Concurrent constipation risk prediction method and system for young and middle-aged first cerebral hemorrhage patients

Abstract

The invention, which belongs to the technical field of medical prediction, discloses a concurrent constipation risk prediction method and system for a young and middle-aged first cerebral hemorrhage patient, and belongs to the technical field of medical prediction. The method comprises the steps of obtaining clinical data of the young and middle-aged first cerebral hemorrhage patient, inputting a pre-trained risk prediction model, and outputting a constipation risk prediction result; according to the constipation risk prediction result, drawing a constipation risk prediction result column graph; through a single-factor analysis method, a multiple linear analysis method and a Logistic regression analysis method, clinical data of the young and middle-aged first cerebral hemorrhage patients are analyzed, influence factors of constipation of the young and middle-aged first cerebral hemorrhage patients are determined, the risk prediction model is constructed, and the risk prediction model is used for the risk of constipation of the young and middle-aged first cerebral hemorrhage patients. The problem that in the prior art, a method for predicting the concurrent constipation risk of the young and middle-aged cerebral hemorrhage patients at the first time is lacked is solved.

Description

Technical Field

[0001] The invention relates to a method and system for predicting the risk of constipation in young and middle-aged patients with first-episode cerebral hemorrhage, belonging to the technical field of medical prediction. Background Art

[0002] Cerebral hemorrhage is an acute cerebrovascular accident with the characteristics of sudden onset and rapid progression of the disease. The risk of death and long-term disability in the acute phase is extremely high. According to statistics, hemorrhagic stroke accounted for 27.9% of new strokes worldwide in 2019, and the age group of 15 to 49 years old accounted for more than 23% of the first disease group, showing a clear trend of younger age. The physiological functions of young and middle-aged patients are still in a relatively healthy state, but the occurrence of cerebral hemorrhage not only interrupts their career and life plans, but also may lead to various complications, causing long-term physical and psychological disorders. Therefore, the prevention and care of complications after cerebral hemorrhage are of great significance to the prognosis of young and middle-aged patients with cerebral hemorrhage. Constipation is one of the most common complications in patients with cerebral hemorrhage, usually referring to reduced bowel movement frequency, difficulty in defecation, or dry and hard stools, with an incidence ranging from 40% to 91%. Studies have shown that straining to defecate can easily induce cardiovascular and cerebrovascular diseases, increase the risk of re-bleeding, interrupt the treatment process of the disease and even endanger life, prolong hospitalization time, increase treatment costs, and seriously affect the patient's recovery and quality of life.

[0003] Young and middle-aged people are at the core stage of society and family. Once cerebral hemorrhage occurs, not only may their original life trajectory be interrupted, but they may also experience various physical disabilities, including motor disorders, sensory disorders, and cognitive and language disorders. They are unable to continue to shoulder family responsibilities and return to work, and bear huge economic burdens and emotional distress. Constipation not only causes physical and psychological pain to patients, but also leads to cardiovascular and cerebrovascular diseases, seriously affecting the quality of life of patients, and at the same time bringing a heavy economic burden to families and patients. Therefore, it is particularly important to screen for constipation risks in young and middle-aged patients with first cerebral hemorrhage and take preventive measures as early as possible. At present, reports on constipation in patients with cerebral hemorrhage are particularly limited, and the existing research population is mainly patients with ischemic stroke alone or combined with cerebral hemorrhage. In addition, existing studies have mainly focused on elderly patients, and there are relatively few studies on risk factors for constipation in young and middle-aged patients with first stroke. Young and middle-aged patients have significant differences from the elderly in terms of physiological, psychological, and social background. Elderly patients are usually accompanied by multiple chronic diseases, and their pathophysiological mechanisms are significantly different from those of young and middle-aged patients. In addition, different lifestyles and psychological states also cause young and middle-aged patients to show different risk factors and rehabilitation challenges when suffering from cerebral hemorrhage. Summary of the invention

[0004] The purpose of the present invention is to provide a method for predicting the risk of constipation in young and middle-aged patients with their first cerebral hemorrhage. By using a univariate analysis method, a multiple linear analysis method and a logistic regression analysis method, the clinical data of young and middle-aged patients with their first cerebral hemorrhage are analyzed, the influencing factors of constipation in young and middle-aged patients with their first cerebral hemorrhage are determined, and a risk prediction model is constructed. The risk prediction model is used to accurately output the patient's constipation risk prediction results, and the constipation risk prediction results are presented in the form of an intuitive nomogram, so that doctors and patients can have a clearer understanding of the constipation risk situation.

[0005] In order to solve the above technical problems, the present invention is implemented by adopting the following technical solutions.

[0006] In a first aspect, the present invention provides a method for predicting the risk of constipation in young and middle-aged patients with first-onset cerebral hemorrhage, comprising:

[0007] Obtain clinical data of young and middle-aged patients with first-onset cerebral hemorrhage;

[0008] Inputting the clinical data of the young and middle-aged patients with first-onset cerebral hemorrhage into a pre-trained risk prediction model, and outputting a constipation risk prediction result;

[0009] According to the constipation risk prediction results, a nomogram of constipation risk prediction results was drawn;

[0010] Among them, the clinical data of the middle-aged and young patients with first-onset cerebral hemorrhage were analyzed by univariate analysis, multiple linear analysis and logistic regression analysis, the influencing factors of constipation in middle-aged and young patients with first-onset cerebral hemorrhage were determined, and the risk prediction model was constructed.

[0011] Furthermore, the clinical data of the young and middle-aged patients with first-onset cerebral hemorrhage are analyzed by univariate analysis, multiple linear analysis and logistic regression analysis to determine the influencing factors of constipation in young and middle-aged patients with first-onset cerebral hemorrhage, and to construct the risk prediction model, including:

[0012] A case-control study method was used to obtain historical clinical data of young and middle-aged patients with first-episode cerebral hemorrhage;

[0013] By using the univariate analysis method, the variables in the historical clinical data of the young and middle-aged patients with first-onset cerebral hemorrhage were analyzed to obtain statistically significant variables;

[0014] By using a multi-linear analysis method, a multi-collinearity analysis is performed on the statistically significant variables to obtain statistically significant variables without multi-collinearity between the variables;

[0015] Assigning values ​​to statistically significant variables without multicollinearity between the variables to obtain assigned variables;

[0016] The variables after the assignment were used as independent variables, and whether constipation occurred was used as the dependent variable. Logistic regression analysis was performed using the backward likelihood method to obtain the influencing factors of constipation in young and middle-aged patients with first-episode cerebral hemorrhage.

[0017] A risk prediction model was constructed based on the influencing factors of constipation in young and middle-aged patients with first-episode cerebral hemorrhage.

[0018] Furthermore, the variables in the historical clinical data of the young and middle-aged patients with first-onset cerebral hemorrhage were analyzed by univariate analysis to obtain statistically significant variables, wherein the probability of random errors in the statistically significant variables was less than 5%.

[0019] Furthermore, by means of a multilinear analysis method, multicollinearity analysis is performed on the statistically significant variables to obtain statistically significant variables without multicollinearity between the variables, wherein when the tolerance of the statistically significant variables is greater than 0.2 and the variance inflation factor is less than 5.0, there is no multicollinearity between the statistically significant variables.

[0020] Furthermore, the influencing factors of constipation in the young and middle-aged patients with first-episode cerebral hemorrhage include Glasgow Coma Scale (GCS), whether they have had surgery, whether they have hemiplegia, whether they have used mechanical ventilation, whether they have used sedatives, and whether they have used calcium ion antagonists.

[0021] Furthermore, after the risk prediction model is constructed, internal verification of the risk prediction model is also performed, including:

[0022] The Bootstrap autonomous sampling method was used to sample the historical clinical data of the young and middle-aged patients with first-episode cerebral hemorrhage to obtain an internal validation data set;

[0023] Inputting the internal validation data set into the risk prediction model, and outputting an internal validation constipation risk prediction result;

[0024] Based on the internal validation constipation risk prediction results, the discrimination of the risk prediction model is detected using the area under the receiver operating characteristic curve;

[0025] Based on the internal validation constipation risk prediction results, the predictive performance of the risk prediction model was evaluated using the Hosmer-Lemeshow test and calibration plot.

[0026] Furthermore, based on the internally validated constipation risk prediction results, the area under the receiver operating characteristic curve is used to detect the discrimination of the risk prediction model, wherein the closer the area under the curve value is to 1, the better the discrimination of the risk prediction model.

[0027] Furthermore, based on the internally validated constipation risk prediction results, the predictive performance of the risk prediction model was evaluated using the Hosmer-Lemeshow test and calibration graph, wherein when the predictive performance of the risk prediction model was tested using the Hosmer-Lemeshow test, the significance level was equal to 0.05, and the goodness of fit of the risk prediction model was evaluated.

[0028] Furthermore, after the risk prediction model is internally validated, the risk prediction model is externally validated:

[0029] The clinical data of young and middle-aged patients with first-episode cerebral hemorrhage in the later period of the internal validation dataset were used as the external validation dataset;

[0030] Inputting the external validation data set into the risk prediction model, and outputting an external validation constipation risk prediction result;

[0031] Based on the externally validated constipation risk prediction results, the discrimination of the risk prediction model is detected using the area under the receiver operating characteristic curve;

[0032] The externally validated constipation risk prediction results are compared with the actual constipation risk results to evaluate the fit of the risk prediction model.

[0033] In a second aspect, the present invention provides a system for predicting the risk of constipation in young and middle-aged patients with first-onset cerebral hemorrhage, comprising:

[0034] Data acquisition module, used to obtain clinical data of young and middle-aged patients with first-episode cerebral hemorrhage;

[0035] A risk prediction module, used to input the clinical data of the young and middle-aged patients with first-episode cerebral hemorrhage into a pre-trained risk prediction model;

[0036] A prediction result output module, used to output the constipation risk prediction result;

[0037] A nomogram generation module is used to draw a nomogram of constipation risk prediction results according to the constipation risk prediction results;

[0038] Among them, the influencing factors of constipation in young and middle-aged patients with first-episode cerebral hemorrhage were determined by univariate analysis, multiple linear analysis and logistic regression analysis, and the risk prediction model was constructed.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] 1. The present invention can accurately output the patient's constipation risk prediction results based on the clinical data of young and middle-aged patients with first-onset cerebral hemorrhage by using a pre-trained risk prediction model, and present the constipation risk prediction results in the form of an intuitive nomogram, so that both doctors and patients can understand the constipation risk more clearly. It not only improves the accuracy of doctors' assessment of patients' constipation risks, but also provides a scientific basis for formulating personalized prevention and intervention measures, thereby helping to reduce the risk of constipation in young and middle-aged patients with first-onset cerebral hemorrhage, improve the quality of life of patients, and reduce the economic and psychological burden caused by this. It solves the problem of the lack of a method for predicting the risk of constipation in young and middle-aged patients with first-onset cerebral hemorrhage in the prior art.

[0041] 2. Based on the historical clinical data of young and middle-aged patients with first-ever cerebral hemorrhage, the present invention determined the influencing factors of constipation in young and middle-aged patients with first-ever cerebral hemorrhage by comprehensively using univariate analysis, multiple linear analysis and logistic regression analysis methods, and constructed a risk prediction model based on the influencing factors of constipation in young and middle-aged patients with first-ever cerebral hemorrhage, which significantly improved the prediction accuracy of the risk of constipation in young and middle-aged patients with first-ever cerebral hemorrhage.

[0042] 3. After constructing the risk prediction model, the present invention also carried out strict internal validation including Bootstrap autonomous sampling method, area under the receiver operating characteristic curve test, Hosmer-Lemeshow test and calibration chart evaluation and external validation, ensuring the stability and reliability of the risk prediction model, which not only improved the accuracy of the constipation risk prediction results, but also provided a solid scientific basis for the wide application of the risk prediction model in clinical practice. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 The figure shows a flow chart of a method for predicting the risk of constipation in young and middle-aged patients with first-onset cerebral hemorrhage provided by an embodiment of the present invention;

[0044] Figure 2 Shown is a nomogram of constipation risk prediction results provided by an embodiment of the present invention;

[0045] Figure 3 Shown is a schematic diagram of a receiver operating characteristic curve based on internally validated constipation risk prediction results provided by an embodiment of the present invention;

[0046] Figure 4 It is a schematic diagram showing a curve comparison between the internal validation constipation risk prediction result and the actual constipation risk result of the risk prediction model provided by an embodiment of the present invention;

[0047] Figure 5 Shown is a schematic diagram of a receiver operating characteristic curve based on externally validated constipation risk prediction results provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0048] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. The embodiments of the present invention and the technical features in the embodiments may be combined with each other unless there is a conflict.

[0049] The term "and / or" is only a description of the association relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " generally indicates that the related objects are in an "or" relationship.

[0050] Example 1

[0051] like Figure 1 As shown, this embodiment introduces a method for predicting the risk of constipation in young and middle-aged patients with first-onset cerebral hemorrhage, including:

[0052] Step 1: Obtain clinical data of young and middle-aged patients with first-onset cerebral hemorrhage.

[0053] The clinical data of young and middle-aged patients with first-onset cerebral hemorrhage are the basis for building a risk prediction model. The clinical data of young and middle-aged patients with first-onset cerebral hemorrhage contain various information about the patients' physiology, pathology, treatment and rehabilitation process, and can reflect the patients' overall health status and the impact of cerebral hemorrhage.

[0054] The present invention uses the clinical data of young and middle-aged patients with first-onset cerebral hemorrhage to analyze the potential influencing factors of concurrent constipation and provides necessary input for building a risk prediction model, which can ensure that the basic data for building the prediction model is comprehensive and accurate, and help improve the accuracy and reliability of the risk prediction model.

[0055] Step 2: Input the clinical data of the young and middle-aged patients with first-onset cerebral hemorrhage into a pre-trained risk prediction model, and output the constipation risk prediction results.

[0056] The risk prediction model provided by the present invention is obtained based on the historical clinical data and statistical analysis methods of a large number of young and middle-aged patients with first-onset cerebral hemorrhage. The risk prediction model can predict the clinical data of newly input young and middle-aged patients with first-onset cerebral hemorrhage by learning the laws and patterns in the historical data.

[0057] The risk prediction model is used to process the clinical data of new middle-aged and young patients with first-ever cerebral hemorrhage, and the constipation risk prediction results are output. Doctors can take targeted prevention and intervention measures in a timely manner based on the prediction results to reduce the incidence of constipation in patients and improve rehabilitation effects.

[0058] Step 3: Based on the constipation risk prediction results, draw a nomogram of constipation risk prediction results.

[0059] The nomogram is an intuitive graphical representation method that can clearly display the constipation risk prediction results and its relationship with various influencing factors.

[0060] The present invention presents the constipation risk prediction results in a graphical manner by drawing a nomogram, which facilitates doctors to quickly and intuitively understand the patient's constipation risk situation and formulate personalized treatment and rehabilitation plans accordingly. At the same time, it also improves doctors' understanding and application of prediction results, promotes doctor-patient communication, and enhances patients' and their families' trust and cooperation in treatment plans.

[0061] Among them, the clinical data of the middle-aged and young patients with first-onset cerebral hemorrhage were analyzed by univariate analysis, multiple linear analysis and logistic regression analysis, the influencing factors of constipation in middle-aged and young patients with first-onset cerebral hemorrhage were determined, and the risk prediction model was constructed.

[0062] The present invention uses a univariate analysis method to analyze the relationship between the influencing factors and constipation of each young and middle-aged patient with first-onset cerebral hemorrhage, screen out variables with statistical significance, and preliminarily determine the factors affecting constipation.

[0063] The present invention utilizes the multilinear analysis method to analyze the linear relationship between multiple statistically significant variables and detect whether there is a multicollinearity problem, thereby ensuring that the variables in the risk prediction model are independent of each other, avoiding prediction errors caused by collinearity, and improving the stability and accuracy of the risk prediction model.

[0064] The present invention utilizes the Logistic regression analysis method to determine the key factors affecting constipation based on the relationship between the binary dependent variable and the independent variable package, and constructs the final prediction model.

[0065] Example 2

[0066] With the same inventive concept as Example 1, this example introduces a method for predicting the risk of constipation in young and middle-aged patients with a first cerebral hemorrhage. Based on the historical clinical data of young and middle-aged patients with a first cerebral hemorrhage, this example deeply explores the influencing factors of constipation in young and middle-aged patients with a first cerebral hemorrhage, and then constructs a risk prediction model, aiming to provide a simple and easy-to-use prediction tool for clinical practice, helping medical staff to identify high-risk patient groups early and accurately. The present invention not only provides a solid theoretical support for preventing constipation in young and middle-aged patients with a first cerebral hemorrhage, but also promotes the realization of personalized management and prevention strategies for constipation in young and middle-aged patients with a first cerebral hemorrhage.

[0067] Step 1: Obtain clinical data of young and middle-aged patients with first-onset cerebral hemorrhage.

[0068] According to the clinical data questionnaire for young and middle-aged patients with first-onset cerebral hemorrhage complicated by constipation, this embodiment uses the hospital electronic medical record data platform system to efficiently complete the data collection work. To ensure the accuracy and consistency of the study, this embodiment conducted comprehensive training for the research team members, and the training content covered the operation of the electronic medical record database, and the effective collection and entry methods of data. In order to minimize the data deviations or errors that may be introduced by human factors, this embodiment adopts a double-check strategy, strictly examines the completeness and accuracy of the data collected from young and middle-aged patients with cerebral hemorrhage, and implements double-entry on this basis to ensure the authenticity and reliability of the data.

[0069] Step 2: Input the clinical data of the young and middle-aged patients with first-onset cerebral hemorrhage into a pre-trained risk prediction model, and output the constipation risk prediction results.

[0070] The clinical data of the young and middle-aged patients with first-onset cerebral hemorrhage were analyzed by univariate analysis, multiple linear analysis and logistic regression analysis, the influencing factors of constipation in young and middle-aged patients with first-onset cerebral hemorrhage were determined, and the risk prediction model was constructed, including:

[0071] Step 2.1: Use case-control study methods to obtain historical clinical data of young and middle-aged patients with first-onset cerebral hemorrhage.

[0072] Using a case-control study method, we obtained the clinical data of young and middle-aged patients with first-episode cerebral hemorrhage who visited the Department of Neurosurgery of a Class-A tertiary hospital in Suzhou from January 2019 to January 2023. They were divided into a constipation group and a non-constipation group based on whether constipation occurred.

[0073] In this embodiment, the inclusion criteria for the historical clinical data of young and middle-aged patients with first-onset cerebral hemorrhage include:

[0074] Meet the diagnostic criteria of the Chinese Guidelines for the Diagnosis and Treatment of Cerebral Hemorrhage (2019) and be diagnosed with cerebral hemorrhage by head CT and MRI;

[0075] Age range: 18 to 59 years old;

[0076] First onset of disease;

[0077] Length of hospital stay ≥ 3 days.

[0078] In this embodiment, the exclusion criteria for the historical clinical data of young and middle-aged patients with first-episode cerebral hemorrhage are:

[0079] Suffering from severe organic gastrointestinal diseases, tumors, or previous gastrointestinal surgery;

[0080] Combined with cardiopulmonary failure, severe liver and kidney diseases or malignant tumors;

[0081] Incomplete case data;

[0082] Are pregnant or breastfeeding.

[0083] Step 2.2: Analyze the variables in the historical clinical data of the young and middle-aged patients with first-onset cerebral hemorrhage by univariate analysis to obtain statistically significant variables.

[0084] In this embodiment, the probability of random error of the statistically significant variables is less than 5%. The results of analyzing the variables in the historical clinical data of young and middle-aged patients with first-episode cerebral hemorrhage using the univariate analysis method are shown in Table 1, where: Represents the total number of historical clinical data.

[0085] Table 1 Results of variables in the historical clinical data of young and middle-aged patients with first-episode cerebral hemorrhage (n=375) project Non-constipation group (n=123) Constipation group (n=252) Test statistic P-value gender <![CDATA[0.554 2) ]]> 0.457 male 90（73.2%） 175（69.4%） female 33（26.8%） 77（30.6%） Age (years) <![CDATA[-0.070 1) ]]> 0.944 M(P25, P75) 49（40，55） 49（40.25，55） Bleeding site <![CDATA[7.089 2) ]]> 0.008 Non-basal ganglia 78（63.4%） 123（48.8%） Basal ganglia 45（36.6%） 129（51.2%） Whether surgery <![CDATA[11.222 2) ]]> 0.001 no 61（49.6%） 80（31.7%） yes 62（50.4%） 172（68.3%） Whether calcium antagonists are used <![CDATA[20.213 2) ]]> ＜0.001 0=No 61（49.6%） 66（26.2%） 1=Yes 62（50.4%） 186（73.8%） GCS <![CDATA[-2.572 1) ]]> 0.010 M(P25, P75) 14（8，15） 12（7，15） Maximum body temperature (℃) <![CDATA[-2.945 1) ]]> 0.003 0=36.5~38.0 84（68.3%） 127（50.4%） 1=38.1~39.0 25（20.3%） 90（35.7%） ≥39.1 14（11.4%） 35（13.9%） Fever days (days) <![CDATA[-3.175 1) ]]> 0.001 M(P25, P75) 0（0，2） 0.5（0，3） Whether mechanical ventilation <![CDATA[13.040 2) ]]> ＜0.001 0=No 101（82.1%） 161（63.9%） 1=Yes 22（17.9%） 91（36.1%） Is it sedative? <![CDATA[25.523 2) ]]> ＜0.001 0=No 99（80.5%） 135（53.6%） 1=Yes 24（19.5%） 117（46.4%） Eating habits <![CDATA[18.101 2) ]]> ＜0.001 0 = Oral 74（60.2%） 93（36.9%） 1 = Nasogastric tube 49（39.8%） 159（63.1%） Is vasopressin <![CDATA[0.006 2) ]]> 0.939 0=No 114（92.7%） 233（92.5%） 1=Yes 9（7.3%） 19（7.5%） Whether there is hemiplegia <![CDATA[12.566 2) ]]> ＜0.001 0=No 82（66.7%） 119（47.2%） 1=Yes 41（33.3%） 133（52.8%） Whether there is intraventricular hemorrhage <![CDATA[4.523 2) ]]> 0.033 0=No 82（66.7%） 139（55.2%） 1=Yes 41（33.3%） 113（44.8%） diabetes <![CDATA[0.039 2) ]]> 0.844 0=No 111（90.2%） 229(90.9%) 1=Yes 12(9.8%) 23(9.1%) Coronary heart disease <![CDATA[0.418 3) ]]> 0.518 0=No 122(99.2%) 246(97.6%) 1=Yes 1(0.8%) 6(2.4%) hypertension <![CDATA[0.760 2) ]]> 0.383 0=No 63（51.2%） 117（46.4%） 1=Yes 60（48.8%） 135（53.6%） Cerebral infarction <![CDATA[0.167 2) ]]> 0.682 0=No 117（95.1%） 242（96.0%） 1=Yes 6（4.9%） 10（4.0%） Anticoagulants <![CDATA[0.003 3) ]]> 0.957 0=No 119（96.7%） 242（96.0%） 1=Yes 4（3.3%） 10（4.0%） Speech disorders <![CDATA[0.221 2) ]]> 0.638 0=No 109（88.6%） 219（86.9%） 1=Yes 14（11.4%） 33（13.1%） Lung infection <![CDATA[0.873 2) ]]> 0.350 0=No 111（90.2%） 219（86.9%） 1=Yes 12（9.8%） 33(13.1%) Whether to stay in ICU <![CDATA[2.209 2) ]]> 0.137 0=No 89(72.4%) 163(64.7%) 1=Yes 34(27.6%) 89(35.3%) Dehydrating agent <![CDATA[-2.110 1) ]]> 0.035 0=No 12（9.8%） 10（4.0%） 1=1~ 102（82.9%） 215（85.3%） 2=4~ 9（7.3%） 27（10.7%） BI <![CDATA[-2.848 1) ]]> 0.004 M(P25, P75) 20（0，50） 10（0，30）

[0086] Step 2.3: Perform multicollinearity analysis on the statistically significant variables by using the multilinear analysis method to obtain statistically significant variables without multicollinearity between the variables.

[0087] In this embodiment, when the tolerances of the statistically significant variables are all greater than 0.2 and the variance inflation factors are all less than 5.0, there is no multicollinearity between the statistically significant variables.

[0088] Step 2.4: Assign values ​​to the statistically significant variables that do not have multicollinearity among the variables, as shown in Table 2. Obtain the assigned variables.

[0089] Table 2. The assignment method of statistically significant variables without multicollinearity between variables project Assignment method Bleeding site Non-basal ganglia = 0, basal ganglia = 1 Surgery No=0, Yes=1 Use of calcium antagonists No=0, Yes=1 GCS score (points) Original value input Highest body temperature during hospitalization (℃) 36.5~38.0=0 38.1~39.0=1 ≥39.1=2 Fever days (days) Original value input Mechanical ventilation No=0, Yes=1 Use of sedatives No=0, Yes=1 Eating habits Oral = 0, nasogastric = 1 Hemiplegia No=0, Yes=1 Intraventricular hemorrhage No=0, Yes=1 Number of dehydrating drugs used (types) No=0, 1~3=1, 4~=2 BI Original value input

[0090] Step 2.5: The assigned variables were used as independent variables, and whether constipation occurred was used as the dependent variable. Logistic regression analysis was performed using the backward likelihood method to obtain the influencing factors of constipation in young and middle-aged patients with first-episode cerebral hemorrhage.

[0091] In this embodiment, the influencing factors for constipation in young and middle-aged patients with first-episode cerebral hemorrhage include Glasgow Coma Scale (GCS), whether they have undergone surgery, whether they have hemiplegia, whether they have used mechanical ventilation, whether they have used sedatives, and whether they have used calcium ion antagonists.

[0092] Among them, the results of logistic regression analysis of constipation in young and middle-aged patients with first-onset cerebral hemorrhage are shown in Table 3.

[0093] Table 3 Logistic regression analysis results of constipation in young and middle-aged patients with first episode of cerebral hemorrhage (n=375) project Beta value Standard error Wald χ2 value P-value OR Value 95% CI constant -1.850 0.633 8.539 0.003 0.157 Surgery 0.629 0.259 5.883 0.015 1.875 1.129～3.117 Use of calcium antagonists 0.758 0.248 9.344 0.002 2.134 1.312～3.469 GCS score 0.084 0.040 4.426 0.035 1.088 1.006～1.176 Mechanical ventilation 0.818 0.362 5.120 0.024 2.267 1.116～4.606 Use of sedatives 0.894 0.277 10.397 0.001 2.445 1.420～4.209 Hemiplegia 0.650 0.249 6.798 0.009 1.915 1.175～3.120

[0094] Step 2.6: Construct a risk prediction model based on the influencing factors of constipation in young and middle-aged patients with first-onset cerebral hemorrhage.

[0095] Step 3: Internal validation of the risk prediction model.

[0096] After the risk prediction model is constructed, internal validation of the risk prediction model is also performed, including:

[0097] Step 3.1: Use the Bootstrap autonomous sampling method to sample the historical clinical data of the young and middle-aged patients with first-episode cerebral hemorrhage to obtain an internal validation data set;

[0098] Step 3.2: inputting the internal validation data set into the risk prediction model, and outputting the internal validation constipation risk prediction result;

[0099] Step 3.3: Based on the internal validation constipation risk prediction results, the discrimination of the risk prediction model is detected using the area under the receiver operating characteristic curve;

[0100] In this embodiment, the closer the area under the curve value is to 1, the better the discrimination of the risk prediction model is. The schematic diagram of the receiver operating characteristic curve based on the internal validation constipation risk prediction result is as follows: Figure 3 shown.

[0101] Based on the internal validation risk prediction results, the area under the receiver operating characteristic curve was 0.733, the sensitivity was 51.2%, the specificity was 87.0%, and the maximum Youden index was 1.382, indicating that the risk prediction model had a good fitting effect and a high predictive value.

[0102] Step 3.4: Based on the internal validation constipation risk prediction results, the prediction performance of the risk prediction model was evaluated using the Hosmer-Lemeshow test and calibration plot.

[0103] In this embodiment, based on the internally validated constipation risk prediction results, the predictive performance of the risk prediction model was evaluated using the Hosmer-Lemeshow test and calibration graph, wherein when the predictive performance of the risk prediction model was tested using the Hosmer-Lemeshow test, the significance level was equal to 0.05, and the goodness of fit of the risk prediction model was evaluated.

[0104] In this embodiment, a schematic diagram of a curve comparison between the internal validation constipation risk prediction result and the actual constipation risk result is shown in FIG. Figure 4 As shown, based on the internal validation risk prediction results, the Hosmer-Lemeshow test showed that the continuous correction test results =8.956, the probability of random error =0.346, indicating that the predicted curve of the risk prediction model in the calibration diagram has a high degree of overlap with the ideal curve. In addition, the area under the receiver operating characteristic curve of the risk prediction model is 0.716, indicating that the prediction model has good discrimination ability.

[0105] Step 4: Externally validate the risk prediction model.

[0106] Step 4.1: Use the clinical data of young and middle-aged patients with first-episode cerebral hemorrhage in the later stage of the internal validation dataset as the external validation dataset.

[0107] In this embodiment, a case-control study method was used to obtain data from 146 young and middle-aged patients with first-onset cerebral hemorrhage admitted to the Department of Neurosurgery of the same hospital from February 2023 to September 2024 to externally validate the risk prediction model, and the patients were divided into a constipation group and a non-constipation group based on whether constipation occurred.

[0108] In this embodiment, the inclusion criteria of the external validation dataset are:

[0109] Meet the diagnostic criteria of the Chinese Guidelines for the Diagnosis and Treatment of Cerebral Hemorrhage (2019) and be diagnosed with cerebral hemorrhage by head CT and MRI;

[0110] Age range: 18 to 59 years old;

[0111] First onset of disease;

[0112] Length of hospital stay ≥ 3 days.

[0113] In this example, the exclusion criteria for the external validation dataset are:

[0114] Suffering from severe organic gastrointestinal diseases, tumors, or previous gastrointestinal surgery;

[0115] Combined with cardiopulmonary failure, severe liver and kidney diseases or malignant tumors;

[0116] Incomplete case data;

[0117] Are pregnant or breastfeeding.

[0118] Step 4.2: Input the external validation data set into the risk prediction model, and output an external validation constipation risk prediction result.

[0119] In this embodiment, among the 146 young and middle-aged patients with first-onset cerebral hemorrhage, there were 101 males and 45 females. The risk prediction model predicted that 123 patients would develop constipation, while 94 patients actually developed constipation.

[0120] Step 4.3: Based on the externally validated constipation risk prediction results, the discrimination of the risk prediction model is detected using the area under the receiver operating characteristic curve.

[0121] In this embodiment, a schematic diagram of a receiver operating characteristic curve based on external validation of constipation risk prediction results is shown in FIG. Figure 5 As shown in the figure. Based on the externally validated constipation risk prediction results, the area under the receiver operating characteristic curve was 0.751, indicating that the risk prediction model had good discrimination.

[0122] Step 4.4: Compare the externally validated constipation risk prediction results with the actual constipation risk results to evaluate the fit of the risk prediction model.

[0123] Hosmer-Lemeshow test shows: Continuous correction test results =7.536, the probability of random error =0.480, indicating that the risk prediction model has good fitting goodness of fit.

[0124] When the optimal critical value of the risk prediction model provided in this embodiment is calculated based on the Youden index to be 0.643, the sensitivity is 73.4% and the specificity is 69.2%, indicating that the risk prediction model has a good fitting effect and a high predictive value.

[0125] Step 5: Based on the constipation risk prediction results, draw a constipation risk prediction nomogram.

[0126] The nomogram of the secretion risk prediction results obtained in this embodiment is as follows: Figure 2 shown.

[0127] In summary, the incidence of constipation in young and middle-aged patients with first-onset cerebral hemorrhage is high, and patients who undergo surgery, mechanical ventilation, hemiplegia, sedatives, and calcium ion antagonists are more likely to suffer from constipation. The risk prediction model constructed in this embodiment has a strong predictive ability and can be used as a screening tool for patients with high risk of constipation, providing a reference for timely formulation of personalized prevention strategies.

[0128] Example 3

[0129] Based on the same inventive concept as Example 1, this example introduces a system for predicting the risk of constipation in young and middle-aged patients with first-onset cerebral hemorrhage, including:

[0130] Data acquisition module, used to obtain clinical data of young and middle-aged patients with first-episode cerebral hemorrhage;

[0131] A risk prediction module, used to input the clinical data of the young and middle-aged patients with first-episode cerebral hemorrhage into a pre-trained risk prediction model;

[0132] A prediction result output module, used to output the constipation risk prediction result;

[0133] A nomogram generation module is used to draw a nomogram of constipation risk prediction results according to the constipation risk prediction results;

[0134] Among them, the influencing factors of constipation in young and middle-aged patients with first-episode cerebral hemorrhage were determined by univariate analysis, multiple linear analysis and logistic regression analysis, and the risk prediction model was constructed.

[0135] The specific functions of the above modules are implemented with reference to the relevant contents of the method in Example 1 or 2, which will not be repeated here.

[0136] Example 4

[0137] With the same inventive concept as other embodiments, the present invention provides a computer-readable storage medium having computer instructions stored thereon, wherein the computer instructions, when executed by a processor, implement the steps of the method of the above-mentioned embodiment 1 or 2.

[0138] Example 5

[0139] With the same inventive concept as other embodiments, the present invention further provides a computer program product, including computer instructions, characterized in that when the computer instructions are executed by a processor, the steps of the method in the above-mentioned embodiment 1 or 2 are implemented.

[0140] In summary, the present invention can accurately output the patient's constipation risk prediction results based on the clinical data of young and middle-aged patients with their first cerebral hemorrhage by using a pre-trained risk prediction model, and present the constipation risk prediction results in the form of an intuitive nomogram, so that both doctors and patients can understand the constipation risk more clearly. It not only improves the accuracy of doctors' assessment of patients' constipation risks, but also provides a scientific basis for formulating personalized prevention and intervention measures, thereby helping to reduce the risk of constipation in young and middle-aged patients with their first cerebral hemorrhage, improve the quality of life of patients, and reduce the resulting economic and psychological burden. It solves the problem of the lack of a method for predicting the risk of constipation in young and middle-aged patients with their first cerebral hemorrhage in the prior art.

[0141] Based on the historical clinical data of young and middle-aged patients with first-ever cerebral hemorrhage, the present invention determined the influencing factors of constipation in young and middle-aged patients with first-ever cerebral hemorrhage by comprehensively using univariate analysis method, multiple linear analysis method and logistic regression analysis method, and constructed a risk prediction model according to the influencing factors of constipation in young and middle-aged patients with first-ever cerebral hemorrhage, which significantly improved the prediction accuracy of constipation risk in young and middle-aged patients with first-ever cerebral hemorrhage.

[0142] After constructing the risk prediction model, the present invention also carried out strict internal validation including Bootstrap autonomous sampling method, area under the receiver operating characteristic curve detection, Hosmer-Lemeshow test and calibration chart evaluation and external validation, ensuring the stability and reliability of the risk prediction model, not only improving the accuracy of the constipation risk prediction results, but also providing a solid scientific basis for the wide application of the risk prediction model in clinical practice. It should be understood by those skilled in the art that the embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0143] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0144] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0145] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0146] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

Claims

1. A method for predicting the risk of constipation in young and middle-aged patients with first-onset cerebral hemorrhage, characterized in that: include: Obtain clinical data of young and middle-aged patients with first-onset cerebral hemorrhage; Inputting the clinical data of the young and middle-aged patients with first-onset cerebral hemorrhage into a pre-trained risk prediction model, and outputting a constipation risk prediction result; According to the constipation risk prediction results, a nomogram of constipation risk prediction results was drawn; Among them, the clinical data of the middle-aged and young patients with first-onset cerebral hemorrhage were analyzed by univariate analysis, multiple linear analysis and logistic regression analysis, the influencing factors of constipation in middle-aged and young patients with first-onset cerebral hemorrhage were determined, and the risk prediction model was constructed.

2. The method for constructing a risk prediction model for constipation in young and middle-aged patients with first-onset cerebral hemorrhage according to claim 1, characterized in that: The method of analyzing the clinical data of the young and middle-aged patients with first-episode cerebral hemorrhage by univariate analysis, multiple linear analysis and logistic regression analysis to determine the influencing factors of constipation in the young and middle-aged patients with first-episode cerebral hemorrhage and constructing the risk prediction model includes: A case-control study method was used to obtain historical clinical data of young and middle-aged patients with first-episode cerebral hemorrhage; By using the univariate analysis method, the variables in the historical clinical data of the young and middle-aged patients with first-onset cerebral hemorrhage were analyzed to obtain statistically significant variables; By using a multi-linear analysis method, a multi-collinearity analysis is performed on the statistically significant variables to obtain statistically significant variables without multi-collinearity between the variables; Assigning values ​​to statistically significant variables without multicollinearity between the variables to obtain assigned variables; The variables after the assignment were used as independent variables, and whether constipation occurred was used as the dependent variable. Logistic regression analysis was performed using the backward likelihood method to obtain the influencing factors of constipation in young and middle-aged patients with first-episode cerebral hemorrhage. A risk prediction model was constructed based on the influencing factors of constipation in young and middle-aged patients with first-episode cerebral hemorrhage.

3. The method for constructing a risk prediction model for constipation in young and middle-aged patients with first episode of cerebral hemorrhage according to claim 2, characterized in that: The variables in the historical clinical data of the young and middle-aged patients with first-onset cerebral hemorrhage were analyzed by univariate analysis to obtain statistically significant variables, wherein the probability of random errors in the statistically significant variables was less than 5%.

4. The method for constructing a risk prediction model for constipation in young and middle-aged patients with first-onset cerebral hemorrhage according to claim 2, characterized in that: By using the multilinear analysis method, multicollinearity analysis was performed on the statistically significant variables to obtain statistically significant variables without multicollinearity between the variables. When the tolerance of the statistically significant variables was greater than 0.2 and the variance inflation factor was less than 5.0, there was no multicollinearity between the statistically significant variables.

5. The method for constructing a risk prediction model for constipation in young and middle-aged patients with first-onset cerebral hemorrhage according to claim 1, characterized in that: The influencing factors of constipation in young and middle-aged patients with first-episode cerebral hemorrhage include Glasgow Coma Scale (GCS), whether they have undergone surgery, whether they have hemiplegia, whether they have used mechanical ventilation, whether they have used sedatives, and whether they have used calcium ion antagonists.

6. The method for constructing a risk prediction model for constipation in young and middle-aged patients with first-onset cerebral hemorrhage according to claim 2, characterized in that: After the risk prediction model is constructed, internal validation of the risk prediction model is also performed, including: The Bootstrap autonomous sampling method was used to sample the historical clinical data of the young and middle-aged patients with first-episode cerebral hemorrhage to obtain an internal validation data set; Inputting the internal validation data set into the risk prediction model, and outputting an internal validation constipation risk prediction result; Based on the internal validation constipation risk prediction results, the discrimination of the risk prediction model is detected using the area under the receiver operating characteristic curve; Based on the internal validation constipation risk prediction results, the predictive performance of the risk prediction model was evaluated using the Hosmer-Lemeshow test and calibration plot.

7. The method for constructing a risk prediction model for constipation in young and middle-aged patients with first-onset cerebral hemorrhage according to claim 6, characterized in that: Based on the internal validation constipation risk prediction results, the area under the receiver operating characteristic curve is used to detect the discrimination of the risk prediction model, wherein the closer the area under the curve value is to 1, the better the discrimination of the risk prediction model.

8. The method for constructing a risk prediction model for constipation in young and middle-aged patients with first episode of cerebral hemorrhage according to claim 6, characterized in that: Based on the internal validation constipation risk prediction results, the predictive performance of the risk prediction model was evaluated using the Hosmer-Lemeshow test and calibration graph, wherein when the predictive performance of the risk prediction model was tested using the Hosmer-Lemeshow test, the significance level was equal to 0.05, and the goodness of fit of the risk prediction model was evaluated.

9. The method for constructing a risk prediction model for constipation in young and middle-aged patients with first-onset cerebral hemorrhage according to claim 6, characterized in that: After the internal validation of the risk prediction model, the external validation of the risk prediction model is also included: The clinical data of young and middle-aged patients with first-episode cerebral hemorrhage in the later period of the internal validation dataset were used as the external validation dataset; Inputting the external validation data set into the risk prediction model, and outputting an external validation constipation risk prediction result; Based on the externally validated constipation risk prediction results, the discrimination of the risk prediction model is detected using the area under the receiver operating characteristic curve; The externally validated constipation risk prediction results are compared with the actual constipation risk results to evaluate the fit of the risk prediction model.

10. A system for predicting the risk of constipation in young and middle-aged patients with first-onset cerebral hemorrhage, characterized in that: include: Data acquisition module, used to obtain clinical data of young and middle-aged patients with first-episode cerebral hemorrhage; A risk prediction module, used to input the clinical data of the young and middle-aged patients with first-episode cerebral hemorrhage into a pre-trained risk prediction model; A prediction result output module, used to output the constipation risk prediction result; A nomogram generation module is used to draw a nomogram of constipation risk prediction results according to the constipation risk prediction results; Among them, the influencing factors of constipation in young and middle-aged patients with first-episode cerebral hemorrhage were determined by univariate analysis, multiple linear analysis and logistic regression analysis, and the risk prediction model was constructed.

Images