Drainage liquid sudden increase rapid early warning system based on artificial intelligence

Abstract

The invention relates to the technical field of drainage liquid sudden increase early warning, in particular to a drainage liquid sudden increase rapid early warning system based on artificial intelligence. The system obtains drainage volume data and various physiological indexes in real time; according to the difference between the drainage volume data changes of the current patient and the preset similar historical patient and the drainage volume data change condition of the current patient, the drainage abnormality degree is obtained, and the drainage abnormality time period is determined; according to the duration of the abnormal drainage time period of the current patient at the current moment, the change of the abnormal drainage degree in the abnormal drainage time period and the abnormal condition of the physiological indexes, the abnormal drainage degree is determined; and acquiring an early warning degree to perform early warning on the drainage sudden increase abnormity of the current patient at the current moment according to the occurrence condition of the drainage abnormal time period of the current patient at the current moment, the overall abnormal condition of the physiological indexes and the change of the interval duration of the adjacent drainage abnormal time periods. According to the invention, early warning is accurately carried out on drainage sudden increase abnormity of the current patient, so that the harm of drainage abnormity is effectively reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sudden increase of drainage liquid early warning, in particular to a rapid early warning system for sudden increase of drainage liquid based on artificial intelligence. BACKGROUND

[0002] In the medical field, patients after pneumothorax, heart surgery or thoracic surgery often need to receive thoracic drainage treatment. The core purpose is to discharge the abnormal accumulation of gas, blood and exudate in the thoracic cavity to the outside of the body through the drainage device, help the thoracic cavity to restore the normal negative pressure environment, and protect the normal expansion and respiratory function of the lung. In the drainage process, the drainage volume of the drainage material needs to be monitored. The sudden increase of the drainage volume is an important signal that the patient may have active bleeding and other critical pathological conditions. Timely and accurate identification of the signal and early warning have important clinical significance for avoiding disease delay and reducing the risk of postoperative complications.

[0003] The current clinical early warning method for sudden increase of drainage liquid mainly relies on fixed threshold judgment logic, that is, the drainage device monitors the drainage volume data in real time, calculates the drainage volume growth rate in a unit of time, and sets a unified alarm threshold for all patients or similar patients. When the patient's drainage volume growth rate exceeds the threshold, the drainage device will automatically alarm. However, in actual clinical application, the patient's normal posture adjustment will change the pressure distribution in the thoracic cavity and the liquid flow state of the drainage tube, which may cause the drainage volume to increase rapidly in a short time. However, this increase in drainage volume is not caused by pathological factors. The drainage volume data will quickly return to the normal range after the posture is stable. The existing early warning method for sudden increase of drainage liquid cannot rule out such physical interference, and thus is prone to frequent false alarms, thereby increasing the risk of missing the real critical pathological conditions (such as active bleeding), and is difficult to meet the needs of fine monitoring and safety protection of postoperative patients. SUMMARY

[0004] In order to solve the technical problem that the existing early warning method for sudden increase of drainage liquid cannot accurately identify real drainage abnormalities, the purpose of the present application is to provide a rapid early warning system for sudden increase of drainage liquid based on artificial intelligence. The technical scheme adopted is as follows: The present application provides a rapid early warning system for sudden increase of drainage liquid based on artificial intelligence, which comprises: A data acquisition module for acquiring the drainage volume data and various physiological indicators of the patient in real time; A drainage abnormal period acquisition module for acquiring the drainage abnormal degree of the patient at each time point according to the difference in drainage volume data change of the current patient and the pre-set similar historical patient at each time point in the drainage process, and the drainage volume data change of the current patient at each time point, determining the drainage abnormal period of the current patient; an early warning degree acquisition module configured to acquire an early warning degree of the current patient at the current time according to a length of the drainage abnormal period in which the current patient is located at the current time, a change of the drainage abnormal degree in the drainage abnormal period in which the current patient is located, and an abnormal condition of various physiological indexes, and an occurrence of the drainage abnormal period of the current patient up to the current time, an overall abnormal condition of various physiological indexes, and a change of the adjacent drainage abnormal period interval length; a data processing module configured to give an early warning to the drainage sudden increase abnormality of the current patient at the current time based on the early warning degree.

[0005] Further, the method for acquiring the drainage abnormal degree comprises: arranging the drainage volume data of the current patient and the drainage volume data of each preset similar historical patient of the current patient according to a time sequence to obtain a drainage volume sequence of the corresponding patient; wherein initial drainage volume data in the drainage volume sequence is not 0, and the drainage volume data at the same position in all the drainage volume sequences correspond to a same time point in a drainage process; for any time point in the drainage process of the current patient and any drainage volume sequence, taking a ratio of the drainage volume data corresponding to the time point in the drainage volume sequence to the initial drainage volume data in the drainage volume sequence as a drainage growth index of the corresponding patient in the drainage volume sequence at the time point; taking an average of the drainage growth indices of all the preset similar historical patients of the current patient at the time point as a drainage growth reference value at the time point; taking a difference between the drainage growth index of the current patient at the time point and the drainage growth reference value at the time point as a first drainage growth analysis value of the current patient at the time point; fitting the drainage volume sequence of the current patient into a curve to obtain a tangent slope of the drainage volume data at the time point on the curve as a second drainage growth analysis value of the current patient at the time point; taking a result of normalizing a sum of the first drainage growth analysis value and the second drainage growth analysis value as the drainage abnormal degree of the current patient at the time point.

[0006] Further, the method for acquiring the drainage abnormal period comprises: when the drainage abnormal degree is greater than a preset drainage abnormal degree threshold value, taking the corresponding time point as a drainage abnormal time point of the current patient; taking a continuous time period formed by at least two continuous drainage abnormal time points as a drainage abnormal period of the current patient.

[0007] Further, the method for acquiring the early warning degree comprises: According to the length of the drainage abnormal period in which the current patient is located at the current time, and the change of the drainage abnormal degree in the drainage abnormal period and the abnormal situation of various physiological indicators, a first abnormal analysis degree of the current patient at the current time is obtained; According to the occurrence of the drainage abnormal period of the current patient up to the current time, the overall abnormal situation of various physiological indicators, and the change of the interval length of adjacent drainage abnormal periods, a second abnormal analysis degree of the current patient at the current time is obtained. The result of adding and normalizing the first abnormal analysis degree and the second abnormal analysis degree is taken as the warning degree of the current patient at the current time.

[0008] Further, the method for obtaining the first abnormal analysis degree is: The drainage abnormal period in which the current patient is located at the current time is taken as a target period; wherein the terminal time of the target period is the current time; The slope of the straight line fitted according to the drainage abnormal degree in the target period in time sequence is taken as the drainage abnormal change value of the current patient at the current time; According to the size of each physiological indicator of the current patient at each time, the deviation degree of each physiological indicator of the current patient at each time is obtained, and the reference abnormal time of each physiological indicator of the current patient is determined; For any physiological indicator, the length between the initial reference abnormal time of the physiological indicator in the target period and the initial time of the target period is taken as the lag length of the physiological indicator; The total number of time points contained from the initial reference abnormal time of the physiological indicator in the target period to the current time is taken as a first number; the total number of reference abnormal times of the physiological indicator in the target period is taken as a second number; and the ratio of the second number to the first number is taken as the abnormal duration analysis value of the physiological indicator; The result of adding and normalizing the total length of the target period, the drainage abnormal change value, the mean of the lag lengths of all physiological indicators, the mean of the abnormal duration analysis values of all physiological indicators, and the number of physiological indicators corresponding to the reference abnormal times in the target period is taken as the first abnormal analysis degree of the current patient at the current time.

[0009] Further, the method for obtaining the deviation degree is: For any time and any physiological indicator of the current patient in the drainage process, the mean of the physiological indicator of all preset similar historical patients of the current patient at the time is obtained as the target analysis value of the physiological indicator; The result of normalizing the difference between the physiological indicator of the current patient at the time and the target analysis value is taken as the deviation degree of the physiological indicator of the current patient at the time.

[0010] Further, the method for obtaining the reference abnormal time point is: When the deviation degree is greater than the preset deviation degree threshold, the corresponding time point is taken as the reference abnormal time point of the physiological indicator of the current patient.

[0011] Further, the method for obtaining the second abnormality analysis degree is: The interval duration between any two adjacent drainage abnormal time periods is taken as the first duration; The first duration is fitted into a curve according to the time sequence as an interval duration curve; the tangent slope of each first duration on the interval duration curve is taken as a change trend analysis value of each first duration; The ratio of the number of change trend analysis values less than 0 to the total number of change trend analysis values is taken as a first value; The product of the first value and the inverse of the mean value of all change trend analysis values less than 0 is taken as the abnormal shortening index of the current patient; The result of normalizing the number of drainage abnormal time periods occurring up to the current time point is taken as the cumulative drainage abnormality analysis value of the current patient at the current time point; The ratio of the number of reference abnormal time points of each physiological indicator of the current patient to the total number of all time points corresponding to the current patient up to the current time point is taken as the cumulative reference abnormality analysis value of each physiological indicator of the current patient at the current time point; The result of normalizing the sum of the mean value of the cumulative reference abnormality analysis value, the abnormal shortening index and the cumulative drainage abnormality analysis value is taken as the second abnormality analysis degree of the current patient at the current time point.

[0012] Further, the method for warning the drainage sudden increase abnormality of the current patient at the current time point based on the warning degree is: When the warning degree is greater than the preset warning degree threshold, the drainage sudden increase abnormality of the current patient at the current time point is warned.

[0013] Further, the method for obtaining the preset similar historical patient is: The preset number of historical patients with the same disease type, the same gender, the same surgical and drainage treatment plan, and the age difference within the first preset range as the current patient are all taken as the preset similar historical patient of the current patient.

[0014] The present application has the following advantages: The application firstly obtains the drainage abnormality degree of the current patient at each time point according to the difference between the drainage data change of the current patient and the preset similar historical patient at each same time point in the drainage process and the drainage data change of the current patient at each time point, accurately reflects the drainage abnormality of the current patient at each time point, and then accurately determines the drainage abnormality period of the current patient based on the drainage abnormality degree, so as to prepare for the subsequent real-time analysis of the drainage abnormality of the current patient. Then, the real drainage abnormality of the current patient at the current time point is preliminarily analyzed according to the length of the drainage abnormality period of the current patient at the current time point, the change of the drainage abnormality degree in the drainage abnormality period and the abnormality of various physiological indexes. Meanwhile, the corresponding situation of the accumulated real drainage abnormality of the current patient is analyzed by combining the occurrence of the drainage abnormality period of the current patient, the overall abnormality of various physiological indexes and the change of the interval length of adjacent drainage abnormality periods, so as to further analyze the real drainage abnormality of the current patient at the current time point, accurately obtain the warning degree of the current patient at the current time point, accurately reflect the real drainage abnormality of the current patient at the current time point, effectively avoid the interference of normal physiological fluctuations such as pose adjustment on the identification of real drainage abnormality, and then accurately warn the drainage sudden increase abnormality of the current patient at the current time point based on the warning degree, effectively improve the accuracy of identifying real drainage abnormality, and help to timely find the real critical pathological condition of the current patient, so that the real drainage abnormality is processed in time, and the harm of the real drainage abnormality to the current patient is effectively reduced. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0016] Figure 1 The structural block diagram of the drainage liquid sudden increase rapid warning system based on artificial intelligence provided by an embodiment of the present application; Figure 2 The flow chart of the acquisition method of the warning degree provided by an embodiment of the present application; Figure 3 The schematic diagram of a computer device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0017] In order to further clarify the technical means and effects taken by the present application to achieve the predetermined inventive purpose, the specific implementation, structure, features and effects of the rapid early warning system for sudden increase of drainage liquid based on artificial intelligence according to the present application are described in detail below in combination with the drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0019] The specific scheme of the rapid early warning system for sudden increase of drainage liquid based on artificial intelligence provided by the present application is described below in combination with the drawings.

[0020] Embodiment 1: The present application proposes a rapid early warning system for sudden increase of drainage liquid based on artificial intelligence, please refer to Figure 1 , which shows the structure block diagram of the rapid early warning system for sudden increase of drainage liquid based on artificial intelligence provided by one embodiment of the present application, which includes a data acquisition module 10, a drainage abnormal period acquisition module 20, an early warning degree acquisition module 30 and a data processing module 40.

[0021] The data acquisition module 10 is used to acquire the drainage volume data and various physiological indicators of the patient in real time.

[0022] Specifically, the present embodiment acquires the drainage volume data of the patient at each time point in the drainage process in real time through the drainage device, and acquires various physiological indicators of the patient at each time point in real time through the body monitor, wherein the physiological indicators include the blood pressure, heart rate, blood oxygen saturation and thoracic negative pressure value of the patient. The present embodiment sets the drainage volume data and various physiological indicators to be acquired synchronously, and the time interval between the adjacent two time points of the collected data is 0.1 seconds. The implementer can set the time interval between the adjacent two time points of the collected data according to the actual situation, which is not limited herein.

[0023] The drainage abnormal period acquisition module 20 is used to acquire the drainage abnormal degree of the current patient at each time point according to the difference between the drainage volume data change of the current patient and the pre-set similar historical patient of the current patient at each same time point in the drainage process, and the drainage volume data change of the current patient at each time point, and determine the drainage abnormal period of the current patient.

[0024] Specifically, it is known that when the drainage volume of a certain patient suddenly increases, it indicates that the drainage of the patient at this time is more likely to be abnormal. In order to accurately analyze whether the sudden increase of the drainage volume of the current patient is abnormal, so as to timely adjust the drainage of the current patient and avoid irreversible harm to the current patient caused by abnormal drainage, the embodiment first acquires the preset similar historical patients of the current patient. Because the drainage volume will present natural fluctuations over time during the drainage process, but the fluctuation rules of different patients are significantly different, which is mainly determined by physiological functions and drainage purposes. For example, the drainage target of a pneumothorax patient is to empty the abnormally accumulated gas in the thoracic cavity, and the drainage volume is usually small and the fluctuation trend is gentle. The drainage target of a postoperative heart patient is to drain the blood accumulated in the thoracic cavity, and the amount of blood is relatively large in the early postoperative period, so the overall drainage volume is larger and the fluctuation amplitude is more obvious. If a unified standard is used to judge whether the drainage volume of all patients is abnormal, misjudgment is easy to occur due to individual differences. Therefore, the similar group of the current patient, i.e., the preset similar historical patients, needs to be matched first, and the fluctuations of the above similar group are used as the benchmark to more accurately screen abnormalities. It should be noted that the preset similar historical patients are all patients who have recovered.

[0025] Further analyze the difference in the change of the drainage volume data of the current patient and the preset similar historical patients of the current patient at a certain same time during the drainage process. The greater the difference, the more different the changes in the drainage of the current patient and the preset similar historical patients of the current patient at the time, which indirectly indicates that the drainage of the current patient at the time is more likely to be unreasonable. Further analyze the change of the drainage volume data of the current patient at the time. The greater the increase of the drainage volume data of the current patient at the time, the more likely the drainage of the current patient at the time is unreasonable. Therefore, according to the difference in the change of the drainage volume data of the current patient and the preset similar historical patients of the current patient at each same time during the drainage process, and the change of the drainage volume data of the current patient at each time, the embodiment obtains the abnormality degree of the drainage of the current patient at each time. The greater the abnormality degree, the more likely the corresponding time is the abnormal time of the sudden increase of the drainage volume of the current patient. Further, the abnormal period of the drainage of the current patient is determined based on the abnormality degree, which prepares for the subsequent real-time and accurate analysis of the abnormality of the drainage volume of the current patient.

[0026] Preferably, in an implementable method of the embodiment, the method for acquiring the preset similar historical patients is that: the historical patients with the same disease type, the same gender, the same surgery and drainage treatment scheme, and the age difference within a first preset range as the current patient are all taken as the preset similar historical patients of the current patient. The embodiment sets the preset number to 50 and the first preset range to 5 years. The implementer can set the preset number and the first preset range according to the actual situation, which is not limited here.

[0027] ​Preferably, in one method that can be implemented in the present embodiment, the method for obtaining the degree of drainage abnormality is as follows: arrange the drainage volume data of the current patient and each of the preset similar historical patients of the current patient according to time sequence to obtain a drainage volume sequence corresponding to the patient; wherein the initial drainage volume data in the drainage volume sequence is not 0, and the drainage volume data at the same position in all drainage volume sequences corresponds to the same time point in the drainage process. It should be noted that the drainage volume sequence of the preset similar historical patient of the current patient corresponds to a complete drainage process. For any time point in the drainage process of the current patient and any drainage volume sequence, the ratio of the drainage volume data corresponding to the time point in the drainage volume sequence to the initial drainage volume data in the drainage volume sequence is taken as the drainage growth index of the patient corresponding to the drainage volume sequence at the time point; the greater the drainage growth index, the greater the degree of increase in the drainage volume of the patient corresponding to the time point compared to the initial time point of the drainage process; In order to accurately analyze the abnormality of the change in the drainage volume of the current patient at the time point, the average of the drainage growth indices of all the preset similar historical patients of the current patient at the time point is taken as the drainage growth reference value at the time point; then the absolute value of the difference between the drainage growth index of the current patient at the time point and the drainage growth reference value at the time point is taken as the first drainage growth analysis value of the current patient at the time point; the greater the first drainage growth analysis value, the more abnormal the drainage volume data of the current patient at the time point, which indirectly indicates that the drainage volume data of the current patient at the time point is more likely to have a sudden increase abnormality; further, the drainage volume sequence of the current patient is fitted into a curve to obtain the tangent slope of the drainage volume data at the time point on the curve, which is taken as the second drainage growth analysis value of the current patient at the time point; the greater the second drainage growth analysis value, the greater the degree of increase in the drainage volume data of the current patient at the time point, and the more likely the drainage volume data of the current patient at the time point to have a sudden increase abnormality; then the result of adding and normalizing the first drainage growth analysis value and the second drainage growth analysis value is taken as the degree of drainage abnormality of the current patient at the time point. In the present embodiment, the norm normalization function is used to normalize the addition result of the first drainage growth analysis value and the second drainage growth analysis value. The curve fitting and the method for obtaining the tangent slope are both known techniques and will not be described in detail.

[0028] Preferably, in the method that can be implemented in the embodiment, the method for obtaining the drainage abnormality period is as follows: when the drainage abnormality degree is greater than the preset drainage abnormality degree threshold, the corresponding time is taken as the drainage abnormality time of the current patient; the preset drainage abnormality degree threshold is set to 0.5 in the embodiment, and the implementer can set the size of the preset drainage abnormality degree threshold according to the actual situation, which is not limited herein. Considering that in actual situations, the isolated drainage abnormality time, that is, the drainage abnormality time that is not the drainage abnormality time of adjacent time, is mostly accidental data fluctuation and not real abnormality, therefore, the isolated drainage abnormality time is not considered in the embodiment, and then the continuous time period composed of at least two continuous drainage abnormality times is taken as the drainage abnormality period of the current patient.

[0029] The warning degree obtaining module 30 is configured to obtain the warning degree of the current patient at the current time according to the length of the drainage abnormality period in which the current patient is located at the current time, the change of the drainage abnormality degree in the drainage abnormality period, the abnormality of various physiological indexes, the occurrence of the drainage abnormality period of the current patient up to the current time, the overall abnormality of various physiological indexes, and the change of the interval length of adjacent drainage abnormality periods.

[0030] It is known that the abnormal increase of drainage volume is the core warning signal in the drainage monitoring scene, but this signal can be caused by two types of completely different reasons, one is the normal physiological fluctuation of patient posture change (no need for emergency intervention), and the other is the real drainage surge abnormality of active bleeding (need for timely medical treatment). Due to the great difference in clinical response needs of the two situations, it is impossible to accurately distinguish them only by using drainage volume data. It is known that the change of posture belongs to physical transient interference, which only affects the drainage volume data by changing the intrathoracic pressure, the position of the drainage tube and other external conditions, and does not cause pathological damage to the physiological state of the patient's body; the real drainage surge abnormality belongs to pathological sustained influence, such as active bleeding, which will cause changes in blood volume and circulatory system compensation of the patient, and then cause imbalance of the overall physiological state, which belongs to the clinical risk that needs emergency intervention; therefore, the feedback degree of the physiological indexes of the current patient after the occurrence of the drainage abnormality needs to be combined, so that the real drainage surge abnormality of the current patient can be determined in a timely manner, and then the abnormal drainage of the current patient can be handled in a timely manner.

[0031] It is known that when the abnormal growth of drainage volume data is caused by the change of patient posture, the feedback of physiological indicators presents instantaneousness and recoverability, that is, the abnormality of physiological indicators appears almost synchronously with the abnormal growth of drainage volume data, without obvious hysteresis; at the same time, the duration of the abnormal state of physiological indicators is short, and it can be quickly recovered with stable posture, without persistent abnormality trend. When the abnormal growth of drainage volume is caused by pathological conditions such as active bleeding, the feedback of physiological indicators presents hysteresis and persistence, that is, the abnormality of physiological indicators lags behind the abnormal growth of drainage volume data, and the drainage volume data first appears abnormal growth in the early stage of active bleeding, but the patient's body has a compensatory mechanism, and the physiological indicators will not immediately appear obvious abnormalities. Usually, after a period of time of bleeding, when the compensatory mechanism is not enough to maintain physiological balance, physiological indicators such as blood pressure drop, continuous heart rate rise, and blood oxygen saturation decrease will appear, and the feedback of physiological indicators presents obvious hysteresis; at the same time, the abnormal state of physiological indicators persists and is difficult to recover by itself, because the pathological factors (bleeding) are not intervened, the patient's physiological imbalance state will continue to worsen, and the physiological indicators will remain abnormal, even if the growth rate of drainage volume temporarily slows down, the physiological indicators will not return to normal by themselves, and medical intervention (such as hemostasis, fluid infusion) is needed to improve.

[0032] Further considering that in the case of drainage abnormality caused by real pathological abnormalities, the degree of drainage abnormality will continue to rise because the pathological process may worsen. Therefore, the present embodiment first analyzes the real drainage abnormality of the current patient at the current time according to the length of the drainage abnormal period in which the current patient is located at the current time, the change of the degree of drainage abnormality in the drainage abnormal period, and the abnormality of various physiological indicators. It is known that the fluctuation of physiological indicators of patients with good drainage effect should gradually tend to be stable, and when the drainage effect of the patient is poor, even if the drainage volume data at a certain time does not show a sudden increase, the physiological indicators of the patient are difficult to maintain stable, and the frequency of drainage abnormality also gradually increases; further, in order to more accurately analyze the drainage abnormality of the current patient at the current time, the long-term historical drainage effect and the cumulative trend of drainage abnormality of the current patient also need to be combined.

[0033] The historical drainage effect is focused on the long-term abnormal trend of the physiological indicators in the drainage process, because the stability of the physiological indicators directly reflects the response of the patient's body to the drainage treatment; the drainage abnormality accumulation trend focuses on analyzing the regular change of the occurrence of the drainage abnormality, and focuses on whether the number of occurrences of the drainage abnormality and the interval time between adjacent two drainage abnormalities are continuously shortened until the current time. When the number of occurrences of the drainage abnormality is more, it means that the disease is more unstable, and the potential risk is higher; when the interval time between adjacent two drainage abnormalities is shorter, it means that the disease may be progressing (such as the bleeding speed is accelerated), and the warning intensity needs to be urgently improved. Therefore, according to the occurrence of the drainage abnormal period of the current patient until the current time, the overall abnormality of various physiological indicators, and the change of the interval length between adjacent drainage abnormal periods, the real drainage abnormality of the current patient at the current time is further analyzed. In order to accurately determine the real drainage abnormality of the current patient at the current time, the embodiment obtains the warning degree of the current patient at the current time according to the length of the drainage abnormal period in which the current patient is located at the current time, the change of the drainage abnormality degree in the drainage abnormal period and the abnormality of various physiological indicators, and the occurrence of the drainage abnormal period of the current patient until the current time, the overall abnormality of various physiological indicators, and the change of the interval length between adjacent drainage abnormal periods. The greater the warning degree, the greater the possibility of real abnormality of the drainage of the current patient at the current time, and the more the warning needs to be performed.

[0034] Preferably, in an implementable manner of the embodiment, the method for obtaining the warning degree can refer to Figure 2 which shows a method flowchart for obtaining a warning degree provided by the embodiment, and the method comprises the following steps: Step S201: obtaining a first abnormality analysis degree of the current patient at the current time according to the length of the drainage abnormal period in which the current patient is located at the current time, and the change of the drainage abnormality degree in the drainage abnormal period and the abnormality of various physiological indicators.

[0035] The greater the first abnormality analysis degree, the more likely the drainage abnormality of the current patient at the current time is caused by the active bleeding real drainage abnormality, and the more the drainage sudden increase abnormality of the current patient at the current time needs to be warned.

[0036] In an implementable manner of the embodiment, the method for obtaining the first abnormality analysis degree comprises: taking a drainage abnormality period in which the current patient is located at the current time as a target period; wherein a terminal time of the target period is the current time; and the current time is essentially a drainage abnormality time. A slope of a straight line fitted according to a time sequence of the drainage abnormality degree in the target period is taken as a drainage abnormality change value of the current patient at the current time; the greater the drainage abnormality change value, the more likely the current patient has active bleeding real drainage abnormality at the current time; further, according to the size of each physiological index of the current patient at each time, a deviation degree of each physiological index of the current patient at each time is obtained, the greater the deviation degree, the more abnormal the corresponding physiological index of the current patient at the corresponding time, and then the reference abnormal time of each physiological index of the current patient can be determined based on the deviation degree; then, for any physiological index, a time length between an initial reference abnormal time of the physiological index in the target period and an initial time of the target period is taken as a lag time of the physiological index; a total number of times contained from the initial reference abnormal time of the physiological index in the target period to the current time is taken as a first number; a total number of reference abnormal times of the physiological index in the target period is taken as a second number; and a ratio of the second number to the first number is taken as an abnormality duration analysis value of the physiological index; the greater the lag time and the abnormality duration analysis value, the more likely the physiological index has lag and persistence, and the more likely the current patient has active bleeding real drainage abnormality at the current time. The greater the time length of the target period, the greater the drainage abnormality change value, the greater the lag time and the abnormality duration analysis value of all physiological indexes in the target period, and the greater the number of types of physiological indexes corresponding to the reference abnormal time in the target period, the more likely the current patient has active bleeding real drainage abnormality at the current time, and then the embodiment adds the total time length of the target period, the drainage abnormality change value, the average of the lag time of all physiological indexes, the average of the deviation change degree of all physiological indexes, and the number of types of physiological indexes corresponding to the reference abnormal time in the target period, and normalizes the result to obtain the first abnormality analysis degree of the current patient at the current time. The embodiment normalizes the addition result of the total time length of the target period, the drainage abnormality change value, the average of the lag time of all physiological indexes, the average of the deviation change degree of all physiological indexes, and the number of types of physiological indexes corresponding to the reference abnormal time in the target period by the norm normalization function. The method for fitting a straight line is a known technology and will not be described in detail. It should be noted that if the current time is not a drainage abnormality time, the first abnormality analysis degree of the current patient at the current time is 0 by default.

[0037] In an implementable manner of the embodiment, the deviation degree obtaining method is: for any time point and any physiological index of the current patient in the drainage process, obtaining the mean value of the physiological index of all preset similar historical patients of the current patient at the time point as the target analysis value of the physiological index; and then normalizing the absolute value of the difference between the physiological index of the current patient at the time point and the target analysis value to obtain the deviation degree of the physiological index of the current patient at the time point.

[0038] In an implementable manner of the embodiment, the reference abnormal time point obtaining method is: when the deviation degree is greater than the preset deviation degree threshold, the corresponding time point is taken as the reference abnormal time point of the physiological index of the current patient. In the embodiment, the preset deviation degree threshold is set to 0.5, and the implementer can set the size of the preset deviation degree threshold according to the actual situation, which is not limited herein.

[0039] Step S202: obtaining the second abnormal analysis degree of the current patient at the current time point according to the occurrence of the drainage abnormal period of the current patient up to the current time point, the overall abnormality of various physiological indexes, and the change of the interval length between adjacent drainage abnormal periods.

[0040] The greater the second abnormal analysis degree is, the more likely it is that the current patient has a real drainage abnormality of active bleeding up to the current time point, and the more necessary it is to give a warning to the drainage sudden increase abnormality of the current patient at the current time point.

[0041] In an implementable manner of the embodiment, the second abnormal analysis degree obtaining method is: obtaining the interval length between any two adjacent drainage abnormal periods as the first length; fitting the first lengths into a curve according to the time sequence as the interval length curve; obtaining the tangent slope of each first length on the interval length curve as the change trend analysis value of each first length; obtaining the ratio of the number of change trend analysis values less than 0 to the total number of change trend analysis values as the first value; the greater the first value is, the more frequent the drainage abnormal period of the current patient appears up to the current time point; when all change trend analysis values less than 0 are smaller, it is more likely that the drainage abnormal period of the current patient appears more frequently; and then the embodiment takes the product of the first value and the inverse of the mean value of all change trend analysis values less than 0 as the abnormal shortening index of the current patient; the greater the abnormal shortening index is, the greater the degree of real drainage abnormality of the current patient up to the current time point is; Further normalize the result of the number of times of the drainage abnormality period occurring before the current time as the cumulative drainage abnormality analysis value of the current patient at the current time; the greater the cumulative drainage abnormality analysis value, the more frequent the drainage abnormality period of the current patient, and the greater the degree of real drainage abnormality of the current patient up to the current time; The ratio of the number of reference abnormal time points of each physiological indicator of the current patient to the total number of all time points corresponding to the current patient up to the current time is taken as the cumulative reference abnormality analysis value of each physiological indicator of the current patient at the current time; the greater the cumulative reference abnormality analysis value, the more unstable the corresponding physiological indicator of the current patient up to the current time, the more abnormal, and indirectly reflects the greater the degree of real drainage abnormality of the current patient up to the current time; In order to reflect the accumulated real drainage abnormality of the current patient up to the current time as a whole, the embodiment adds the mean value of the cumulative reference abnormality analysis value, the abnormality shortening index, and the cumulative drainage abnormality analysis value, and then normalizes the result as the second abnormality analysis degree of the current patient at the current time. The embodiment normalizes the addition result of the mean value of the cumulative reference abnormality analysis value, the abnormality shortening index, and the cumulative drainage abnormality analysis value by the norm normalization function.

[0042] Step S203: Add the first abnormality analysis degree and the second abnormality analysis degree, and then normalize the result as the warning degree of the current patient at the current time.

[0043] The greater the first abnormality analysis degree and the second abnormality analysis degree, the more need to warn the drainage surge abnormality of the current patient at the current time, and then the embodiment normalizes the result of adding the first abnormality analysis degree and the second abnormality analysis degree as the warning degree of the current patient at the current time. The embodiment normalizes the addition result of the first abnormality analysis degree and the second abnormality analysis degree by the norm normalization function.

[0044] The data processing module 40 is configured to warn the drainage surge abnormality of the current patient at the current time based on the warning degree.

[0045] Specifically, the greater the warning degree, the more likely the drainage surge abnormality of the current patient at the current time is a real drainage abnormality of active bleeding, and the more need to be warned, therefore, the embodiment warns the drainage surge abnormality of the current patient at the current time based on the warning degree.

[0046] Preferably, in an implementable manner of the present embodiment, the method for early warning of the drainage surge anomaly of the current patient at the current time based on the early warning degree is: when the early warning degree is greater than the preset early warning degree threshold, the drainage surge anomaly of the current patient at the current time is early warned, that is, the drainage device issues an early warning sound, timely reminding the staff to handle the drainage abnormality of the current patient, effectively reducing the adverse effects of the drainage abnormality on the current patient. The preset early warning degree threshold is set to 0.5 in the present embodiment, and the implementer can set the size of the preset early warning degree threshold according to the actual situation, which is not limited here.

[0047] In summary, the present embodiment acquires drainage volume data and various physiological indicators in real time; determines the drainage abnormality period according to the difference between the drainage volume data change of the current patient and the preset similar historical patient, and the drainage volume data change of the current patient; and early warns the drainage surge anomaly of the current patient at the current time according to the length of the drainage abnormality period in which the current patient is located, the change of the drainage abnormality degree in the drainage abnormality period, and the abnormality of the physiological indicators, as well as the occurrence of the drainage abnormality period of the current patient up to the current time, the overall abnormality of the physiological indicators, and the change of the adjacent drainage abnormality period interval length. The present application accurately early warns the drainage surge anomaly of the current patient, effectively reducing the harm of the drainage abnormality.

[0048] Embodiment 2: The present application also provides an artificial intelligence-based drainage fluid surge rapid early warning device. The device includes a memory and a processor, wherein the memory stores executable program code, and the processor is configured to call and execute the executable program code to execute the artificial intelligence-based drainage fluid surge rapid early warning system provided by the present application. The device can be a chip, a component or a module. The chip can include a processor and a memory connected thereto. When the processor calls and executes the instructions, the chip can execute the artificial intelligence-based drainage fluid surge rapid early warning system provided by the above-mentioned embodiments.

[0049] In addition, the present application also protects a computer device, please refer to Figure 3 The computer device includes a memory 401, a processor 402, and a computer program 403 stored in the memory 401 and running on the processor 402. When the processor 402 executes the computer program 403, the computer device can execute any artificial intelligence-based drainage fluid surge rapid early warning system introduced above.

[0050] Embodiment 3: The application further provides a computer readable storage medium, which stores computer program codes, and when the computer program codes are run on a computer, the computer is caused to execute the related method steps to realize the artificial intelligence-based rapid early warning system for sudden increase of drainage liquid provided by the above embodiments.

[0051] Embodiment 4: The application further provides a computer program product, which, when run on a computer, causes the computer to execute the above related steps to realize the artificial intelligence-based rapid early warning system for sudden increase of drainage liquid provided by the above embodiments.

[0052] Among them, the device, computer readable storage medium, computer program product or chip provided by the embodiment can be used to execute the corresponding method provided above, so the beneficial effects that can be achieved are referable to the beneficial effects in the corresponding method provided above, which will not be described here.

[0053] It should be noted that the above-mentioned sequence of the embodiments of the application is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are possible or can be advantageous.

[0054] Each of the embodiments in the specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the difference from other embodiments.

Claims

1. An artificial intelligence-based rapid early warning system for sudden increases in drainage fluid, characterized in that, The system includes: The data acquisition module is used to acquire patients' drainage volume data and various physiological indicators in real time; The abnormal drainage period acquisition module is used to obtain the degree of abnormal drainage at each moment of the current patient based on the difference in drainage volume data changes between the current patient and preset similar historical patients at each same moment during the drainage process, as well as the changes in drainage volume data of the current patient at each moment, and to determine the abnormal drainage period of the current patient. The warning level acquisition module is used to obtain the warning level of the current patient at the current moment based on the duration of the abnormal drainage period, the changes in the degree of abnormal drainage during the abnormal drainage period, the abnormality of various physiological indicators, the occurrence of the abnormal drainage period, the overall abnormality of various physiological indicators, and the changes in the interval between adjacent abnormal drainage periods. The data processing module is used to issue early warnings for sudden increases in drainage in the current patient at the current moment, based on the level of warning.

2. The artificial intelligence-based rapid early warning system for sudden increases in drainage fluid as described in claim 1, characterized in that, The method for obtaining the degree of abnormality in the drainage is as follows: The drainage data of the current patient and each of its preset similar historical patients are arranged in chronological order to obtain the drainage sequence of the corresponding patient; wherein, the initial drainage data in the drainage sequence is not 0, and the drainage data at the same position in all drainage sequences correspond to the same moment in the drainage process; For any moment and any drainage volume sequence of the current patient during the drainage process, the ratio of the drainage volume data corresponding to that moment in the drainage volume sequence to the initial drainage volume data in the drainage volume sequence is used as the drainage growth index of the patient at that moment for that drainage volume sequence. The mean of the drainage growth index of all pre-set similar historical patients of the current patient at this moment is used as the reference value for drainage growth at this moment; The difference between the current patient's drainage growth index at that moment and the drainage growth reference value at that moment is used as the first drainage growth analysis value for the current patient at that moment. The current patient's drainage volume sequence is fitted into a curve, and the slope of the tangent line of the drainage volume data at that moment is obtained on the curve, which is used as the second drainage growth analysis value of the current patient at that moment. The result of adding the first and second drainage growth analysis values ​​and normalizing them is taken as the degree of drainage abnormality of the patient at that time.

3. The artificial intelligence-based rapid early warning system for sudden increases in drainage fluid as described in claim 1, characterized in that, The method for obtaining the abnormal drainage period is as follows: When the degree of drainage abnormality exceeds the preset threshold for drainage abnormality, the corresponding time will be taken as the current time of drainage abnormality for the patient. The continuous time period consisting of at least two consecutive moments of abnormal drainage is taken as the current period of abnormal drainage for the patient.

4. The artificial intelligence-based rapid early warning system for sudden increases in drainage fluid as described in claim 1, characterized in that, The method for obtaining the warning level is as follows: Based on the duration of the abnormal drainage period in which the patient is currently experiencing the current moment, as well as the changes in the degree of abnormal drainage and the abnormality of various physiological indicators during the abnormal drainage period, the first degree of abnormality analysis of the patient at the current moment is obtained. Based on the occurrence of abnormal drainage periods for the current patient up to the current moment, the overall abnormality of various physiological indicators, and the changes in the interval between adjacent abnormal drainage periods, the degree of second abnormality analysis for the current patient at the current moment is obtained. The sum of the first and second abnormality analysis levels, after normalization, is used as the warning level for the current patient at the current moment.

5. The artificial intelligence-based rapid early warning system for sudden increases in drainage fluid as described in claim 4, characterized in that, The method for obtaining the first level of anomaly analysis is as follows: The current period of abnormal drainage experienced by the patient at the current moment is taken as the target period; the end time of the target period is the current moment. The slope of the straight line fitted to the degree of drainage abnormality in the target time period according to the time sequence is used as the value of the change in drainage abnormality of the current patient at the current moment. Based on the magnitude of each physiological indicator of the current patient at each time point, obtain the degree of deviation of each physiological indicator of the current patient at each time point, and determine the reference abnormal time of each physiological indicator of the current patient; For any physiological indicator, the duration between the initial reference abnormality time of the physiological indicator within the target time period and the initial time of the target time period is taken as the lag duration of the physiological indicator. The total number of times from the initial reference abnormal time to the current time within the target time period is taken as the first quantity; the total number of reference abnormal times of the physiological indicator within the target time period is taken as the second quantity; and the ratio of the second quantity to the first quantity is taken as the abnormal persistence analysis value of the physiological indicator. The result of normalizing the sum of the total duration of the target period, the abnormal change value of drainage, the mean of the lag duration of all physiological indicators, the mean of the abnormal persistence analysis value of all physiological indicators, and the number of types of physiological indicators corresponding to the reference abnormal time within the target period is used as the first degree of abnormality analysis for the current patient at the current time.

6. The artificial intelligence-based rapid early warning system for sudden increases in drainage fluid as described in claim 5, characterized in that, The method for obtaining the degree of deviation is as follows: For any moment and any physiological indicator of the current patient during the drainage process, obtain the mean value of the physiological indicator of all preset similar historical patients at that moment, and use it as the target analysis value of the physiological indicator. The result of normalizing the difference between the current patient's physiological indicator at that moment and the target analysis value is taken as the degree of deviation of the current patient's physiological indicator at that moment.

7. The artificial intelligence-based rapid early warning system for sudden increases in drainage fluid as described in claim 6, characterized in that, The method for obtaining the reference anomaly time is as follows: When the deviation exceeds the preset deviation threshold, the corresponding time will be used as the reference abnormal time for the current patient's physiological indicator.

8. The artificial intelligence-based rapid early warning system for sudden increases in drainage fluid as described in claim 4, characterized in that, The method for obtaining the second level of anomaly analysis is as follows: The interval between any two adjacent abnormal drainage periods is taken as the first duration. The first duration is fitted into a curve according to the time sequence to serve as the interval duration curve; the slope of the tangent line for each first duration on the interval duration curve is obtained as the value of the change trend analysis for each first duration. The ratio of the number of trend analysis values ​​less than 0 to the total number of trend analysis values ​​is taken as the first value; The product of the first value and the negative of the mean of all trend analysis values ​​less than 0 is used as the abnormal shortening index for the current patient. The result of normalizing the number of times the abnormal drainage period occurred up to the current time is used as the cumulative abnormal drainage analysis value for the current patient at the current time. The ratio of the number of reference abnormal times for each physiological indicator of the current patient to the total number of times corresponding to the current patient up to the current time is used as the cumulative reference abnormality analysis value for each physiological indicator of the current patient at the current time. The result of normalizing the sum of the mean of the cumulative reference abnormality analysis values, the abnormal shortening index, and the cumulative drainage abnormality analysis values ​​is used as the second degree of abnormality analysis for the current patient at the current moment.

9. The artificial intelligence-based rapid early warning system for sudden increases in drainage fluid as described in claim 1, characterized in that, The method for issuing early warnings based on the level of warning for sudden increases in drainage abnormalities in the current patient at the current moment is as follows: When the warning level exceeds the preset warning level threshold, an early warning is issued for the sudden increase in drainage abnormality of the current patient at the current moment.

10. The artificial intelligence-based rapid early warning system for sudden increases in drainage fluid as described in claim 1, characterized in that, The method for obtaining the preset patients with similar historical data is as follows: A preset number of historical patients who share the same disease type, gender, surgical and drainage treatment plan, and age difference within a first preset range as the current patient will be considered as the current patient's preset similar historical patients.

Images