Acute kidney injury detection method and system based on missing data

Abstract

The invention relates to the technical field of medical diagnosis, in particular to an acute kidney injury detection method and system based on missing data. The method comprises the following steps: acquiring the preoperative creatinine anomaly according to the difference of serum creatinine data of a patient to be detected and a similar patient in a preoperative analysis period, and acquiring the postoperative creatinine initial anomaly according to the change rate and the prominence degree of the serum creatinine data at each moment in the postoperative analysis period of the patient to be detected; based on the difference between the preoperative creatinine abnormity degree and the postoperative creatinine initial abnormity degree and the difference between the blood loss amount and the blood transfusion amount, the final abnormity degree of the postoperative creatinine is obtained, and acute kidney injury detection is conducted on the to-be-detected patient in combination with the preoperative creatinine abnormity degree and the total urine amount in the postoperative analysis period. According to the change condition of the abnormal degree of the serum creatinine before and after the operation and the difference between the blood loss amount and the blood transfusion amount, the influence of extracorporeal circulation on the abnormal degree of the serum creatinine after the operation is reduced, and the accuracy of acute kidney injury detection of the patient to be detected is improved by combining the abnormal degree of the creatinine before the operation and the urine amount.

Description

Technical Field

[0001] The present invention relates to the field of medical diagnosis technology, and in particular to a method and system for detecting acute kidney injury based on missing data. Background Art

[0002] Acute Kidney Injury (AKI) refers to a clinical syndrome caused by a sudden decline in renal function in a short period of time. During cardiothoracic surgery, cardiopulmonary bypass (CPB) technology is often used to maintain the patient's vital signs when the heart stops beating, but the hemodynamic changes, continuous blood anticoagulation, and tissue ischemia-reperfusion injury caused by CPB technology can induce a variety of postoperative complications; acute kidney injury is a common and serious complication after cardiothoracic surgery. Therefore, acute kidney injury detection in patients after cardiothoracic surgery can help ensure the patient's life safety.

[0003] Existing methods usually determine acute kidney injury based on changes in serum creatinine levels and urine volume after cardiothoracic surgery. However, since extracorporeal circulation during cardiothoracic surgery may cause changes in the patient's blood volume, the postoperative serum creatinine concentration cannot accurately reflect the patient's kidney injury, which in turn leads to low accuracy in acute kidney injury detection. Summary of the invention

[0004] In order to solve the technical problem that the serum creatinine concentration cannot accurately reflect the kidney injury due to extracorporeal circulation during surgery, resulting in low accuracy of acute kidney injury detection, the purpose of the present invention is to provide an acute kidney injury detection method and system based on missing data, and the technical scheme adopted is as follows: In a first aspect, an embodiment of the present invention provides a method for detecting acute kidney injury based on missing data, the method comprising: Obtain the serum creatinine data of the patient to be tested and similar patients at each moment in the preoperative analysis period, as well as the intraoperative blood loss, blood transfusion volume, total urine volume and serum creatinine data of the patient to be tested at each moment in the postoperative analysis period; According to the difference in serum creatinine data between the patient to be tested and similar patients during the preoperative analysis period, the preoperative creatinine abnormality is obtained; according to the change rate of the serum creatinine data of the patient to be tested at each moment during the postoperative analysis period, and the position at each moment and the prominence of the serum creatinine data, the postoperative creatinine initial abnormality is obtained; According to the difference between the preoperative creatinine abnormality and the postoperative creatinine initial abnormality, and the difference between the blood loss and blood transfusion amount of the patient to be tested during the operation, obtaining the final postoperative creatinine abnormality; The patient to be tested is tested for acute kidney injury based on the preoperative creatinine abnormality, the final postoperative creatinine abnormality and the total urine volume of the patient to be tested during the postoperative analysis period.

[0005] Furthermore, obtaining the abnormal degree of creatinine before surgery includes: The patients to be tested and similar patients are recorded as analysis patients, and the mean and variance of the serum creatinine data of each analysis patient at all times during the preoperative analysis period are recorded as the creatinine overall value and creatinine discrete value of the corresponding analysis patient respectively; Obtaining a creatinine global difference value between the patient to be tested and each similar patient according to a difference between the creatinine global value of the patient to be tested and each similar patient and a difference between the creatinine discrete values; The cumulative sum of the overall creatinine difference values ​​between the patient to be tested and all similar patients is negatively correlated and normalized to obtain the preoperative creatinine abnormality.

[0006] Furthermore, obtaining the initial abnormality of creatinine after surgery includes: The result of normalizing the difference between the serum creatinine data of the patient at each moment in the postoperative analysis period and the previous moment and the ratio of the time interval are used as the abnormal index of the serum creatinine data at each moment in the postoperative analysis period; According to the prominence of the serum creatinine data of the patient to be tested at each moment in the postoperative analysis period, the time interval between each moment and its adjacent previous moment, and the position of each moment in the postoperative analysis period, the creatinine weight of each moment in the postoperative analysis period is obtained; The abnormal index of the serum creatinine data at all times in the postoperative analysis period is weighted and summed according to the creatinine weight, and the result of the weighted summation is normalized to obtain the initial abnormal degree of postoperative creatinine.

[0007] Furthermore, the step of obtaining the creatinine weight at each moment in the postoperative analysis period includes: Arranging the serum creatinine data of the patient to be tested at all times in the local preset time period at each time in the postoperative analysis period in time sequence to obtain a local creatinine sequence at each time in the postoperative analysis period; obtaining a first-order difference sequence of the local creatinine sequence, normalizing the mean of all elements in the first-order difference sequence, and obtaining a prominent index of the serum creatinine data at each time in the postoperative analysis period; The time interval between each moment in the postoperative analysis period and its adjacent previous moment is used as the time error index of each moment in the postoperative analysis period; The time interval between the start time of the postoperative analysis period and each moment is taken as the postoperative duration of each moment in the postoperative analysis period; According to the prominent index, the time error index and the postoperative duration, the creatinine weight at each moment in the postoperative analysis period is obtained; the prominent index, the time error index and the postoperative duration are all positively correlated with the creatinine weight.

[0008] Furthermore, obtaining the final abnormal degree of creatinine after surgery includes: Normalizing the difference between the initial abnormal degree of creatinine after surgery and the abnormal degree of creatinine before surgery to obtain the abnormal change value of creatinine before and after surgery; The difference between the blood transfusion volume and blood loss volume of the patient to be tested during the operation is normalized to obtain the extracorporeal circulation impact value; According to the abnormal change value of creatinine before and after and the extracorporeal circulation influence value, the final abnormal degree of creatinine after surgery is obtained; the abnormal change value of creatinine before and after and the final abnormal degree of creatinine after surgery are positively correlated, and the extracorporeal circulation influence value and the final abnormal degree of creatinine after surgery are negatively correlated.

[0009] Furthermore, the acute kidney injury detection of the patient to be tested is performed based on the preoperative creatinine abnormality, the postoperative final creatinine abnormality and the total urine volume of the patient to be tested during the postoperative analysis period, including: Performing negative correlation mapping on the total urine volume of the patient to be tested during the postoperative analysis period, normalizing the product of the preoperative creatinine abnormality, the postoperative creatinine final abnormality and the mapping result to obtain a renal injury risk index; Based on the renal injury risk indicator, the patient to be tested is tested for acute renal injury.

[0010] Furthermore, obtaining the overall creatinine difference value between the patient to be tested and each similar patient according to the difference between the overall creatinine value of the patient to be tested and each similar patient and the difference between the discrete values ​​of creatinine includes: The product of the absolute value of the difference between the overall creatinine value of the patient to be tested and each similar patient and the absolute value of the difference between the discrete values ​​of creatinine is taken as the overall creatinine difference value between the patient to be tested and each similar patient.

[0011] Furthermore, the duration of the local preset time period is shorter than the duration of the postoperative analysis period.

[0012] Furthermore, the duration of the pre-operative analysis period is shorter than the duration of the post-operative analysis period.

[0013] In a second aspect, another embodiment of the present invention provides an acute kidney injury detection system based on missing data, the system comprising: A data acquisition module is used to obtain the serum creatinine data of the patient to be tested and similar patients at each moment in the preoperative analysis period, as well as the intraoperative blood loss, blood transfusion volume, total urine volume and serum creatinine data of the patient to be tested at each moment in the postoperative analysis period; The creatinine initial abnormality analysis module is used to obtain the preoperative creatinine abnormality according to the difference in serum creatinine data between the patient to be tested and similar patients during the preoperative analysis period; and to obtain the postoperative creatinine initial abnormality according to the change rate of the serum creatinine data of the patient to be tested at each moment during the postoperative analysis period, as well as the position at each moment and the prominence of the serum creatinine data; The creatinine final abnormality analysis module is used to obtain the final abnormality of creatinine after surgery according to the difference between the preoperative creatinine abnormality and the postoperative initial creatinine abnormality, and the difference between the blood loss and blood transfusion amount of the patient to be tested during the operation; The renal injury detection module is used to perform acute renal injury detection on the patient to be tested based on the preoperative creatinine abnormality, the final postoperative creatinine abnormality and the total urine volume of the patient to be tested during the postoperative analysis period.

[0014] The present invention has the following beneficial effects: In the embodiment of the present invention, the preoperative creatinine abnormality analyzes the risk of acute kidney injury after surgery in the patient to be tested from a preoperative perspective; the change rate of serum creatinine data at each moment in the postoperative analysis period reflects the abnormality of serum creatinine data, and the position of each moment and the prominence of serum creatinine data show the importance of serum creatinine data at each moment to the analysis of the abnormality of postoperative serum creatinine data. The overall abnormality of postoperative serum creatinine data is analyzed in combination with the above factors to obtain the initial abnormality of postoperative creatinine; compared with simply analyzing the abnormality based on the postoperative serum creatinine data of the patient to be tested, the difference between the preoperative creatinine abnormality and the initial abnormality of postoperative creatinine can better highlight the abnormality of postoperative serum creatinine data, and at the same time, combined with the difference in blood loss and blood transfusion caused by intraoperative extracorporeal circulation, the analysis is performed to reduce the influence of the change in patient blood volume caused by extracorporeal circulation on the abnormality of postoperative serum creatinine data, so that the final abnormality of postoperative creatinine obtained more accurately reflects the risk of acute kidney injury after surgery in the patient to be tested. This protocol analyzes the abnormality of serum creatinine data of the patients to be tested from both preoperative and postoperative perspectives, taking into account the risk of acute kidney injury in the patients to be tested after surgery. The final abnormality of creatinine after surgery reduces the impact of extracorporeal circulation on abnormal analysis. Combined with the total urine volume that reflects acute kidney injury, this protocol greatly improves the accuracy of acute kidney injury detection in the patients to be tested. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 A flowchart of a method for detecting acute kidney injury based on missing data provided by one embodiment of the present invention; Figure 2 A flowchart of the steps of a method for obtaining the initial abnormality of creatinine after surgery provided by one embodiment of the present invention; Figure 3 A system structure diagram of an acute kidney injury detection system based on missing data provided by one embodiment of the present invention; Figure 4 A schematic diagram of a computer device for an acute kidney injury detection device based on missing data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the acute kidney injury detection method and system based on missing data proposed by the present invention, its specific implementation, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

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

[0019] The specific scheme of the method and system for detecting acute kidney injury based on missing data provided by the present invention is described in detail below with reference to the accompanying drawings.

[0020] Embodiment 1: The present invention proposes a method for detecting acute kidney injury based on missing data, please refer to Figure 1 , which shows a flowchart of a method for detecting acute kidney injury based on missing data provided by an embodiment of the present invention, the method comprising: Step S1: Obtain serum creatinine data of the patient to be tested and similar patients at each moment in the preoperative analysis period, as well as blood loss and blood transfusion volume during the operation and total urine volume and serum creatinine data at each moment in the postoperative analysis period of the patient to be tested.

[0021] Before the patient undergoes cardiothoracic surgery, the basic information of the patient is collected; the basic information of patients who undergo cardiothoracic surgery and develop acute kidney injury after surgery is selected from the hospital's medical record database, and these patients are recorded as candidate patients. The attributes of the basic information include: age, gender, years of smoking, years of drinking, symptoms, and type of surgery; age, years of smoking, and years of drinking are numerical data; gender, symptoms, and type of surgery are text data.

[0022] In order to analyze the risk of acute kidney injury in patients undergoing cardiothoracic surgery, patients with similar basic information to the patients to be tested were selected as similar patients of the patients to be tested. The specific method for obtaining similar patients is as follows: The patient to be tested and the patient to be selected are recorded as target patients, and the difference between each attribute of the patient to be tested and the patient to be selected is obtained; if the basic information of each attribute of the target patient is text data, the basic information of each attribute of the target patient is converted into a word vector, and the DTW value of the word vector of each attribute of the patient to be tested and the patient to be selected is calculated as the difference of the corresponding attribute; if the basic information of each attribute of the target patient is numerical data, the absolute value of the difference between the basic information of each attribute of the patient to be tested and the patient to be selected is used as the difference of the corresponding attribute; the cumulative sum of the difference of all attributes of the patient to be tested and the patient to be selected is normalized to obtain the overall difference between the patient to be tested and the patient to be selected. Among them, the dynamic time warping algorithm (Dynamic Time Warping, DTW) is a well-known technology for those skilled in the art and will not be repeated here.

[0023] Since the smaller the overall difference, the higher the similarity between the basic information of the patient to be tested and the patient to be selected, the patient to be selected corresponding to the overall difference less than the preset similarity threshold is recorded as a similar patient to the patient to be tested.

[0024] It should be noted that the embodiment of the present invention uses the bag-of-words model to convert text data into word vectors, the normalization method is the Norm function, and the preset similarity threshold is an empirical value of 0.3, which can be set by the implementer according to the specific situation; other embodiments can also use TF-IDF and Word2Vec and other technologies to convert word vectors, function conversion, maximum and minimum normalization and other normalization methods. Among them, the bag-of-words model is a well-known technology for those skilled in the art and will not be repeated here.

[0025] Serum creatinine level is an important indicator for measuring renal function. Creatinine is a metabolite produced by muscle metabolism and is mainly excreted through the kidneys. The creatinine level in plasma reflects the filtration function of the kidneys. The 24 hours before and 48 hours after the patient's cardiothoracic surgery were recorded as the preoperative analysis period and the postoperative analysis period respectively; the patient to be tested and similar patients were recorded as analysis patients, and the serum creatinine concentration of the blood samples of the analysis patients was randomly collected N1 times during the preoperative analysis period using a colorimetric analysis instrument, and recorded as the serum creatinine data of the analysis patients at each moment during the preoperative analysis period. The serum creatinine concentration of the blood samples of the patient to be tested was randomly collected N2 times during the postoperative analysis period using a colorimetric analysis instrument, and recorded as the serum creatinine data at each moment during the postoperative analysis period.

[0026] It should be noted that since the postoperative analysis period is more important than the serum creatinine data in the preoperative analysis period for the detection of acute kidney injury in the patient to be tested, the length of the preoperative analysis period is shorter than the length of the postoperative analysis period, and N1 is less than N2, and both N1 and N2 are positive integers; among them, N1 takes the empirical value of 3, and N2 takes the empirical value of 50, which can be set by the implementer according to the specific situation. Because the serum creatinine data in the postoperative analysis period is randomly collected, there are differences in the time intervals between two adjacent moments in the postoperative analysis period.

[0027] During the cardiothoracic surgery, the amount of blood transfused into the patient's body by extracorporeal circulation is measured and recorded as the intraoperative blood transfusion volume, and the amount of blood loss is measured and recorded as the intraoperative blood loss volume; the total urine volume of the patient during the postoperative analysis period is collected.

[0028] Step S2: According to the difference in serum creatinine data between the patient to be tested and similar patients during the preoperative analysis period, the preoperative creatinine abnormality is obtained; according to the change rate of the serum creatinine data of the patient to be tested at each moment during the postoperative analysis period, as well as the position at each moment and the prominence of the serum creatinine data, the postoperative initial creatinine abnormality is obtained.

[0029] Similar patients of the patient to be tested suffer from acute kidney injury after undergoing cardiothoracic surgery. The difference in serum creatinine data between the patient to be tested and the similar patients during the preoperative analysis period reflects the risk of the patient to be tested suffering from acute kidney injury after surgery, thereby reflecting the abnormal degree of serum creatinine data of the patient to be tested during the preoperative analysis period, and obtaining the preoperative creatinine abnormality.

[0030] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the preoperative creatinine abnormality includes: recording the patient to be tested and similar patients as analysis patients, and recording the mean and variance of the serum creatinine data of each analysis patient at all times in the preoperative analysis period as the corresponding analysis patient's overall creatinine value and creatinine discrete value, respectively; obtaining the overall creatinine difference value between the patient to be tested and each similar patient based on the difference in the overall creatinine value and the difference in the creatinine discrete value between the patient to be tested and each similar patient; negatively correlating the cumulative sum of the overall creatinine difference values ​​between the patient to be tested and all similar patients and normalizing them to obtain the preoperative creatinine abnormality.

[0031] The overall creatinine value reflects the overall level of serum creatinine data of the analyzed patient during the preoperative analysis period, and the discrete creatinine value reflects the degree of dispersion of serum creatinine data of the analyzed patient during the preoperative analysis period. In this embodiment, the difference in serum creatinine data between the patient to be tested and similar patients during the preoperative analysis period is measured by the overall creatinine value and the discrete creatinine value. The specific difference is calculated by multiplying the absolute value of the difference between the overall creatinine value of the patient to be tested and each similar patient by the absolute value of the difference between the discrete creatinine values ​​as the overall creatinine difference value between the patient to be tested and each similar patient.

[0032] If the overall difference in creatinine between the patient to be tested and all similar patients is smaller, it means that the serum creatinine level of the patient to be tested before surgery is closer to the serum creatinine level of similar patients with postoperative acute kidney injury before surgery, which makes the risk of postoperative acute kidney injury in the patient to be tested higher. The greater the abnormality of the serum creatinine data of the patient to be tested during the preoperative analysis period, the greater the abnormality of the preoperative creatinine.

[0033] In a specific implementation of the embodiment of the present invention, the abnormal degree of creatinine before surgery is expressed by the formula: In the formula, The abnormal degree of creatinine before surgery; is the overall creatinine value of the patient to be tested; is the overall creatinine value of the ath similar patient to the patient to be tested; LS is the discrete value of creatinine of the patient to be tested; is the discrete value of creatinine of the ath similar patient to the patient to be tested; is the overall difference in creatinine between the patient to be tested and the ath similar patient; is the absolute value function; exp is the exponential function with the natural constant as the base. It should be noted that the present embodiment uses the exponential function with the natural constant as the base to Perform negative correlation and normalization.

[0034] The change rate of serum creatinine data at each moment in the postoperative analysis period reflects the abnormal degree of serum creatinine data at each moment. The position of each moment and the prominence of serum creatinine data show the importance of serum creatinine data at each moment to the analysis of the abnormal degree of postoperative serum creatinine data. The overall abnormal degree of postoperative serum creatinine data is analyzed in combination with the above factors to obtain the initial abnormal degree of postoperative creatinine.

[0035] See also Figure 2 , which shows a flowchart of the steps of a method for obtaining the initial abnormality of creatinine after surgery provided by an embodiment of the present invention, the method comprising: Step S210: normalizing the difference between the serum creatinine data of the patient at each moment in the postoperative analysis period and the previous moment and the time interval ratio as an abnormality index of the serum creatinine data at each moment in the postoperative analysis period.

[0036] It is known that when kidney function is impaired, the filtration function of the kidney will decrease, leading to an increase in plasma creatinine levels.

[0037] If the increase in the serum creatinine data of the patient to be tested at each moment during the postoperative analysis period is greater than that of the previous moment, it means that the renal physiological function of the patient to be tested is declining faster at each moment, and the greater the abnormality of the serum creatinine data at each moment during the postoperative analysis period, the greater the abnormal index of the serum creatinine data at each moment.

[0038] Step S220: Obtain the creatinine weight of each moment in the postoperative analysis period according to the prominence of the serum creatinine data of the patient at each moment in the postoperative analysis period, the time interval between each moment and its adjacent previous moment, and the position of each moment in the postoperative analysis period.

[0039] Preferably, in some possible implementation modes of the embodiments of the present invention, the method for obtaining the creatinine weight includes: arranging the serum creatinine data of the patient to be tested at all times within the local preset time period at each time within the postoperative analysis period in time series to obtain a local creatinine sequence at each time within the postoperative analysis period; obtaining a first-order difference sequence of the local creatinine sequence, normalizing the means of all elements in the first-order difference sequence to obtain a prominent index of the serum creatinine data at each time within the postoperative analysis period; taking the time interval between each time within the postoperative analysis period and its adjacent previous time as the time error index at each time within the postoperative analysis period; taking the time interval between the start time of the postoperative analysis period and each time as the postoperative duration at each time within the postoperative analysis period; and obtaining the creatinine weight at each time within the postoperative analysis period according to the prominent index, the time error index and the postoperative duration.

[0040] If the patient's serum creatinine concentration level rises or stabilizes at a high level in the early stage after cardiothoracic surgery, the patient is more likely to suffer from acute kidney injury after the operation, and the degree of serum creatinine abnormality is greater; because extracorporeal circulation during surgery may cause the patient's blood volume to increase, the serum creatinine concentration level will drop sharply after the operation, and then gradually return to normal levels. The above-mentioned sharp drop in serum creatinine concentration level is a normal phenomenon, the degree of abnormality is small, and it needs to be excluded; after the serum creatinine concentration returns to normal levels, the serum creatinine concentration level still rises, which is also considered to be abnormal.

[0041] Therefore, if the prominence index is closer to the constant 1, it means that the serum creatinine data in the local preset time period at each moment in the postoperative analysis period has an obvious upward trend, and the abnormality of the serum creatinine data at this moment is more obvious. The serum creatinine data at this moment is more important for the detection of acute kidney injury, and the weight of the serum creatinine data at this moment should be greater. If the prominence index is closer to the constant 0, it means that the serum creatinine data in the local preset time period at each moment in the postoperative analysis period has an obvious downward trend, which is a normal phenomenon. The importance of the serum creatinine data at this moment to the detection of acute kidney injury is lower, and the weight of the serum creatinine data at this moment should be smaller.

[0042] It should be noted that in the embodiment of the present invention, the Norm function is used for normalization processing, and other normalization methods may also be selected, such as function conversion, maximum and minimum normalization, etc., which are not limited here; the local preset time period of each moment in the postoperative analysis period includes three moments, and the moment is located in the middle of the three moments; the length of the local preset time period is less than the length of the postoperative analysis period. In this embodiment, the prominence index of the moment without the local preset time period in the postoperative analysis period is directly set to a constant of 0.

[0043] If the time error index is smaller, the time interval between each moment in the postoperative analysis period and its adjacent previous moment is shorter, and the possibility of random deviation in serum creatinine data due to factors such as measurement technology, equipment accuracy and changes in the sample collection process is greater, resulting in a lower accuracy rate of the abnormal index of the serum creatinine data at each moment in step S210. In order to ensure the accuracy of the abnormal degree analysis of the serum creatinine data in the postoperative analysis period, the weight of the serum creatinine data at each moment in the postoperative analysis period should be smaller.

[0044] The postoperative time represents the position of each moment in the postoperative analysis period; if the postoperative time is longer, it means that each moment in the postoperative analysis period is farther away from the end of the operation, the patient's condition is more stable, and the serum creatinine data at that moment has more reference value, then the weight of the serum creatinine data at each moment should be greater.

[0045] In summary, the prominence index, time error index and postoperative duration at each moment in the postoperative analysis period are positively correlated with the creatinine weight. In an embodiment of the present invention, the product of the prominence index, time error index and postoperative duration at each moment in the postoperative analysis period is normalized to obtain the creatinine weight at the corresponding moment. The embodiment of the present invention uses the Norm function for normalization.

[0046] Step S230: performing weighted summation on the abnormal indicators of serum creatinine data at all times within the postoperative analysis period according to the creatinine weight, and normalizing the result of the weighted summation to obtain the initial abnormal degree of postoperative creatinine.

[0047] By analyzing the abnormal index and creatinine weight of serum creatinine data at all times during the postoperative analysis period, the overall abnormality of serum creatinine data after surgery was analyzed to obtain the initial abnormality of creatinine after surgery. The initial abnormality of creatinine after surgery is expressed as: In the formula, is the initial abnormal degree of creatinine after surgery; N2 is the total number of moments in the postoperative analysis period; is the creatinine weight at the n2th moment except the first moment in the postoperative analysis period; It is the abnormal index of serum creatinine data at the n2th moment except the first moment in the postoperative analysis period; Serum creatinine data at the n2th moment except the first moment during the postoperative analysis period; Serum creatinine data at the n2-1st moment except the first moment during the postoperative analysis period; is the time interval between the n2-1th moment and the n2th moment in the postoperative analysis period except the first moment; Norm is the normalization function. It should be noted that the greater the initial abnormality of postoperative creatinine, the greater the overall abnormality of the serum creatinine concentration level of the patient to be tested after cardiothoracic surgery.

[0048] Step S3: According to the difference between the preoperative creatinine abnormality and the initial postoperative creatinine abnormality, and the difference between the blood loss and blood transfusion volume of the patient to be tested during the operation, the final postoperative creatinine abnormality is obtained.

[0049] Compared with analyzing the abnormality only based on the postoperative serum creatinine data of the tested patients, the difference between the preoperative creatinine abnormality and the initial postoperative creatinine abnormality can better highlight the abnormality of the postoperative serum creatinine data. At the same time, combined with the difference in blood loss and blood transfusion caused by extracorporeal circulation during surgery, the final postoperative creatinine abnormality can more accurately reflect the risk of acute kidney injury in the tested patients after surgery.

[0050] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the final abnormality of creatinine after surgery includes: normalizing the difference between the initial abnormality of creatinine after surgery and the abnormality of creatinine before surgery to obtain the abnormal change value of creatinine before and after surgery; normalizing the difference between the amount of blood transfusion and the amount of blood loss of the patient to be tested during surgery to obtain the extracorporeal circulation influence value; and obtaining the final abnormality of creatinine after surgery based on the abnormal change value of creatinine before and after surgery and the extracorporeal circulation influence value.

[0051] The degree of change between the initial postoperative creatinine abnormality and the preoperative creatinine abnormality reflects the degree of abnormality in the postoperative serum creatinine concentration level of the patient to be tested. If the difference between the initial postoperative creatinine abnormality and the preoperative creatinine abnormality, that is, the abnormal change value of creatinine before and after, is greater, it means that the postoperative serum creatinine data is more abnormal than before the operation, and the degree of abnormality of the postoperative serum creatinine data is greater.

[0052] In order to ensure the physiological health of patients during cardiothoracic surgery, extracorporeal circulation usually delivers sufficient blood to patients, that is, the blood transfusion volume is greater than the blood loss, resulting in an increase in the patient's blood volume. The greater the extracorporeal circulation impact value, the greater the patient's blood volume, the greater the impact on the initial abnormal degree of creatinine after surgery, and thus the abnormal creatinine change value before and after has a lower reference value.

[0053] Therefore, the abnormal change value of creatinine before and after surgery is positively correlated with the final abnormal degree of creatinine after surgery, and the impact value of extracorporeal circulation is negatively correlated with the final abnormal degree of creatinine after surgery.

[0054] The final abnormal degree of creatinine after surgery is expressed by the formula: In the formula, is the final abnormal degree of creatinine after surgery; R is the abnormal change value of creatinine before and after; IM is the extracorporeal circulation impact value; is the initial abnormal degree of creatinine after surgery; The abnormal degree of creatinine before surgery; The amount of blood transfused during surgery; The amount of blood transfused during surgery; It is a preset positive number with an empirical value of 0.1. It is used to prevent the denominator from being zero, which makes the fraction meaningless. Norm is the normalization function.

[0055] By adjusting the amount of blood transfusion and blood loss of the patient during surgery, the degree of abnormality of the preoperative and postoperative serum creatinine data is adjusted to reduce the impact of changes in the patient's blood volume due to extracorporeal circulation during surgery on the abnormality of the postoperative serum creatinine data, thereby effectively improving the accuracy of abnormal analysis of the postoperative serum creatinine data of the patient.

[0056] Step S4: Based on the preoperative creatinine abnormality, the final postoperative creatinine abnormality and the total urine volume of the patient during the postoperative analysis period, the patient is tested for acute kidney injury.

[0057] The abnormal degree of pre-operative creatinine and the final abnormal degree of post-operative creatinine reflect the abnormal degree of serum creatinine data of the patient before and after the operation, and further reflect the risk of acute kidney injury in the patient after the operation; decreased urine volume is a key indicator for analyzing acute kidney injury. The above factors are combined to perform acute kidney injury detection on the patient.

[0058] In an embodiment of the present invention, the method for obtaining the renal injury risk index is: negative correlation mapping is performed on the total urine volume of the patient to be tested during the postoperative analysis period, and the product of the preoperative creatinine abnormality, the postoperative creatinine final abnormality and the mapping result is normalized to obtain the renal injury risk index.

[0059] When the preoperative creatinine abnormality and the final postoperative creatinine abnormality are greater, it means that the patient is at higher risk of acute kidney injury after surgery and the renal injury impact coefficient is greater; when the total urine volume is less, the patient is at higher risk of acute kidney injury after surgery. Therefore, the preoperative creatinine abnormality and the final postoperative creatinine abnormality are positively correlated with the renal injury risk index, and the total urine volume is negatively correlated with the renal injury risk index.

[0060] The renal injury risk index is expressed as follows: In the formula, P is the renal injury risk index of the patient to be tested; The abnormal degree of creatinine before surgery; is the final abnormal degree of creatinine after surgery; UV is the total urine volume of the patient to be tested during the postoperative analysis period; exp is an exponential function with a natural constant as the base; Norm is a normalization function. It should be noted that the embodiment of the present invention uses The function performs negative correlation mapping on UV.

[0061] The greater the renal injury risk index, the greater the risk of the patient suffering from acute renal injury after cardiothoracic surgery. Based on the renal injury risk index, the assistant doctor determines whether the patient suffers from acute renal injury after cardiothoracic surgery. During the judgment process, the doctor also needs to combine auxiliary examinations such as renal echocardiography, blood gas analysis and blood electrolyte measurement.

[0062] So far, the present invention is completed.

[0063] Embodiment 2: The present invention proposes an acute kidney injury detection system based on missing data, please refer to Figure 3, which shows a system structure diagram of an acute kidney injury detection system based on missing data provided by an embodiment of the present invention, the system comprising: The data acquisition module 510 is used to obtain the serum creatinine data of the patient to be tested and similar patients at each moment in the preoperative analysis period, as well as the blood loss and blood transfusion volume of the patient to be tested during the operation and the total urine volume and serum creatinine data at each moment in the postoperative analysis period; The creatinine initial abnormality analysis module 520 is used to obtain the preoperative creatinine abnormality according to the difference in serum creatinine data between the patient to be tested and similar patients during the preoperative analysis period; and to obtain the postoperative creatinine initial abnormality according to the change rate of the serum creatinine data of the patient to be tested at each moment during the postoperative analysis period, and the position at each moment and the prominence of the serum creatinine data; The creatinine final abnormality analysis module 530 is used to obtain the final abnormality of creatinine after surgery according to the difference between the abnormality of creatinine before surgery and the initial abnormality of creatinine after surgery, and the difference between the blood loss and blood transfusion amount of the patient to be tested during surgery; The renal injury detection module 540 is used to perform acute renal injury detection on the patient to be tested based on the preoperative creatinine abnormality, the final postoperative creatinine abnormality and the total urine volume of the patient to be tested during the postoperative analysis period.

[0064] It should be noted that: the device provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the acute kidney injury detection system based on missing data and the acute kidney injury detection method based on missing data provided in the above embodiment belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0065] Embodiment 3: Figure 4 A computer device schematic diagram of an acute kidney injury detection device based on missing data provided by an embodiment of the present invention. Figure 4 As shown, the computer device includes: a memory 601, a processor 602, and a computer program 603 stored in the memory 601 and running on the processor 602, wherein when the processor 602 executes the computer program 603, the computer device can execute any one of the acute kidney injury detection methods based on missing data introduced above.

[0066] In addition, an embodiment of the present application also protects a device, which may include a memory and a processor, wherein the memory stores executable program code, and the processor is used to call and execute the executable program code to execute an acute kidney injury detection method based on missing data provided in an embodiment of the present application.

[0067] In this embodiment, the functional modules of the device can be divided according to the above method example. For example, each functional module can be corresponded, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0068] It should be understood that the device provided in this embodiment is used to execute the above-mentioned acute kidney injury detection method based on missing data, and thus can achieve the same effect as the above-mentioned implementation method.

[0069] In the case of an integrated unit, the device may include a processing module and a storage module. When the device is applied to a device, the processing module may be used to control and manage the actions of the device. The storage module may be used to support the device to execute mutual program codes, etc.

[0070] The processing module may be a processor or a controller, which may implement or execute various exemplary logic blocks, modules and circuits included in the disclosure of this application. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc. The storage module may be a memory.

[0071] It should be noted that the sequence of the above embodiments of the present invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0072] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0073] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for detecting acute kidney injury based on missing data, characterized in that: The method includes: Obtain the serum creatinine data of the patient to be tested and similar patients at each moment in the preoperative analysis period, as well as the intraoperative blood loss, blood transfusion volume, total urine volume and serum creatinine data of the patient to be tested at each moment in the postoperative analysis period; According to the difference in serum creatinine data between the patient to be tested and similar patients during the preoperative analysis period, the preoperative creatinine abnormality is obtained; according to the change rate of the serum creatinine data of the patient to be tested at each moment during the postoperative analysis period, and the position at each moment and the prominence of the serum creatinine data, the postoperative creatinine initial abnormality is obtained; According to the difference between the preoperative creatinine abnormality and the postoperative creatinine initial abnormality, and the difference between the blood loss and blood transfusion amount of the patient to be tested during the operation, obtaining the final postoperative creatinine abnormality; Performing an acute kidney injury test on the patient to be tested based on the preoperative creatinine abnormality, the postoperative final creatinine abnormality and the total urine volume of the patient to be tested during the postoperative analysis period; The obtaining of the initial abnormal degree of creatinine after surgery includes: The result of normalizing the difference between the serum creatinine data of the patient at each moment in the postoperative analysis period and the previous moment and the ratio of the time interval are used as the abnormal index of the serum creatinine data at each moment in the postoperative analysis period; According to the prominence of the serum creatinine data of the patient to be tested at each moment in the postoperative analysis period, the time interval between each moment and its adjacent previous moment, and the position of each moment in the postoperative analysis period, the creatinine weight of each moment in the postoperative analysis period is obtained; According to the creatinine weight, the abnormal index of the serum creatinine data at all times in the postoperative analysis period is weighted and summed, and the result of the weighted summation is normalized to obtain the initial abnormal degree of postoperative creatinine; The method of obtaining the final abnormal degree of creatinine after surgery includes: Normalizing the difference between the initial abnormal degree of creatinine after surgery and the abnormal degree of creatinine before surgery to obtain the abnormal change value of creatinine before and after surgery; The difference between the blood transfusion volume and blood loss volume of the patient to be tested during the operation is normalized to obtain the extracorporeal circulation impact value; According to the abnormal change value of creatinine before and after and the extracorporeal circulation influence value, the final abnormal degree of creatinine after surgery is obtained; the abnormal change value of creatinine before and after and the final abnormal degree of creatinine after surgery are positively correlated, and the extracorporeal circulation influence value and the final abnormal degree of creatinine after surgery are negatively correlated.

2. The method for detecting acute kidney injury based on missing data according to claim 1, characterized in that: The obtaining of the abnormal degree of creatinine before surgery includes: The patients to be tested and similar patients are recorded as analysis patients, and the mean and variance of the serum creatinine data of each analysis patient at all times during the preoperative analysis period are recorded as the creatinine overall value and creatinine discrete value of the corresponding analysis patient respectively; Obtaining a creatinine global difference value between the patient to be tested and each similar patient according to a difference between the creatinine global value of the patient to be tested and each similar patient and a difference between the creatinine discrete values; The cumulative sum of the overall creatinine difference values ​​between the patient to be tested and all similar patients is negatively correlated and normalized to obtain the preoperative creatinine abnormality.

3. The method for detecting acute kidney injury based on missing data according to claim 1, characterized in that: The step of obtaining the creatinine weight at each moment in the postoperative analysis period includes: Arranging the serum creatinine data of the patient to be tested at all times in the local preset time period at each time in the postoperative analysis period in time sequence to obtain a local creatinine sequence at each time in the postoperative analysis period; obtaining a first-order difference sequence of the local creatinine sequence, normalizing the mean of all elements in the first-order difference sequence, and obtaining a prominent index of the serum creatinine data at each time in the postoperative analysis period; The time interval between each moment in the postoperative analysis period and its adjacent previous moment is used as the time error index of each moment in the postoperative analysis period; The time interval between the start time of the postoperative analysis period and each moment is taken as the postoperative duration of each moment in the postoperative analysis period; According to the prominent index, the time error index and the postoperative duration, the creatinine weight at each moment in the postoperative analysis period is obtained; the prominent index, the time error index and the postoperative duration are all positively correlated with the creatinine weight.

4. The method for detecting acute kidney injury based on missing data according to claim 1, characterized in that: The method of performing acute kidney injury detection on the patient to be tested based on the preoperative creatinine abnormality, the postoperative final creatinine abnormality and the total urine volume of the patient to be tested during the postoperative analysis period includes: Performing negative correlation mapping on the total urine volume of the patient to be tested during the postoperative analysis period, normalizing the product of the preoperative creatinine abnormality, the postoperative creatinine final abnormality and the mapping result to obtain a renal injury risk index; Based on the renal injury risk indicator, the patient to be tested is tested for acute renal injury.

5. The method for detecting acute kidney injury based on missing data according to claim 2, characterized in that: The step of obtaining the overall creatinine difference value between the patient to be tested and each similar patient according to the difference between the overall creatinine value of the patient to be tested and each similar patient and the difference between the discrete values ​​of creatinine comprises: The product of the absolute value of the difference between the overall creatinine value of the patient to be tested and each similar patient and the absolute value of the difference between the discrete values ​​of creatinine is taken as the overall creatinine difference value between the patient to be tested and each similar patient.

6. The method for detecting acute kidney injury based on missing data according to claim 3, characterized in that: The duration of the local preset time period is shorter than the duration of the postoperative analysis period.

7. The method for detecting acute kidney injury based on missing data according to claim 1, characterized in that: The duration of the pre-operative analysis period is shorter than the duration of the post-operative analysis period.

8. An acute kidney injury detection system based on missing data, characterized in that: The system includes: A data acquisition module is used to obtain the serum creatinine data of the patient to be tested and similar patients at each moment in the preoperative analysis period, as well as the intraoperative blood loss, blood transfusion volume, total urine volume and serum creatinine data of the patient to be tested at each moment in the postoperative analysis period; The creatinine initial abnormality analysis module is used to obtain the preoperative creatinine abnormality according to the difference in serum creatinine data between the patient to be tested and similar patients during the preoperative analysis period; and to obtain the postoperative creatinine initial abnormality according to the change rate of the serum creatinine data of the patient to be tested at each moment during the postoperative analysis period, as well as the position at each moment and the prominence of the serum creatinine data; The creatinine final abnormality analysis module is used to obtain the final abnormality of creatinine after surgery according to the difference between the preoperative creatinine abnormality and the postoperative initial creatinine abnormality, and the difference between the blood loss and blood transfusion amount of the patient to be tested during the operation; A renal injury detection module, used for performing acute renal injury detection on the patient to be tested according to the preoperative creatinine abnormality, the postoperative final creatinine abnormality and the total urine volume of the patient to be tested during the postoperative analysis period; The obtaining of the initial abnormal degree of creatinine after surgery includes: The result of normalizing the difference between the serum creatinine data of the patient at each moment in the postoperative analysis period and the previous moment and the ratio of the time interval are used as the abnormal index of the serum creatinine data at each moment in the postoperative analysis period; According to the prominence of the serum creatinine data of the patient to be tested at each moment in the postoperative analysis period, the time interval between each moment and its adjacent previous moment, and the position of each moment in the postoperative analysis period, the creatinine weight of each moment in the postoperative analysis period is obtained; According to the creatinine weight, the abnormal index of the serum creatinine data at all times in the postoperative analysis period is weighted and summed, and the result of the weighted summation is normalized to obtain the initial abnormal degree of postoperative creatinine; The method of obtaining the final abnormal degree of creatinine after surgery includes: Normalizing the difference between the initial abnormal degree of creatinine after surgery and the abnormal degree of creatinine before surgery to obtain the abnormal change value of creatinine before and after surgery; The difference between the blood transfusion volume and blood loss volume of the patient to be tested during the operation is normalized to obtain the extracorporeal circulation impact value; According to the abnormal change value of creatinine before and after and the extracorporeal circulation influence value, the final abnormal degree of creatinine after surgery is obtained; the abnormal change value of creatinine before and after and the final abnormal degree of creatinine after surgery are positively correlated, and the extracorporeal circulation influence value and the final abnormal degree of creatinine after surgery are negatively correlated.

Images