Prediction method and device for recurrence risk after treatment of small and medium-sized unbroken intracranial aneurysm stent spring ring, electronic equipment and storage medium

Abstract

The invention discloses a method and device for predicting the recurrence risk after treatment of a small and medium-sized unruptured intracranial aneurysm support spring ring, electronic equipment and a storage medium, and belongs to the field of clinical medicines.The method comprises the steps that parameter values of all prediction parameters of a patient to be evaluated are obtained; wherein the prediction parameters comprise the aneurysm size, the treatment mode, the postoperative immediate occlusion degree and the preoperative high-resolution magnetic resonance blood vessel wall imaging tumor wall strengthening level; determining a target category corresponding to each prediction parameter according to the parameter value of each prediction parameter; according to the target category, based on a preset scoring rule, determining a recurrence risk score of each prediction parameter; according to the recurrence risk score of each prediction parameter, determining the recurrence risk probability of the to-be-evaluated patient after the treatment of the small and medium-sized unbroken intracranial aneurysm stent spring ring; by implementing the method, the problem that the recurrence risk of small and medium-sized unbroken intracranial aneurysm after interventional operation is not accurately predicted in the prior art can be solved.

Description

Technical Field

[0001] The present invention relates to the field of clinical medical technology, and in particular to a method, device, electronic device and storage medium for predicting the recurrence risk of small and medium-sized unruptured intracranial aneurysms after treatment with stent coils. Background Art

[0002] Intracranial aneurysms (IAs) are cerebrovascular diseases caused by pathological dilation of the cerebral artery wall, with a prevalence of approximately 3%-5% in the adult population. Rupture of an unruptured intracranial aneurysm (UIA) can trigger catastrophic spontaneous subarachnoid hemorrhage, resulting in extremely high morbidity and mortality. Currently, endovascular interventional therapies (such as simple microcoil embolization and stent-assisted microcoil embolization) and craniotomy and microsurgical clipping are the mainstays of treatment for UIAs and reduce the risk of rupture. However, even after mainstream endovascular interventional therapy, particularly for small and medium-sized UIAs, patients still face a high recurrence rate of 13%-26% during follow-up. Aneurysm recurrence not only increases the risk of rebleeding but may also require secondary or even multiple interventions, placing additional physical, psychological, and financial burdens on patients and potentially leading to adverse clinical outcomes. Therefore, accurately assessing and predicting the recurrence risk of patients with small and medium-sized UIA after endovascular interventional treatment is of vital clinical significance for guiding the clinic to formulate individualized postoperative follow-up strategies, optimize the allocation of medical resources, promptly detect and treat high-risk recurrent lesions, and ultimately improve patient prognosis.

[0003] To effectively assess the risk of aneurysm recurrence after surgery, researchers at home and abroad have attempted to construct various prediction models. For example, the ARSS (Aneurysm Recanalization Stratification Scale) model incorporates factors such as aneurysm size, rupture status, intratumoral thrombosis, treatment method, and degree of immediate postoperative occlusion to predict the probability of postoperative aneurysm retreatment. Other studies have constructed scoring systems based on variables such as aneurysm size, degree of immediate postoperative embolization, and whether stent-assisted embolization was used. Most of these existing models focus on factors such as aneurysm morphological characteristics, patient basic conditions, and immediate treatment efficacy. However, these models generally fail to incorporate key indicators that can directly reflect the degree of local inflammation in the aneurysm wall, particularly the important factor of the aneurysm wall enhancement grade revealed by preoperative high-resolution magnetic resonance vascular wall imaging (HR-VWI). This results in inaccurate prediction of the recurrence risk of small and medium-sized unruptured intracranial aneurysms after stent-coil treatment. Summary of the Invention

[0004] The embodiments of the present invention provide a method, device, electronic device and storage medium for predicting the recurrence risk of small and medium-sized unruptured intracranial aneurysms after stent coil treatment, which can solve the problem of inaccurate prediction of the recurrence risk of small and medium-sized unruptured intracranial aneurysms after interventional treatment in the prior art.

[0005] An embodiment of the present invention provides a method for predicting the risk of recurrence of small to medium-sized unruptured intracranial aneurysms after stent-coil treatment, comprising: obtaining parameter values ​​of various prediction parameters of a patient to be evaluated; wherein the prediction parameters include aneurysm size, treatment method, degree of immediate postoperative occlusion, and preoperative high-resolution magnetic resonance imaging (MRI) of the aneurysm wall enhancement grade;

[0006] Determining the target category corresponding to each prediction parameter according to the parameter value of each prediction parameter;

[0007] Determining a recurrence risk score for each prediction parameter based on the target category and a preset scoring rule;

[0008] Based on the recurrence risk scores of each prediction parameter, the recurrence risk probability of the patients undergoing stent-coil treatment for small and medium-sized unruptured intracranial aneurysms was determined;

[0009] The preset scoring rules are determined in the following ways:

[0010] Acquiring a plurality of clinical data samples, wherein a clinical data sample includes observation values ​​of a plurality of prediction parameters and corresponding recurrence of unruptured intracranial aneurysms;

[0011] Based on the clinical data samples, a first multivariate regression model is constructed with the recurrence of unruptured intracranial aneurysms corresponding to the observed values ​​as the outcome variable, and a regression coefficient of each prediction parameter is determined according to the first multivariate regression model;

[0012] According to the regression coefficient corresponding to each prediction parameter, based on the preset reference value of each category of each prediction parameter, a preset scoring rule is determined.

[0013] Furthermore, the preset reference value of each category of a prediction parameter is determined by the following method:

[0014] If the current prediction parameter is a continuous variable, the value range of the current prediction parameter is divided into at least two category intervals, and the midpoint value of each category interval is used as the preset reference value of the corresponding category interval;

[0015] If the current prediction parameter is a categorical variable, each category of the current prediction parameter is numerically encoded, and the encoding result is used as the preset reference value of the corresponding category.

[0016] Furthermore, the predetermined scoring rules are determined based on the regression coefficient corresponding to each prediction parameter and the predetermined reference value of each category of each prediction parameter, including:

[0017] Determine the offset corresponding to each category of each prediction parameter based on the regression coefficient corresponding to each prediction parameter, the preset reference value of each category of each prediction parameter, and the preset reference value of the benchmark category;

[0018] Based on the preset conversion unit, the offset corresponding to each category of each prediction parameter is converted to generate the recurrence risk score corresponding to each category of each prediction parameter;

[0019] The preset scoring rules are determined based on the recurrence risk scores corresponding to each category of each prediction parameter.

[0020] Furthermore, the recurrence risk probability of the patient undergoing stent-coil treatment for small and medium-sized unruptured intracranial aneurysms is determined based on the recurrence risk score of each prediction parameter, including:

[0021] According to the recurrence risk score of each prediction parameter, the total recurrence risk score of the patient to be evaluated is calculated;

[0022] Based on the preset logistic regression formula, the total recurrence risk score of the evaluated patients was converted into the recurrence risk probability of the evaluated patients after stent-coil treatment of small and medium-sized unruptured intracranial aneurysms.

[0023] Furthermore, the preset logistic regression formula is generated in the following way:

[0024] For each clinical data sample, determine the recurrence risk score corresponding to the observed value of each prediction parameter in the current clinical data sample; calculate and generate the total recurrence risk score corresponding to the current clinical data sample based on the recurrence risk score corresponding to the observed value of each prediction parameter;

[0025] The recurrence risk total score was used as the independent variable, and the recurrence of unruptured intracranial aneurysms corresponding to the clinical data sample of the recurrence risk total score was used as the outcome variable to construct a second multivariate regression model, and the intercept of the second multivariate regression model was determined.

[0026] According to the intercept of the second multivariate regression model and the preset conversion unit, a logistic regression formula for converting the total recurrence risk score into the recurrence risk probability was constructed.

[0027] Based on the above method embodiments, the present invention provides corresponding device embodiments.

[0028] An embodiment of the present invention provides a device for predicting the recurrence risk of small to medium-sized unruptured intracranial aneurysms after stent-coil treatment, comprising: a module for acquiring parameters to be evaluated, a module for determining target categories, and a module for outputting the probability of recurrence risk;

[0029] The module for obtaining parameters to be evaluated is used to obtain parameter values ​​of various prediction parameters of the patient to be evaluated; wherein the prediction parameters include aneurysm size, treatment method, degree of occlusion immediately after surgery, and aneurysm wall enhancement grade based on preoperative high-resolution magnetic resonance imaging of the vascular wall;

[0030] The target category determination module is used to determine the target category corresponding to each prediction parameter according to the parameter value of each prediction parameter;

[0031] The recurrence risk probability output module is configured to determine a recurrence risk score for each prediction parameter based on the target category and a preset scoring rule; and to determine a recurrence risk probability for a patient undergoing stent-coil treatment for a small or medium-sized unruptured intracranial aneurysm based on the recurrence risk score for each prediction parameter.

[0032] The preset scoring rules are determined in the following ways:

[0033] Acquiring a plurality of clinical data samples, wherein a clinical data sample includes observation values ​​of a plurality of prediction parameters and corresponding recurrence of unruptured intracranial aneurysms;

[0034] Based on the clinical data samples, a first multivariate regression model is constructed with the recurrence of unruptured intracranial aneurysms corresponding to the observed values ​​as the outcome variable, and a regression coefficient of each prediction parameter is determined according to the first multivariate regression model;

[0035] According to the regression coefficient corresponding to each prediction parameter, based on the preset reference value of each category of each prediction parameter, a preset scoring rule is determined.

[0036] Furthermore, the recurrence risk probability output module determines a preset scoring rule based on the regression coefficient corresponding to each prediction parameter and the preset reference value of each category of each prediction parameter, including:

[0037] Determine the offset corresponding to each category of each prediction parameter based on the regression coefficient corresponding to each prediction parameter, the preset reference value of each category of each prediction parameter, and the preset reference value of the benchmark category;

[0038] Based on the preset conversion unit, the offset corresponding to each category of each prediction parameter is converted to generate the recurrence risk score corresponding to each category of each prediction parameter;

[0039] The preset scoring rules are determined based on the recurrence risk scores corresponding to each category of each prediction parameter.

[0040] Based on the above method embodiment, the present invention provides a corresponding electronic device embodiment.

[0041] An embodiment of the present invention provides an electronic device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the method for predicting the risk of recurrence of small and medium-sized unruptured intracranial aneurysms after stent coil treatment as described in any one of the above-mentioned method embodiments is implemented.

[0042] Based on the above method embodiment, the present invention provides a corresponding storage medium embodiment.

[0043] An embodiment of the present invention provides a storage medium having a computer program stored thereon, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the method for predicting the risk of recurrence after treatment with stent coils for small and medium-sized unruptured intracranial aneurysms as described in any one of the above-mentioned method embodiments.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] An embodiment of the present invention provides a method, device, electronic device and storage medium for predicting the risk of recurrence after treatment with stent coils for small and medium-sized unruptured intracranial aneurysms. The method obtains parameter values ​​of various prediction parameters of the patient to be evaluated, including aneurysm size, treatment method, degree of occlusion immediately after surgery, and key preoperative high-resolution magnetic resonance vascular wall imaging aneurysm wall enhancement level. Based on these parameter values, the target category corresponding to each parameter is determined, and a recurrence risk score is determined for the target category of each prediction parameter based on a preset scoring rule, and finally the overall recurrence risk probability of the patient is determined by combining the risk scores of each prediction parameter. Among them, the preset scoring rule is formulated by analyzing a clinical data sample set, constructing a first multivariate regression model to obtain the regression coefficients of each prediction parameter, and combining these regression coefficients with the preset reference values ​​of each category of each prediction parameter.

[0046] This invention significantly improves the accuracy of predicting the risk of recurrence after stent-coil treatment of small and medium-sized unruptured intracranial aneurysms by incorporating the important factor of the aneurysm wall enhancement grade revealed by preoperative high-resolution magnetic resonance vascular wall imaging (HR-VWI) into the prediction considerations, providing strong technical support for early clinical identification of high-risk individuals and optimized follow-up management. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 The present invention provides a flowchart of a method for predicting the risk of recurrence of small and medium-sized unruptured intracranial aneurysms after stent coil treatment, according to an embodiment of the present invention.

[0048] Figure 2 4 is a ROC curve diagram provided by an embodiment of the present invention.

[0049] Figure 3 3 is a comparison diagram of the ROC curve of the method provided by one embodiment of the present invention and the ROC curve of the prior art.

[0050] Figure 4 This is a statistical diagram of recurrence of the method provided in one embodiment of the present invention.

[0051] Figure 5 This is a statistical diagram of recurrence in the prior art provided by an embodiment of the present invention.

[0052] Figure 6 This is a comparison chart of recurrence situations between the risk stratification of the method provided in one embodiment of the present invention and that of the prior art.

[0053] Figure 7 This is a schematic structural diagram of a device for predicting the risk of recurrence of small and medium-sized unruptured intracranial aneurysms after stent coil treatment, provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0054] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0055] like Figure 1 As shown, one embodiment of the present invention provides a method for predicting the risk of recurrence of small to medium-sized unruptured intracranial aneurysms after stent coil treatment, comprising at least the following steps:

[0056] Step S1, obtaining parameter values ​​of various prediction parameters of the patient to be evaluated; wherein the prediction parameters include aneurysm size, treatment method, degree of occlusion immediately after surgery, and aneurysm wall enhancement grade of preoperative high-resolution magnetic resonance vascular wall imaging.

[0057] It should be noted that the treatment methods include simple microcoil embolization (Simple Coiling) and stent-assisted microcoil embolization (Stent-Assisted Coiling, SAC).

[0058] It should be noted that the degree of immediate postoperative occlusion refers to the immediate state and level of occlusion observed by digital subtraction angiography (DSA) immediately after the completion of endovascular interventional treatment for an unruptured intracranial aneurysm (UIA), indicating that the aneurysm is effectively filled with embolic material and blood flow no longer enters the aneurysm cavity. The modified Raymond-Roy classification (MRRC) is used to assess the degree of immediate postoperative occlusion, which is divided into MRRC grades I, II, and III (IIIa, IIIb). The MRRC classification specifically divides the occlusion state into: MRRC grade I, which refers to complete occlusion of the aneurysm with no blood flow entering; MRRC grade II, which refers to the majority of the aneurysm being occluded, but with a small amount of residual blood flow at the neck; and MRRC grade III, which refers to incomplete occlusion with significant blood flow still filling the aneurysm cavity (this grade can be further subdivided into subtypes IIIa and IIIb to more accurately describe the extent and nature of residual blood flow).

[0059] The criteria for judging aneurysm recurrence are based on long-term imaging follow-up after surgery. Patients usually undergo DSA reexamination at key time points such as the 3rd, 6th, and 12th months after embolization. In these follow-up examinations, MRRC grading is also used to dynamically evaluate the evolution of the aneurysm's occlusion status. If, during a follow-up review, the DSA image shows that the MRRC grading of the aneurysm has increased in grade compared to the MRRC grade immediately after surgery (for example, from MRRC grade I to grade II or grade III, or from grade II to grade III), it is clearly determined that the aneurysm has recurred after interventional treatment.

[0060] It should be noted that the preoperative high-resolution magnetic resonance vascular wall imaging (HR-VWI) aneurysm wall enhancement grade refers to an indicator used by specific magnetic resonance imaging technology to evaluate whether and to what extent there is an inflammatory response or pathological change in the aneurysm wall before treatment of an unruptured intracranial aneurysm (UIA).

[0061] This assessment is primarily carried out through the following methods:

[0062] Before treatment of a UIA, a Philips 3.0T MRI device is typically used for HR-VWI scanning. High-resolution T1-weighted (T1-weighted) images before and after contrast agent injection are compared to assess the enhancement of the aneurysm wall.

[0063] Secondly, based on the enhancement characteristics, the degree of tumor wall enhancement is qualitatively graded, usually from 0 to 3:

[0064] Grade 0, no enhancement of the tumor wall or enhancement of the suspicious part;

[0065] Grade 1: The tumor wall showed localized and thick wall enhancement, with the measured enhancement thickness >1 mm;

[0066] Grade 2: The tumor wall is annular and thinly enhanced, with the measured enhancement thickness ≤ 1 mm;

[0067] Grade 3: The tumor wall shows an entire ring-shaped enhancement, and the enhancement thickness is locally or throughout the tumor wall >1 mm.

[0068] In addition, quantitative analysis can be performed on aneurysm images obtained through three-dimensional HR-VWI scanning to further objectively assess their degree of enhancement. This method measures the maximum signal intensity value of the aneurysm wall (SIwall) and the maximum signal intensity value of the pituitary stalk (SIstalk) as a reference on high-resolution T1-weighted enhanced sequence images. By calculating the ratio of these two values, namely the contrast ratio of the enhanced signals of the aneurysm to the pituitary stalk (Contrast Ratio, CRstalk = SIwall / SIstalk), an objective and quantitative index of the degree of aneurysm wall enhancement can be obtained.

[0069] In a preferred embodiment, the aneurysm wall enhancement grade of preoperative high-resolution magnetic resonance vessel wall imaging (HR-VWI) is determined as a predictive parameter for predicting the recurrence risk of unruptured intracranial aneurysm (UIA) after treatment by the following method:

[0070] (1) Research subjects and data collection: This example uses 69 cystic UIAs from 65 patients as the research data source. The patients' demographic information (such as age and gender), aneurysm characteristics (such as size, location, neck width, and top-neck ratio), existing risk assessment scores (such as PHASES score and ELAPSS score), HR-VWI imaging characteristics, specific endovascular treatment methods (simple microcoil embolization or stent-assisted coil embolization, SAC), and the degree of aneurysm occlusion immediately after surgery (modified Raymond-Roy classification, MRRC) were recorded in detail. All UIAs were followed up with continuous imaging (average 12.6±4.7 months) to determine whether there was aneurysm recurrence. The recurrence rate in this study was 20.3% (14 / 69 UIAs).

[0071] (2) Assessment and classification of HR-VWI aneurysm wall enhancement grade: HR-VWI examination was completed for each UIA before treatment. Based on the enhancement of the aneurysm wall in HR-VWI images, it was divided into grades 0 to 3 (grade 0: no enhancement or suspicious partial enhancement; grade 1: local thick wall enhancement, thickness >1mm; grade 2: annular thin wall enhancement, thickness ≤1mm; grade 3: annular enhancement, local or overall thickness >1mm). Among the 69 UIAs in this study, 47 were grade 0, 11 were grade 1, 4 were grade 2, and 7 were grade 3; among them, 11 had annular aneurysm wall enhancement (grades 2 and 3). At the same time, the contrast ratio of aneurysm to pituitary stalk enhancement (CRstalk value) was calculated and recorded. There were 10 UIAs with CRstalk ≥ 0.5.

[0072] (3) Statistical analysis to determine the independent predictive value of predictive parameters: Statistical methods were used to analyze the relationship between various factors and recurrence after UIA endovascular intervention. First, a univariate analysis was performed to preliminarily screen potential predictive parameters related to UIA recurrence. The results showed that patient gender (male), smoking history, aneurysm size, aneurysm location (internal carotid artery), PHASES score, ELAPSS score, aneurysm top-neck ratio, HR-VWI aneurysm wall enhancement (such as annular aneurysm wall enhancement grade 2 or 3, annular aneurysm wall enhancement grade 3, CRstalk ≥ 0.5), treatment method (simple coil embolization), and MRRC grade of immediate postoperative occlusion degree and coil compression during follow-up were all significantly associated with UIA recurrence.

[0073] Subsequently, to control for confounding factors and identify independent risk factors, factors that were significant or clinically considered important in univariate analysis (including smoking history, aneurysm size, aneurysm location, immediate postoperative occlusion, treatment modality, and different definitions of HR-VWI aneurysm wall enhancement) were incorporated into a multivariate logistic regression model. Specifically, the following model was constructed:

[0074] Model 1: Including annular aneurysm wall enhancement grade 2 or 3 as HR-VWI indicators, the analysis results showed that aneurysm size and annular aneurysm wall enhancement grade 2 or 3 were independent risk factors for recurrence after UIA interventional treatment.

[0075] Model 2: Including annular aneurysm wall enhancement grade 3 as HR-VWI indicator, the analysis results showed that smoking, aneurysm size and annular aneurysm wall enhancement grade 3 were independent risk factors for recurrence after UIA interventional treatment.

[0076] Model 3: CRstalk ≥ 0.5 was included as the HR-VWI indicator. The analysis results showed that aneurysm size and CRstalk ≥ 0.5 were independent risk factors for recurrence after UIA interventional treatment.

[0077] (4) Conclusive determination: Based on the results of the above univariate and multivariate analyses, this example determined that aneurysm wall enhancement (whether it is ring enhancement defined by grade or enhancement quantified by CRstalk value) shown by preoperative high-resolution magnetic resonance vascular wall imaging is an independent and important predictive parameter for predicting the risk of recurrence of small and medium-sized unruptured intracranial aneurysms after endovascular treatment (especially stent coiling).

[0078] Step S2: Determine the target category corresponding to each prediction parameter according to the parameter value of each prediction parameter.

[0079] In an optional embodiment, determining the target category corresponding to each prediction parameter according to the parameter value of each prediction parameter includes:

[0080] For each prediction parameter, a set of classification criteria or category division rules are predefined.

[0081] If the prediction parameter is a continuous variable type, such as aneurysm size, its original parameter value (such as the specific number of millimeters) will be accurately classified into one of the predefined target categories according to the preset numerical range (for example, 3.0mm to 4.9mm is divided into one category, 5.0mm to 6.9mm is divided into another category, 7.0mm to 10.0mm is divided into another category, etc.).

[0082] If the predictive parameter itself is a categorical or ordinal variable, such as treatment type (e.g., simple microcoil embolization or stent-assisted microcoil embolization), immediate postoperative occlusion degree (e.g., MRRC grade I, II, or III), or preoperative high-resolution magnetic resonance angiography (AWE) aneurysm wall enhancement grade (e.g., AWE grade 0, 1, 2, or 3), the actual parameter value of the parameter for the patient to be evaluated (i.e., the specific treatment type, occlusion grade, or enhancement grade) will be directly identified and determined as the corresponding target category.

[0083] Step S3: Determine the recurrence risk score of each prediction parameter according to the target category and based on a preset scoring rule.

[0084] In a preferred embodiment, the preset scoring rules are determined in the following manner:

[0085] Acquiring a plurality of clinical data samples, wherein a clinical data sample includes observation values ​​of a plurality of prediction parameters and corresponding recurrence of unruptured intracranial aneurysms;

[0086] Based on the clinical data samples, a first multivariate regression model is constructed with the recurrence of unruptured intracranial aneurysms corresponding to the observed values ​​as the outcome variable, and a regression coefficient of each prediction parameter is determined according to the first multivariate regression model;

[0087] According to the regression coefficient corresponding to each prediction parameter, based on the preset reference value of each category of each prediction parameter, a preset scoring rule is determined.

[0088] For example, to determine the regression coefficients for each predictive parameter, this approach first performs statistical modeling based on the acquired clinical data samples. During this process, the recurrence of unruptured intracranial aneurysms (UIAs) recorded in each clinical data sample is used as the outcome variable, and the corresponding observed values ​​are selected as independent variables to construct a first multivariate regression model (preferably a multivariate logistic regression model). The core objective of this model is to predict the risk of UIA recurrence after specific endovascular treatments (particularly stent-assisted coiling).

[0089] The predictive parameters included in this first multivariate regression model and their specific treatment were as follows: aneurysm size was directly included as a continuous variable. Treatment was treated as a dichotomous variable based on, for example, the distinction between stent-assisted coil embolization (SAC) and simple microcoil embolization. Similarly, the degree of immediate postoperative occlusion was also treated as a dichotomous variable based on, for example, MRRC grade I (complete occlusion) versus MRRC grade II or III (incomplete occlusion). Preoperative high-resolution magnetic resonance angiography (HR-VWI) aneurysm wall enhancement grade was also included as a dichotomous variable based on, for example, AWE grade 0-1 (none or partial enhancement) versus AWE grade 2-3 (ring enhancement).

[0090] As shown in the table below, through model construction and fitting, the regression coefficient (β value) for each predictive parameter can be determined based on the first multivariate regression model. This regression coefficient (β value) accurately quantifies the independent contribution of the predictive parameter to the risk of UIA recurrence after controlling for the influence of other factors.

[0091] Prediction parameters <![CDATA[Regression coefficient (β i )]]> intercept -7.2962 Aneurysm size 0.7447 Treatment 1.5289 Immediate postoperative occlusion degree 1.1071 Preoperative high-resolution MRI tumor wall enhancement 2.3778

[0092] Specifically, the preset reference value of each category of a prediction parameter is determined by the following method:

[0093] If the current prediction parameter is a continuous variable, the value range of the current prediction parameter is divided into at least two category intervals, and the midpoint value of each category interval is used as the preset reference value of the corresponding category interval;

[0094] If the current prediction parameter is a categorical variable, each category of the current prediction parameter is numerically encoded, and the encoding result is used as the preset reference value of the corresponding category.

[0095] For example, if the prediction parameter is a continuous variable (e.g., aneurysm size), its numerical range is first divided into three pre-set category intervals (e.g., 3.0 mm to 4.9 mm is divided into one category, 5.0 mm to 6.9 mm is divided into another category, and 7.0 mm to 10.0 mm is divided into another category). Subsequently, the midpoint value of each category interval is usually selected as the reference value for the corresponding category interval.

[0096] If the prediction parameter is a categorical variable (e.g., treatment modality, immediate postoperative occlusion degree, or preoperative high-resolution magnetic resonance angiography (HRA) wall enhancement grade), each inherent category of the prediction parameter is numerically encoded, and the resulting encoding result is designated as the reference value for the corresponding category. A commonly used encoding method is binary encoding, in which the category that generally represents the reference standard or lower risk level can be designated as the baseline category and assigned a value of 0, while other relevant categories are assigned a value of 1. For example, among the prediction parameters involved in the present invention, for treatment modality, stent-assisted coil embolization (SAC) can be used as the baseline category (reference value 0), and simple microcoil embolization can be used as the other category (reference value 1); for immediate postoperative occlusion degree, MRRC grade I can be used as the baseline category (reference value 0), and MRRC grades II or III can be used as the other category (reference value 1); for preoperative high-resolution HRA wall enhancement grade, AWE grades 0-1 can be used as the baseline category (reference value 0), and AWE grades 2-3 can be used as the other category (reference value 1), as shown in the following table.

[0097] Prediction parameters Classification Reference value Aneurysm size (mm) 3.0-4.9 4.0 5.0-6.9 6.0 7.0-10.0 8.5 Treatment SAC 0 Simple coil embolization 1 Immediate postoperative occlusion degree MRRC Level I 0 MRRC level II or III 1 Preoperative high-resolution MRI of the tumor wall AWE Level 0-1 0 AWE Level 2-3 1

[0098] In a preferred embodiment, the step of determining a preset scoring rule based on the regression coefficient corresponding to each prediction parameter and the preset reference value of each category of each prediction parameter includes:

[0099] Determine the offset corresponding to each category of each prediction parameter based on the regression coefficient corresponding to each prediction parameter, the preset reference value of each category of each prediction parameter, and the preset reference value of the benchmark category;

[0100] Based on the preset conversion unit, the offset corresponding to each category of each prediction parameter is converted to generate the recurrence risk score corresponding to each category of each prediction parameter;

[0101] The preset scoring rules are determined based on the recurrence risk scores corresponding to each category of each prediction parameter.

[0102] For example, as shown in the table below, the benchmark categories for each prediction parameter are: aneurysm size: 3.0-4.9 mm; treatment method: SAC; immediate postoperative occlusion degree: MRRC grade I; preoperative high-resolution HR-VWI imaging: AWE grade 0-1.

[0103] Prediction parameters Classification Reference value Aneurysm size (mm) 3.0-4.9 <![CDATA[4.0=W 1,REF ]]> 5.0-6.9 6.0 7.0-10.0 8.5 Treatment SAC <![CDATA[0=W 2,REF ]]> Simple coil embolization 1 Immediate postoperative occlusion degree MRRC Level I <![CDATA[0=W 3,REF ]]> MRRC level II or III 1 Preoperative high-resolution MRI tumor wall enhancement AWE Level 0-1 <![CDATA[0=W 4,REF ]]> AWE Level 2-3 1

[0104] Exemplarily, the offset of each category of a prediction parameter is calculated by the following formula:

[0105] D i，j =β i (W i，j -W i，REF )

[0106] Among them, D i，j is the offset of the j-th category of the prediction parameter i; β i is the regression coefficient of the prediction parameter i; W i，j is the reference value of the jth category of the prediction parameter i; W i,REF is the reference value of the benchmark category for the prediction parameter i.

[0107] The offset calculation results for each category of each prediction parameter are shown in the following table.

[0108]

[0109]

[0110] Exemplarily, based on a preset conversion unit, the offset corresponding to each category of each prediction parameter is converted to generate a recurrence risk score corresponding to each category of each prediction parameter, including:

[0111] A constant B is set as the conversion unit, so that each increase of B regression units corresponds to an increase of 1 point. The present invention selects the minimum non-zero offset of the predictive factor of aneurysm size as the conversion unit, B=1.4894.

[0112] The recurrence risk score corresponding to each category of each prediction parameter is calculated using the following formula:

[0113] G i，j =D i，j / B=[β i (W i，j -W i,REF )] / B

[0114] Among them, G i，j is the recurrence risk score of the jth category of the prediction parameter i; D i，j is the offset of the j-th category of the prediction parameter i; β iis the regression coefficient of the prediction parameter i; W i，j is the reference value of the jth category of the prediction parameter i; W i,REF is the reference value of the benchmark category of the prediction parameter i; B is the conversion unit.

[0115] The recurrence risk scores corresponding to each category of each prediction parameter were calculated using the formula and rounded to an integer, as shown in the following table.

[0116]

[0117]

[0118] Exemplarily, based on the recurrence risk scores corresponding to each category of each prediction parameter, a preset scoring rule is determined, including:

[0119] After conversion using the above method, a scoring rule was obtained, which includes four prediction parameters: aneurysm size, treatment method, degree of immediate postoperative occlusion, and preoperative high-resolution magnetic resonance imaging (HRMRI) aneurysm wall enhancement, as shown in the table below.

[0120] Prediction parameters Classification Score Aneurysm size (mm) 3.0-4.9 0 5.0-6.9 1 7.0-10.0 2 Treatment Stent-assisted coil embolization 0 Simple coil embolization 1 Immediate postoperative occlusion degree MRRC Level I 0 MRRC level II or III 1 Preoperative high-resolution MRI tumor wall enhancement AWE Level 0-1 0 AWE Level 2-3 2

[0121] Step S4: determining the recurrence risk probability of the patient's small or medium-sized unruptured intracranial aneurysm after stent coil treatment based on the recurrence risk score of each prediction parameter;

[0122] In an optional embodiment, determining the recurrence risk probability of a small or medium-sized unruptured intracranial aneurysm after stent coiling treatment in a patient to be evaluated based on the recurrence risk score of each prediction parameter includes:

[0123] According to the recurrence risk score of each prediction parameter, the total recurrence risk score of the patient to be evaluated is calculated;

[0124] Based on the preset logistic regression formula, the total recurrence risk score of the evaluated patients was converted into the recurrence risk probability of the evaluated patients after stent-coil treatment of small and medium-sized unruptured intracranial aneurysms.

[0125] In a preferred embodiment, the preset logistic regression formula is generated in the following manner:

[0126] For each clinical data sample, determine the recurrence risk score corresponding to the observed value of each prediction parameter in the current clinical data sample; calculate and generate the total recurrence risk score corresponding to the current clinical data sample based on the recurrence risk score corresponding to the observed value of each prediction parameter;

[0127] The recurrence risk total score was used as the independent variable, and the recurrence of unruptured intracranial aneurysms corresponding to the clinical data sample of the recurrence risk total score was used as the outcome variable to construct a second multivariate regression model, and the intercept of the second multivariate regression model was determined.

[0128] According to the intercept of the second multivariate regression model and the preset conversion unit, a logistic regression formula for converting the total recurrence risk score into the recurrence risk probability is constructed. In the specific implementation, in order to accurately convert the total recurrence risk score of each sample calculated by the scoring scale (for example, in a system of 0 to 6 points) into an individualized recurrence risk probability, this scheme adopts a specific probability calibration step. This step is first based on the existing clinical data sample set, with the total recurrence risk score of each sample as the independent variable, and the recurrence of unruptured intracranial aneurysms corresponding to these samples as the outcome variable, to construct a second multivariate regression model. Through the fitting analysis of this second multivariate regression model, the intercept and regression coefficient of the model are generated, as shown in the following table.

[0129] Regression coefficient intercept -4.2160 Total recurrence risk score 1.3410

[0130] Subsequently, the intercept of the second multivariate regression model was used, combined with the conversion unit (B value) used to standardize the score in the original scoring rule development process (for example, B value ≈ 1.4894), and the individual patient's total recurrence risk score, to construct a logistic regression formula for converting the total recurrence risk score into the recurrence risk probability:

[0131]

[0132] Where P is the probability of recurrence risk; T is the total recurrence risk score; β0 is the intercept of the second multivariate regression model; B is the conversion unit; and e is the base of the natural logarithm, e≈2.7183.

[0133] The total recurrence risk score of the clinical data sample is converted into the recurrence risk probability, as shown in the following table.

[0134] Total recurrence risk score Recurrence risk probability 0 0.0145 1 0.0614 2 0.2249 3 0.5628 4 0.8509 5 0.9620 6 0.9912

[0135] In order to evaluate the predictive ability of the method provided by the present invention for predicting the risk of recurrence after stent coil treatment of small and medium-sized unruptured intracranial aneurysms, the receiver operating characteristic curve (ROC) was used to evaluate the predictive effect of the present invention.

[0136] The specific method is as follows: using the recurrence risk probability (i.e., the probability corresponding to the total sample score) and the actual recurrence outcome variable as input variables, calculate the ROC curve, and estimate its area under the curve (AUC) and 95% confidence interval.

[0137] like Figure 2As shown, the AUC value of the ROC curve of the present invention is 0.893, 95% CI: 0.771-1.000, the optimal threshold probability is 0.394, the sensitivity is 71.4%, and the specificity is 94.5%.

[0138] In addition, to determine the optimal cutoff threshold of the ROC curve of the present invention, the Youden Index was used to determine the optimal cutoff value: Youden Index = Sensitivity + Specificity - 1; the optimal cutoff value is the prediction probability threshold that maximizes the Youden Index. In the present invention, the optimal prediction probability threshold is 0.3938.

[0139] The risk stratification criteria of the present invention are determined based on the optimal prediction probability threshold, with high risk defined as a risk probability greater than or equal to the optimal prediction probability threshold, and low risk defined as a risk probability less than the optimal prediction probability threshold. Combining the total recurrence risk score, recurrence risk probability, and the optimal prediction probability threshold of the present invention corresponding to the clinical data sample, the model risk stratification is determined as follows:

[0140]

[0141]

[0142] As can be seen from the table above, the optimal prediction probability threshold (0.3938) is between the recurrence risk probabilities corresponding to the total recurrence risk scores of 2 (0.2249) and 3 (0.5628). Based on our definition of high risk and low risk, a score of ≥3 is considered high risk, and a score of <3 is considered low risk.

[0143] Compared with the previous rating scales provided by other prediction methods in the prior art,

[0144] The previous rating scale is as follows.

[0145] Prediction parameters Classification Score Aneurysm size (mm) <10.0 0 10.0-25.0 1 >25 3 Treatment SAC 0 Simple coil embolization 1 Immediate postoperative occlusion degree MRRC Level I 0 MRRC level II or III 1

[0146] Convert each total score of the previous scoring scale into the corresponding recurrence risk probability, as shown in the following table:

[0147] Total recurrence risk score Recurrence risk probability Risk stratification 0 0.0187 Low risk 1 0.0722 Low risk 2 0.2410 High risk 3 0.5644 High risk 4 0.8409 High risk 5 0.9556 High risk

[0148] like Figure 3As shown, receiver operating characteristic (ROC) curves were calculated for the prediction methods of the present invention (Model 1) and the prior art (Model 2), and the area under the curve (AUC) and 95% confidence interval (95% CI) were estimated. For the present invention, the AUC value for Model 1 was 0.893, 95% CI: 0.771-1.000, while the AUC value for Model 2 was 0.717, 95% CI: 0.572-0.861. The DeLong test showed that Model 1 had superior discriminatory power compared to Model 2, with a statistically significant difference (P = 0.007).

[0149] The recurrence of small and medium-sized unruptured intracranial aneurysms treated with stent-coil therapy under different scores of the present invention (Model 1) in the data set was calculated, and the actual recurrence rate was calculated.

[0150] Total recurrence risk score Aneurysm (number) Follow-up of recurrent aneurysms (number) Actual recurrence rate Predicting recurrence rate 0 16 1 0.06 0.0145 1 25 0 0.00 0.0614 2 15 3 0.20 0.2249 3 3 2 0.67 0.5628 4 5 3 0.60 0.8509 5 5 5 1.00 0.9620 6 0 0 0.00 0.9912 total 69 14 0.20 /

[0151] like Figure 4 As shown, the recurrence of unruptured intracranial aneurysms after treatment corresponding to the total scores of the recurrence samples of the present invention are as follows: there are 16 aneurysms with a total score of 0, 1 recurrence after stent coil treatment, and the actual recurrence rate is 0.06; there are 25 aneurysms with a total score of 1, 0 recurrence after stent coil treatment, and the actual recurrence rate is 0.00; there are 15 aneurysms with a total score of 2, 3 recurrences after stent coil treatment, and the actual recurrence rate is 0.20; there are 3 aneurysms with a total score of 3, 2 recurrences after stent coil treatment, and the actual recurrence rate is 0.67; there are 5 aneurysms with a total score of 4, 3 recurrences after stent coil treatment, and the actual recurrence rate is 0.60; there are 5 aneurysms with a total score of 5, 5 recurrences after stent coil treatment, and the actual recurrence rate is 1.00; there are 0 aneurysms with a total score of 6, 0 recurrences after stent coil treatment, and the actual recurrence rate is 0.00.

[0152] The risk of recurrence after stent-coil treatment of small and medium-sized unruptured intracranial aneurysms using existing technology was calculated under different scores of the previous model (Model 2), and the actual recurrence rate was calculated, as shown in the following table.

[0153]

[0154]

[0155] like Figure 5As shown, the recurrence of unruptured intracranial aneurysms after treatment corresponding to the total recurrence risk scores of previous models in the prior art are as follows: with a total score of 0, there were 41 aneurysms in total, 4 of which recurred after treatment with stents and coils, and the actual recurrence rate was 0.10; with a total score of 1, there were 26 aneurysms in total, 8 of which recurred after treatment with stents and coils, and the actual recurrence rate was 0.31; with a total score of 2, there were 2 aneurysms in total, 2 of which recurred after treatment with stents and coils, and the actual recurrence rate was 1.00; with a total score of 3, there were 0 aneurysms in total, 0 of which recurred after treatment with stents and coils, and the actual recurrence rate was 0.00; with a total score of 4, there were 0 aneurysms in total, 0 of which recurred after treatment with stents and coils, and the actual recurrence rate was 0.00; with a total score of 5, there were 0 aneurysms in total, 0 of which recurred after treatment with stents and coils, and the actual recurrence rate was 0.00.

[0156] Comparison of the recurrence of each risk stratification in the present invention (Model 1) and the previous model (Model 2) of the prior art is as follows Figure 6 shown.

[0157] Results showed that Model 1 predicted 13 high-risk cases, of which 10 actually relapsed and 3 did not; and predicted 56 low-risk cases, of which 4 actually relapsed and 52 did not. The true positive rate was 0.714 (10 / 14) and the true negative rate was 0.945 (52 / 55). Model 2 predicted 2 high-risk cases, of which 2 actually relapsed and 0 did not; and predicted 67 low-risk cases, of which 12 actually relapsed and 55 did not. The true positive rate was 0.143 (2 / 14) and the true negative rate was 1.000 (55 / 55).

[0158] like Figure 7 As shown, an embodiment of the present invention provides a device for predicting the recurrence risk of small and medium-sized unruptured intracranial aneurysms after stent coil treatment, comprising: a module for acquiring parameters to be evaluated, a module for determining target categories, and a module for outputting the probability of recurrence risk;

[0159] The module for obtaining parameters to be evaluated is used to obtain parameter values ​​of various prediction parameters of the patient to be evaluated; wherein the prediction parameters include aneurysm size, treatment method, degree of occlusion immediately after surgery, and aneurysm wall enhancement grade based on preoperative high-resolution magnetic resonance imaging of the vascular wall;

[0160] The target category determination module is used to determine the target category corresponding to each prediction parameter according to the parameter value of each prediction parameter;

[0161] The recurrence risk probability output module is configured to determine a recurrence risk score for each prediction parameter based on the target category and a preset scoring rule; and to determine a recurrence risk probability for a patient undergoing stent-coil treatment for a small or medium-sized unruptured intracranial aneurysm based on the recurrence risk score for each prediction parameter.

[0162] The preset scoring rules are determined in the following ways:

[0163] Acquiring a plurality of clinical data samples, wherein a clinical data sample includes observation values ​​of a plurality of prediction parameters and corresponding recurrence of unruptured intracranial aneurysms;

[0164] Based on the clinical data samples, a first multivariate regression model is constructed with the recurrence of unruptured intracranial aneurysms corresponding to the observed values ​​as the outcome variable, and a regression coefficient of each prediction parameter is determined according to the first multivariate regression model;

[0165] According to the regression coefficient corresponding to each prediction parameter, based on the preset reference value of each category of each prediction parameter, a preset scoring rule is determined.

[0166] Specifically, the recurrence risk probability output module determines a preset scoring rule based on the regression coefficient corresponding to each prediction parameter and the preset reference value of each category of each prediction parameter, including:

[0167] Determine the offset corresponding to each category of each prediction parameter based on the regression coefficient corresponding to each prediction parameter, the preset reference value of each category of each prediction parameter, and the preset reference value of the benchmark category;

[0168] Based on the preset conversion unit, the offset corresponding to each category of each prediction parameter is converted to generate the recurrence risk score corresponding to each category of each prediction parameter;

[0169] The preset scoring rules are determined based on the recurrence risk scores corresponding to each category of each prediction parameter.

[0170] It should be noted that the embodiment of the device described above corresponds to the above-mentioned embodiment of the present invention, and it can implement the method for predicting the risk of recurrence after treatment of small and medium-sized unruptured intracranial aneurysms with stent coils as described in any of the above-mentioned embodiments of the present invention. In addition, the embodiment of the above-mentioned device is merely schematic, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the embodiment of the device provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which can be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art can understand and implement it without paying any creative work.

[0171] Based on the above method embodiment of the present invention, a corresponding electronic device embodiment is provided.

[0172] An embodiment of the present invention provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the method for predicting the risk of recurrence after treatment of small and medium-sized unruptured intracranial aneurysms with stent coils as described in any one of the present inventions is implemented, or, when the processor executes the computer program, the functions of the modules in the above-mentioned device embodiments are implemented.

[0173] Exemplarily, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

[0174] The terminal device may be a computing device such as a desktop computer, a notebook computer, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0175] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, connecting various parts of the entire terminal device using various interfaces and lines.

[0176] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created based on the use of the mobile phone, etc. In addition, the memory can include a high-speed random access memory and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0177] Based on the above method embodiment, the present invention provides a corresponding storage medium embodiment;

[0178] Another embodiment of the present invention provides a storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute any of the above-mentioned methods for predicting the risk of recurrence after treatment of small and medium-sized unruptured intracranial aneurysms with stent coils of the present invention.

[0179] The storage medium is a computer-readable storage medium, and the computer program includes computer program code, which may be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium may include any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunications signal, and a software distribution medium.

[0180] In the description of this specification, the reference terms "one embodiment," "some embodiments," "example," "specific example," or "some examples" mean that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials, or characteristics described may be combined in any appropriate manner in any one or more embodiments or examples. In addition, those skilled in the art may combine and integrate different embodiments or examples described in this specification, as well as features of different embodiments or examples, unless they are mutually inconsistent.

[0181] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for predicting the risk of recurrence of small to medium-sized unruptured intracranial aneurysms after stent-coil treatment, characterized in that: include: Obtaining parameter values ​​for each prediction parameter of the patient to be evaluated; wherein the prediction parameters include aneurysm size, treatment method, degree of occlusion immediately after surgery, and aneurysm wall enhancement grade as determined by preoperative high-resolution magnetic resonance angiography; Determining the target category corresponding to each prediction parameter according to the parameter value of each prediction parameter; Determining a recurrence risk score for each prediction parameter based on the target category and a preset scoring rule; Based on the recurrence risk scores of each prediction parameter, the recurrence risk probability of the patients undergoing stent-coil treatment for small and medium-sized unruptured intracranial aneurysms was determined; The preset scoring rules are determined in the following ways: Acquiring a plurality of clinical data samples, wherein a clinical data sample includes observation values ​​of a plurality of prediction parameters and corresponding recurrence of unruptured intracranial aneurysms; Based on the clinical data samples, a first multivariate regression model is constructed with the recurrence of unruptured intracranial aneurysms corresponding to the observed values ​​as the outcome variable, and a regression coefficient of each prediction parameter is determined according to the first multivariate regression model; According to the regression coefficient corresponding to each prediction parameter, based on the preset reference value of each category of each prediction parameter, a preset scoring rule is determined.

2. A method for predicting the risk of recurrence of small to medium-sized unruptured intracranial aneurysms after stent coil treatment as claimed in claim 1, characterized in that: The preset reference values ​​for each category of a prediction parameter are determined by: If the current prediction parameter is a continuous variable, the value range of the current prediction parameter is divided into at least two category intervals, and the midpoint value of each category interval is used as the preset reference value of the corresponding category interval; If the current prediction parameter is a categorical variable, each category of the current prediction parameter is numerically encoded, and the encoding result is used as the preset reference value of the corresponding category.

3. The method for predicting the risk of recurrence of small to medium-sized unruptured intracranial aneurysms after stent coil treatment according to claim 2, characterized in that: The step of determining a preset scoring rule based on the regression coefficient corresponding to each prediction parameter and the preset reference value of each category of each prediction parameter includes: Determine the offset corresponding to each category of each prediction parameter based on the regression coefficient corresponding to each prediction parameter, the preset reference value of each category of each prediction parameter, and the preset reference value of the benchmark category; Based on the preset conversion unit, the offset corresponding to each category of each prediction parameter is converted to generate the recurrence risk score corresponding to each category of each prediction parameter; The preset scoring rules are determined based on the recurrence risk scores corresponding to each category of each prediction parameter.

4. The method for predicting the risk of recurrence of small to medium-sized unruptured intracranial aneurysms after stent coil treatment according to claim 3, characterized in that: The step of determining the offset corresponding to each category of each prediction parameter according to the regression coefficient corresponding to each prediction parameter, the preset reference value of each category of each prediction parameter, and the preset reference value of the benchmark category includes: For each category of the current prediction parameter, calculate the difference between the preset reference value corresponding to the current category and the reference value of the preset benchmark category to generate the offset difference corresponding to the current category; multiply the offset difference corresponding to the current category by the regression coefficient of the current prediction parameter as the offset of the current category of the current prediction parameter.

5. The method for predicting the risk of recurrence of small to medium-sized unruptured intracranial aneurysms after stent coil treatment according to claim 4, characterized in that: The recurrence risk probability of patients undergoing stent-coil treatment for small and medium-sized unruptured intracranial aneurysms is determined based on the recurrence risk score of each predictive parameter, including: According to the recurrence risk score of each prediction parameter, the total recurrence risk score of the patient to be evaluated is calculated; Based on the preset logistic regression formula, the total recurrence risk score of the evaluated patients was converted into the recurrence risk probability of the evaluated patients after stent-coil treatment of small and medium-sized unruptured intracranial aneurysms.

6. The method for predicting the risk of recurrence of small to medium-sized unruptured intracranial aneurysms after stent coil treatment according to claim 5, characterized in that: Generate the preset logistic regression formula in the following way: For each clinical data sample, determine the recurrence risk score corresponding to the observed value of each prediction parameter in the current clinical data sample; calculate and generate the total recurrence risk score corresponding to the current clinical data sample based on the recurrence risk score corresponding to the observed value of each prediction parameter; The recurrence risk total score was used as the independent variable, and the recurrence of unruptured intracranial aneurysms corresponding to the clinical data sample of the recurrence risk total score was used as the outcome variable to construct a second multivariate regression model, and the intercept of the second multivariate regression model was determined. According to the intercept of the second multivariate regression model and the preset conversion unit, a logistic regression formula for converting the total recurrence risk score into the recurrence risk probability was constructed.

7. A device for predicting the risk of recurrence of small to medium-sized unruptured intracranial aneurysms after stent-coil treatment, characterized in that: include: Module for obtaining parameters to be evaluated, module for determining target categories, and module for outputting recurrence risk probability; The module for obtaining parameters to be evaluated is used to obtain parameter values ​​of various prediction parameters of the patient to be evaluated; wherein the prediction parameters include aneurysm size, treatment method, degree of occlusion immediately after surgery, and aneurysm wall enhancement grade based on preoperative high-resolution magnetic resonance imaging of the vascular wall; The target category determination module is used to determine the target category corresponding to each prediction parameter according to the parameter value of each prediction parameter; The recurrence risk probability output module is configured to determine a recurrence risk score for each prediction parameter based on the target category and a preset scoring rule; and to determine a recurrence risk probability for a patient undergoing stent-coil treatment for a small or medium-sized unruptured intracranial aneurysm based on the recurrence risk score for each prediction parameter. The preset scoring rules are determined in the following ways: Acquiring a plurality of clinical data samples, wherein a clinical data sample includes observation values ​​of a plurality of prediction parameters and corresponding recurrence of unruptured intracranial aneurysms; Based on the clinical data samples, a first multivariate regression model is constructed with the recurrence of unruptured intracranial aneurysms corresponding to the observed values ​​as the outcome variable, and a regression coefficient of each prediction parameter is determined according to the first multivariate regression model; According to the regression coefficient corresponding to each prediction parameter, based on the preset reference value of each category of each prediction parameter, a preset scoring rule is determined.

8. The device for predicting the risk of recurrence of small and medium-sized unruptured intracranial aneurysms after stent coil treatment according to claim 7, characterized in that: The recurrence risk probability output module determines a preset scoring rule based on the regression coefficient corresponding to each prediction parameter and the preset reference value of each category of each prediction parameter, including: Determine the offset corresponding to each category of each prediction parameter based on the regression coefficient corresponding to each prediction parameter, the preset reference value of each category of each prediction parameter, and the preset reference value of the benchmark category; Based on the preset conversion unit, the offset corresponding to each category of each prediction parameter is converted to generate the recurrence risk score corresponding to each category of each prediction parameter; The preset scoring rules are determined based on the recurrence risk scores corresponding to each category of each prediction parameter.

9. An electronic device, characterized in that: It comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements the method for predicting the risk of recurrence after treatment of small and medium-sized unruptured intracranial aneurysms with stent coils as described in any one of claims 1 to 6.

10. A storage medium, characterized in that: The storage medium includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the method for predicting the recurrence risk after treatment of small and medium-sized unruptured intracranial aneurysms with stent coils as described in any one of claims 1 to 6.

Images