment is a fragment formed by cleavage of a volatile organic compound molecule in a sample detection module. The ion fragment may provide information about the molecular structure of the biomarker. For example, mass spectrometers analyze samples by measuring the mass-to-charge ratio (m/z, mass-to-charge ratio) of ion fragments. The m/z value of each ion fragment may provide information about its mass and charge state to determine molecular structure information of the biomarker to which it corresponds.

In some embodiments, the ionic fragments of the VOCs may be acquired based on a sample detection module. The sample detection module may also acquire ionic fragments of VOCs in a sample of the subject by means of an external detection device (e.g., a mass spectrometer).

In some embodiments, the ionic fragments of the biomarker comprise an ionic fragment of m/z=43 and/or an ionic fragment of m/z=61. In some embodiments, the ionic fragment of the biomarker comprises an ionic fragment of m/z=43. In some embodiments, the ionic fragment of the biomarker comprises an ionic fragment of m/z=61. In some embodiments, the ionic fragments of the biomarker include an ionic fragment of m/z=43 and an ionic fragment of m/z=43.

As can be seen from the later validation data, some VOCs as biomarkers are significantly correlated with wound stress related indicators.

The wound stress related index refers to a physiological index closely related to the wound stress response. Cortisol, for example, is a stress hormone, the elevation of which is often associated with stress events. The wound stress related index may include concentration of norepinephrine, epinephrine, cortisol, and the like. Wherein, the concentration of norepinephrine, epinephrine and cortisol can be measured by collecting blood. In some validation examples, the wound stress related indicator is norepinephrine.

It can be seen further based on the validation data that the ion fragments of m/z=43 and/or the ion fragments of m/z=61 are significantly correlated with the wound stress related indicators.

In some embodiments, an ultraviolet light ionization time-of-flight mass spectrometry detection device may be employed to obtain a signal response value for an ion fragment corresponding to each of at least one VOC based on the gas exhaled by the subject. In some embodiments, the ion source of the ultraviolet light ionization time-of-flight mass spectrometry detection apparatus may be configured to include: pressure: 7.7kPa; flow rate: 70mL/min; sample temperature: 80-100 ℃; ion source temperature: 80-100 ℃; current flow: 0.99mA. In some embodiments, the ion source of the ultraviolet ionisation time-of-flight mass spectrometry detection means is arranged to: pressure: 7.7kPa; flow rate: 70mL/min; sample temperature: 100 ℃; ion source temperature: 100 ℃; current flow: 0.99mA. The state of the ultraviolet ionization time-of-flight mass spectrum detection device comprises: temperature: 20-40 ℃; humidity is 25-35%. For example, the temperature may be 20 ℃, 30 ℃ or 40 ℃, and the humidity may be 25%, 30% or 35%. In some embodiments, the state of the ultraviolet light ionization time-of-flight mass spectrometry detection device is: the temperature is 38.49 ℃; humidity 31.26%.

Fig. 3 is an exemplary schematic diagram of a stress model shown in accordance with some embodiments of the present description.

In some embodiments, the processor may predict the wound stress response information through the stress model based on the abundance of the biomarker at the plurality of time points.

As shown in fig. 3, stress model 320 may be used to predict wound stress response information 330. In some embodiments, stress model 320 may be a machine learning model.

In some embodiments, the input to the stress model 320 may be the abundance of the biomarker at a plurality of time points 310 and the output may be the wound stress response information 330. In some embodiments, the input of the stress model 320 at each of the plurality of time points may include the abundance of the ion fragment of m/z=43. In some embodiments, the input of the stress model 320 at each of the plurality of time points may include the abundance of the ion fragment of m/z=61. In some embodiments, the input of stress model 320 at each of a plurality of time points may include both the abundance of the ion fragment of m/z=43 and the abundance of the ion fragment of m/z=61.

The plurality of time points may include any time point. The abundance of the plurality of time points can include an abundance of the biomarker detected at a first time (e.g., pre-operative) and an abundance of the biomarker detected at a second time (e.g., intra-operative). For a detailed description of biomarkers and wound stress response information, see fig. 2 and its related content.

In some embodiments, stress model 320 may include a neural network layer or a time-based model layer.

For example, the neural network layer may be any feasible model layer such as a deep neural network (Deep Neural Networks, DNN). The model layer model based on the time sequence can be any feasible model layer such as a cyclic neural network (Recurrent Neural Network, RNN), a Long Short-Term Memory (LSTM) and the like.

When the stress model comprises a time-based model layer, the input to the stress model may be an abundance sequence. The abundance sequence comprises biomarker abundances recorded in chronological order. In some embodiments, the processor may process the abundance of the biomarker for a plurality of acquisition time points stored in the storage device to obtain an abundance sequence. The stress model outputs wound stress response information based on the processing of the abundance sequences. For example, an abundance sequence consisting of abundances of biomarkers corresponding to different time points can be represented by the processor as a= [ (T1, 5%), (T2, 8%), (T3, 7%) ], representing that the abundance of biomarker a is 5% at time T1, 8% at time T2, and 7% at time T3. The stress model outputs wound stress response information based on the abundance change of the biomarker a: the traumatic stress response was grade 1. Wherein, level 1 represents the extent of the traumatic stress response. The extent of the traumatic stress response may be represented by any feasible means such as a grade, score size, etc. The stress model output may also include whether a stress response is occurring. The output of whether a stress response occurs can be expressed in a variety of ways. For example, 0 and 1 may be used to indicate whether a stress response occurs. When the predicted result is 1, indicating that the historical wound stress response occurs; when the predicted result is 0, it indicates that no historical wound stress response occurs.

In some embodiments of the present disclosure, the wound stress response information is determined based on the abundance sequence, which can more accurately reflect the change of the wound stress response of the subject between different time points, and further improves the accuracy of determining the wound stress response information by the stress model.

In some embodiments of the present description, predicting the wound stress response information based on the stress model is beneficial to improving the efficiency and accuracy of predicting the wound stress response information. Predicting the traumatic stress response information based on the stress model may identify at an early stage the risk that the subject may be exposed to the traumatic stress response, enabling the physician to take early intervention to prevent the development of the traumatic stress response or to reduce its severity.

In some embodiments, the processor may train to derive the stress model 320 based on a number of first training samples with first tags.

Wherein the first training sample comprises the abundance of the sample biomarker of the sample subject at a plurality of time points in the sample, e.g., the abundance of the sample biomarker of the sample subject measured prior to the historical surgery and the abundance of the sample biomarker of the sample subject detected during the historical surgery. In some embodiments, the first training sample may be obtained based on historical data.

The first label corresponding to the first training sample is a predicted result of historical wound stress response corresponding to abundance of the sample biomarker of the sample subject at a plurality of time points of the sample. The predicted outcome may include 0 and 1, or may be a historical degree of traumatic stress response. Wherein, when the predicted result is 1, it indicates that a historical wound stress response occurs; when the predicted result is 0, it indicates that no historical wound stress response occurs. The degree of historical wound stress may be expressed in any feasible manner, such as grade, size, etc. For example, the degree of historical wound stress response may be of the order of 1-3, with the greater the number of orders, the more severe the degree of historical wound stress response.

In some embodiments, the first tag may be obtained by manual labeling. In some embodiments, the first tag may be obtained from historical data.

In some embodiments, the first label corresponding to the first training sample may also be obtained based on a historical wound stress related indicator corresponding to the sample.

The historical wound stress related index refers to an index related to a historical wound stress response. The historical wound stress related indicators may include one or more of the concentration of phenylephrine, epinephrine, cortisol, and the like. In some embodiments, the historical wound stress related indicator may be measured by blood collection.

In some embodiments, the first label may be determined from the amount of change in the historical wound stress related indicator over the first time and the second time.

For example, the first label may be determined based on the historical norepinephrine amounts of change in the first time and the second time for the first training sample.

The first time may be a point in time prior to surgery. The second time may be a point in time during surgery. The amount of change in the first time and the second time is the concentration difference between the two time points. For example, the historical norepinephrine concentration is 100pg/mol at a first time and 250pg/mol at a second time, and the amount of change is 150pg/mol. The change amount can reflect the intensity of the historical wound stress response, and the greater the change amount is, the greater the historical wound stress response is.

In some embodiments, the first tag may be determined by looking up a preset table based on historical norepinephrine variation. For example, a table of correspondence between the amount of change of historical norepinephrine and the degree of response to historical wound stress determined manually may be preset, for example, the degree of response to historical wound is 1 when the amount of change of historical norepinephrine is 100-200pg/mol, the degree of response to historical wound is 2 when 200-300pg/mol, etc., and the first label may be obtained by looking up a table. In some embodiments, the first tag may be obtained based on historical norepinephrine variation. For example, the processor may take as the first label of the first training sample the historical wound stress response level of other subjects having the same amount of change in historical norepinephrine as the first training sample.

The first label is determined according to the change amounts of the historical norepinephrine corresponding to the first training sample in the first time and the second time, so that the determined historical wound stress response degree is more reasonable, and the actual operation condition is met.

The label is determined based on the historical wound stress related index corresponding to the sample, so that factors related to the historical wound stress response can be fully considered, and the accuracy of the label is improved.

In the training process, a plurality of first training samples with labels may be input into the initial stress model 320, a loss function is constructed through the labels and the results of the initial stress model 320, and parameters of the initial stress model 320 are iteratively updated based on the loss function. And when the loss function of the initial stress model 320 meets the preset condition, model training is completed, and a trained stress model 320 is obtained. The preset condition may be that the loss function converges, the number of iterations reaches a threshold value, etc.

FIG. 4 is an exemplary schematic diagram of another stress model shown in accordance with some embodiments of the present description.

In some embodiments, the input of the stress model further comprises a sign parameter characteristic of the subject.

The physical sign parameter is characterized by an observable quantitative indicator that is used to describe the physiological state and health of the subject. In some embodiments, the physical parameter characteristics include at least one of pulse oximetry, an electrocardiogram, a heart rate, non-invasive arterial blood pressure, end-tidal partial pressure of carbon dioxide, and a depth of anesthesia monitoring indicator. The anesthesia depth monitoring index is an index for evaluating the anesthesia and sedation depth of a patient based on brain electrical data. The representation of the anesthesia depth monitoring indicator is typically a dimensionless number that quantifies the extent of inhibition of the nervous system activity of the subject during surgical anesthesia. The anesthesia depth monitoring index may include a dual frequency index (BIS), a Patient Status Index (PSI), an entropy index, etc.

In some embodiments, the subject's vital parameter characteristics may be represented as a characteristic vector by processing by a processor. For example, construct a sign parameter feature vectorWherein each element in the vital parameter feature vector represents a vital parameter of a subject.

In some embodiments, the vital parameter characteristics may be obtained by measuring and recording using a corresponding instrument, device or tool. For example, the concentration of norepinephrine in the blood of a subject is measured using a pulse oximeter, a pulse meter, or touching an artery (e.g., radial, ulnar, carotid, femoral), calculating the pulse rate per minute, measuring the heart rate per minute using a heart rate meter or an Electrocardiogram (ECG), or collecting a blood sample of the subject using a disposable syringe. In some embodiments, the vital parameter characteristics may be based on manual acquisition. For example, the breathing of the subject can be observed manually, the fluctuation of the chest or the abdomen of the patient is observed directly, the breathing times per minute are calculated, the physical sign parameter characteristics of the subject are recorded, and the recorded results are uploaded and processed by the processor. For another example, brain electrical data of the subject may be collected by brain electrical monitoring to determine anesthesia depth monitoring indicators. For example, a score of 0 to 100 is used to represent the subject's score in terms of depth of anesthesia, where 0 represents the deepest state of anesthesia and 100 represents complete wakefulness.

In some embodiments, the vital parameter characteristics may be obtained based on a parameter monitoring module that continuously monitors the subject.

The parameter monitoring module is configured to continuously monitor a physical sign parameter characteristic of the subject. In some embodiments, the parameter detection module and the sample detection module may be integrated in one module. In some embodiments, the parameter detection module and the sample detection module may share a storage device, or may have respective storage devices.

In some embodiments, the vital parameter characteristics may be obtained in a variety of ways based on a parameter monitoring module that continuously monitors the subject. For example, the vital sign parameter characteristics uploaded by the parameter monitoring module in real time may be obtained. For another example, the parameter monitoring module may obtain the vital parameter characteristics uploaded at intervals (e.g., 5 minutes).

The sign parameter feature may be used to determine stress response information of the subject in a stress situation. For example, when a subject is exposed to mechanical stimuli such as knife cuts, sticks, or other non-damaging stimuli such as current, high temperature, strong acids, strong bases, or other physical and chemical factors, the characteristic features may help record and quantify these changes, thereby better responding to stress. By measuring the change in the characteristic of the physical sign parameter, the method is helpful for understanding the influence of stress on the emotion of the subject, understanding the adaptability of the individual under stress and the response strategy, or determining whether the intervention treatment is helpful for relieving stress response or improving the mental health of the subject, etc.

According to the embodiment of the specification, the physical sign parameter characteristics of the subject are used as the input of the stress model, the wound stress response information is output, and the anesthesia doctor and the operation team can be helped to manage the anesthesia and the overall condition of the subject in real time.

In some embodiments, the input of the stress model further comprises a baseline characteristic of the subject.

Baseline characteristics refer to the fundamental characteristics of the subject. For example, the baseline characteristic may include at least one of age, sex, disease history, diagnostic information, treatment history, wound tolerance of the subject. In some embodiments, the baseline characteristic may be represented by a characteristic vector. For example, constructing baseline feature vectorsWherein each element in the baseline characteristic vector represents one baseline characteristic of the subject.

In some embodiments, the baseline characteristic may be obtained based on historical data. For example, the processor may obtain a historical diagnosis and treatment record, case, etc. of the subject to obtain the baseline characteristic. In some embodiments, the baseline characteristics may also be obtained manually. For example, information of age, sex, etc. manually filled out may be acquired, and the subject's wound tolerance may be acquired through a pre-operative evaluation.

The baseline characteristics may also be obtained by other means, for example, the pre-operative anesthesia risk stratification of the subject may be determined in conjunction with the ASA stratification system of the American anesthesiologist Association, thereby determining the baseline characteristics of the subject:

ASA I: normal health status. The patient has no systemic disease and good physiological and psychological conditions.

ASA II: mild systemic disease. Patients may have mild systemic disease but do not affect activities of daily living.

ASA III: moderate systemic disease. Patients may have one or more moderate systemic diseases that may affect activities of daily living.

ASA IV: severe systemic disease. Patients have serious systemic diseases that have severely affected activities of daily living.

ASA V: approaching a dead state. The patient is in near-dead condition and emergency surgery is required to save lives.

According to the embodiment of the specification, the baseline characteristic of the subject is taken as the input of the model, the traumatic stress response information is output, the influence of individual differences of different subjects on the traumatic stress response prediction is fully considered, the traumatic stress response information of the subject can be predicted more accurately, and doctors are guided to formulate anti-inflammatory, anti-stress and analgesic related operations suitable for different subjects, so that the prognosis of the subject is improved.

In some embodiments, the input of the stress model further comprises surgical information.

The operation information refers to information related to operation. In some embodiments, the surgical information includes at least one of a surgical type, a surgical stimulation intensity, and surgical plan information. The types of surgery include, but are not limited to, open surgery, laparoscopic surgery, minimally invasive surgery, and robotic-assisted surgery.

In some embodiments, the surgical information may be obtained in a variety of ways. For example, the type of surgery, the information of surgery, etc. may be obtained by manual input, or may be obtained based on a doctor's diagnosis and treatment record, a surgical plan record, etc. In some embodiments, the surgical stimulus intensity may be obtained by an intensity determination model. Specifically, the input of the intensity determination model is an operation image of a doctor, and the operation stimulus intensity is output. Wherein the surgical operation image may be acquired by the image acquisition device. The surgical stimulus intensity may be represented by any feasible means, such as numbers, text, etc.

According to the embodiment of the specification, the surgical information is used as the input of the stress model, so that the corresponding wound stress response information of different surgical type subjects and different doctors under the surgical stimulus intensity can be predicted more accurately.

As shown in fig. 4, inputs to the stress model 450 may include abundance of biomarkers 410 at various time points, subject's vital parameter characteristics 420, baseline characteristics 430, and surgical information 440, and the output may be wound stress response information 460.

In some embodiments, the processor may train the stress model 450 based on a number of second training samples with second labels. Wherein the second training sample comprises abundance of sample biomarkers of the sample subject at a plurality of sample time points, sample sign parameter characteristics, sample baseline characteristics, and sample surgical information. In some embodiments, the second training sample may be obtained based on historical data. The second label of the second training sample may be a predicted result of the historical wound stress response and a degree of the historical wound stress response corresponding to the abundance of the sample biomarker of the sample subject at the plurality of time points of the sample, the sample sign parameter characteristic, and the sample baseline characteristic. In some embodiments, the second label may be determined from the abundance of the sample biomarker of the sample subject at a plurality of time points in the sample, the sample vital sign parameter characteristics, and the amount of change in the historical wound stress related indicator corresponding to the sample baseline characteristics at different time points. Specifically, the second tag is obtained in a similar manner to the first tag, and will not be described herein.

During the training process, the processor may input a second training sample into the initial stress model 450, resulting in a trained stress model 450. Specifically, the stress model 450 is trained in a similar manner to the stress model 320, and will not be described in detail herein.

In some embodiments of the present disclosure, the physical parameter characteristics, baseline characteristics and operation information of the subject are used as inputs of a stress model, and various factors related to the traumatic stress response are fully considered, so that a suitable anti-stress scheme can be formulated for different subjects and different operation conditions, thereby reducing the occurrence rate of postoperative complications. Meanwhile, the stress model is used for predicting the stress response information of the wound, so that the efficiency and accuracy of predicting the stress response information of the wound are improved.

Verification data for traumatic stress response monitoring using VOC markers

A total of 105 patients were collected from Huaxi Hospital, university of Sichuan, and were all patients with general anesthesia for downstream abdominal phase selection surgery. The baseline characteristics of the patients are shown in table 1.

Table 1 baseline characteristics of patients

The study was approved by the biomedical ethics committee of the university of Sichuan Huaxi hospital and written informed consent was obtained for each subject prior to surgery.

After the patient enters the operating room, pulse oxygen saturation (SpO) is continuously monitored ₂ ) Electrocardiogram (ECG), heart Rate (HR), non-invasive arterial blood pressure (SBP, DBP, MBP), partial end-tidal carbon dioxide pressure (P) _ET CO ₂ ) Patient State Index (PSI). Anesthesia induction protocol: 2mg of midazolam was intravenously administered, 2 minutes later followed by controlled infusion of propofol (TCI), plasma concentration was set to 3.0. Mu.g/mL, and after patient consciousness had been lost, 0.2mg/kg of cis-atracurium and 0.4ug/kg of sufentanil were intravenously administered and assisted ventilation was performed, and tracheal intubation was performed and mechanical ventilation was performed with 2 or less of four consecutive myofibrillation reactions (TOF.ltoreq.2) obtained by electrically stimulating nerves. Setting air/oxygen mixing (1L: 1L) ventilation, and maintaining P with respiratory rate of 10-18 times/min and tidal volume of 6-10 mL/kg _ET CO ₂ The value is 35-45 mmHg, spO ₂ 97% -100%. Anesthesia maintenance protocol: continuing to infuse propofol TCI after anesthesia induction is finished, adjusting the infusion rate according to the PSI value, and adjusting to be difluoride to maintain anesthesia after the propofol infusion is finished; intravenous infusion of remifentanil is started 5min before operation, and the infusion rate is 0.1-0.3 ug/(kg.min). The PSI value is maintained between 25 and 50, and cis-atracurium is intermittently administered according to TOF value. The heart rate and blood pressure changes of the patient are maintained within 20% of the basal level. Hydromorphone 20ug/kg was expected to be administered intravenously about 30 minutes before the end of the procedure and 0.5% subcutaneous infusion of ropivacaine after the end of the skin incision. The procedure Bi Bachu endotracheal tube was delivered to the post clinical recovery unit (PACU).

After the patient enters the room, basic information such as age, sex, ASA classification, BMI value, operation mode and the like is recorded, and the Pre-operation (Pre-op) after anesthesia induction and the operation time (Intr) after 2 hours are recordeda-op), invasive arterial systolic pressure, diastolic pressure, mean arterial pressure, heart rate, pulse, PSI, spO of a patient before tracheal tube removal (End-op) after the End of the procedure ₂ 、P _ET CO ₂ And collecting 2mL of patient exhale at the corresponding moment and radial arterial blood sample at the corresponding moment. The nursing operation of injecting normal saline into the airway for humidifying and sucking phlegm is avoided 15min before the expired air sample is collected, the pipeline is replaced by a dry threaded pipe of a sterilizing breathing machine during collection, high-purity nitrogen is cleaned, high-temperature treatment is carried out, and a Tedlar gas sampling bag after vacuum suction is connected in series between the expired air end of the pipeline of the anesthesia machine and the anesthesia machine, so that platform-stage alveolar expired air collection is carried out. Each collection time was filled with 500mL Tedlar PVF sampling bags. The environment in the operating room is controlled at 20-25 ℃ and the humidity is controlled at 45-50%. The obtained specimen is immediately placed in a refrigerator at 4 ℃ for preservation. And (3) performing EDTA anticoagulation treatment on the blood sample, centrifuging at 1000r/min for 10min, taking supernatant, and storing in a refrigerator at-80 ℃ to be tested. The collected breath sample is stored at low temperature for no more than 48 hours until analyzed. And after the sample is rewarmed, a UVI-TOF MS (ultraviolet ionization time-of-flight mass spectrometer) is directly adopted for sample injection analysis without pretreatment. The ion source is set as follows: (a) pressure: 7.7kPa; (b) flow rate: 70mL/min; (c) sample temperature: 100 ℃; (d) ion source temperature: 100 ℃; (e) current: 0.99mA.

Tables 2 and 3 are the change in VOC abundance between preoperative and operative, and between operative and postoperative, respectively, R > 0 indicates positive correlation of VOC with norepinephrine, and R <0 indicates negative correlation of VOC with norepinephrine.

A first set of VOCs significantly correlated with the presence of norepinephrine preoperatively and intraoperatively is shown in table 2, and a second set of VOCs significantly correlated with the presence of norepinephrine postoperatively and intraoperatively is shown in table 3. VOCs contained in both the first and second sets of VOCs are selected as biomarkers for subsequent wound stress monitoring.

TABLE 2 variation in VOC abundance between preoperative and intraoperative anesthesia surgery

/>

TABLE 3 variation in VOC concentration between intra-and post-anesthesia surgery

/>

The 25 VOCs determined above, which were detected and significantly changed throughout the subject anesthesia procedure, were correlated with the plasma wound stress related indicators, and it can be seen from the table that VOCs significantly negatively correlated with norepinephrine, i.e., comp_6 and comp_23. From the ROC curves obtained for comp_6 for detecting the presence or absence of traumatic stress reaction before and during surgery and for detecting the presence or absence of traumatic stress reaction during and after surgery, ROC curve analysis gave comp_6 as differentiating the area under the curve between before and during surgery from 0.736 (95% CI:0.6680-0.8040, p <0.0001, a in fig. 5) and the area under the ROC curve between during and after surgery from 0.599 (95% CI:0.5225-0.6756, p=0.0131, B in fig. 5). From the ROC curves obtained for comp_23 for detecting the presence or absence of traumatic stress reaction before and during surgery and for detecting the presence or absence of traumatic stress reaction during and after surgery, ROC curve analysis gave comp_23 as distinguishing the area under the curve before and during surgery as 0.7004 (95% CI:0.6302-0.7706, p <0.0001, C in fig. 5) and the area under the ROC curve after and during surgery as 0.5832 (95% CI:0.5062-0.6603, p= 0.0372, D in fig. 5). These results demonstrate that the extent of wound stress response at different surgical moments can be effectively differentiated based on the corresponding VOC biomarkers.

The above results demonstrate that potential VOC biomarkers from some embodiments of the present description can monitor the extent of wound stress response to varying degrees. The stress model predicts the wound stress response information through processing the biomarker abundance corresponding to different time points and the physical sign parameter characteristics of the subject, the baseline characteristics of the subject and/or the operation information, and can fully identify and learn the relation between the relevant characteristics in the subject data and the wound stress response information, thereby helping doctors understand the potential influencing factors of the wound stress response and the mechanism of the wound stress response. In addition, the stress model can provide personalized, accurate and efficient wound stress response prediction for each subject by analyzing a large amount of individual data, so that medical professionals can be helped to better know the condition of each patient, and an individual treatment scheme can be adopted. Meanwhile, based on accurate prediction of the stress model, the risk that a subject may face a traumatic stress response can be identified in an early stage, so that a medical team can take early intervention measures to prevent the development of the traumatic stress response or reduce the severity of the traumatic stress response.

The system for detecting the wound stress response can realize rapid, convenient and accurate monitoring of the wound stress response, and has good application prospect.

The embodiments in this specification are for illustration and description only and do not limit the scope of applicability of the specification. Various modifications and changes may be made by those skilled in the art in light of the present description while remaining within the scope of the present description.

Furthermore, certain features, structures, or characteristics of one or more embodiments of the present description may be combined as suitable.

In some embodiments, numbers describing the components, number of attributes are used, it being understood that such numbers being used in the description of embodiments are modified in some examples by the modifier "about," approximately, "or" substantially. Unless otherwise indicated, "about," "approximately," or "substantially" indicate that the number allows for a 20% variation. Accordingly, in some embodiments, numerical parameters set forth in the specification and claims are approximations that may vary depending upon the desired properties sought to be obtained by the individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and employ a method for preserving the general number of digits. Although the numerical ranges and parameters set forth herein are approximations that may be employed in some embodiments to confirm the breadth of the range, in particular embodiments, the setting of such numerical values is as precise as possible.

If the description, definition, and/or use of a term in this specification makes reference to a material that is inconsistent or conflicting with the disclosure provided herein, the description, definition, and/or use of the term in this specification controls.

Claims

1. A system for detecting a wound stress response, comprising:

a sample detection module configured to determine an abundance of a biomarker in a sample of a subject, wherein the biomarker comprises a volatile organic compound exhaled by the subject; and

a processor configured to predict traumatic stress response information for the subject based on the abundance of the biomarker.

2. The system of claim 1, wherein the sample detection module is further configured to:

acquiring a signal response value of an ion fragment corresponding to at least one volatile organic compound;

based on the signal response value, determining the abundance of at least one of the biomarkers.

3. The system of claim 2, wherein the ion fragments comprise ion fragments of m/z=43 and/or ion fragments of m/z=61.

4. The system of claim 1, wherein the predicting the subject's traumatic stress response information comprises:

Predicting the wound stress response information by a stress model based on the abundance of the biomarker at a plurality of time points, the stress model being a machine learning model.

5. The system of claim 4, wherein the input of the stress model further comprises a vital parameter characteristic of the subject, wherein the vital parameter characteristic comprises at least one of pulse oxygen saturation, an electrocardiogram, a heart rate, non-invasive arterial blood pressure, end-tidal carbon dioxide partial pressure, and an anesthesia depth monitoring index, the vital parameter characteristic being obtained based on a parameter monitoring module that continuously monitors the subject.

6. A method implemented on a computing device having at least one processor and at least one storage device to detect a wound stress response, the method comprising:

determining the abundance of a biomarker in a sample of a subject, wherein the biomarker comprises a volatile organic compound exhaled by the subject;

based on the abundance of the biomarker, wound stress response information of the subject is predicted.

7. The method of claim 6, wherein determining the abundance of the biomarker in the sample of the subject comprises:

8. The method of claim 7, wherein the ion fragments comprise ion fragments of m/z=43 and/or ion fragments of m/z=61.

9. The method of claim 6, wherein predicting the subject's traumatic stress response information comprises:

10. A non-transitory computer-readable medium comprising at least one set of instructions, wherein the at least one set of instructions, when executed by at least one processor of a computing device, cause the computing device to perform a method comprising: