Hemorrhagic transformation risk prediction method, device, equipment and medium

Abstract

The invention relates to the technical field of intelligent diagnosis of acute ischemic stroke, in particular to a hemorrhagic transformation risk prediction method, device and equipment and a medium, and the method comprises the steps: obtaining magnetic resonance imaging data of brain tissues of a target object and clinical baseline data of the target object; determining iconography characteristics of brain tissues of the target object according to the magnetic resonance imaging data; determining clinical baseline features of the target object according to the clinical baseline data; and determining a hemorrhagic transformation risk prediction result of the target object based on the iconography features and the clinical baseline features. According to the hemorrhagic transformation risk prediction method, the accuracy of HT risk prediction can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent diagnosis of acute ischemic stroke, and in particular to a method, device, equipment and medium for predicting the risk of hemorrhagic transformation. Background Art

[0002] Acute ischemic stroke (AIS) is one of the main causes of disability, which has a wide impact on individuals, families and society. One of the main complications of AIS patients is hemorrhagic transformation (HT), which will seriously affect the patient's prognosis.

[0003] The occurrence of HT is affected by multiple factors, such as blood-brain barrier disruption, neuroinflammation, immune system activation, reperfusion therapy (such as intravenous thrombolysis with tissue-type plasminogen activator or mechanical thrombectomy), and other baseline characteristics (such as age, hypertension, atrial fibrillation, and diabetes). Studies have shown that patients receiving AIS treatment are more likely to develop HT, accompanied by a higher mortality rate. Therefore, reliably predicting the risk of HT in AIS patients is crucial to guide treatment decisions and improve patient prognosis.

[0004] At present, magnetic resonance imaging (MRI) technology has been widely used in the evaluation of AIS patients, especially diffusion weighted imaging (DWI) and perfusion weighted imaging (PWI), which have been shown to have extremely high sensitivity in diagnosing cerebral infarction, but research based on these images for predicting HT risk is still insufficient. At present, the HT prediction method based on existing images still has certain limitations, and the prediction accuracy is low, which is difficult to meet clinical needs. Summary of the invention

[0005] In view of the above problems in the prior art, the object of the present invention is to provide a method, device, equipment and medium for predicting the risk of hemorrhagic transformation, which can improve the accuracy of HT risk prediction.

[0006] In order to solve the above problems, the present invention provides a method for predicting the risk of hemorrhagic transformation, which comprises: obtaining magnetic resonance imaging data of brain tissue of the target subject and clinical baseline data of the target subject; determining imaging features of brain tissue of a target subject based on magnetic resonance imaging data; Determine the clinical baseline characteristics of the target subjects based on clinical baseline data; Based on the imaging features and clinical baseline characteristics, the hemorrhagic transformation risk prediction results of the target subjects were determined.

[0007] Optionally, the magnetic resonance imaging data include perfusion-weighted imaging data, and the clinical baseline data include NIHSS score; Determine the imaging features of the target subject's brain tissue based on magnetic resonance imaging data, including: determining a maximum time image of residual function of brain tissue of the target subject based on the perfusion-weighted imaging data; Based on the maximum residual function time image, the imaging characteristics of the target subject's brain tissue are determined.

[0008] Furthermore, the imaging characteristics of the target subject's brain tissue are determined based on the residual function maximum time image, including: The volume of the low perfusion area in the brain tissue of the target object and the corresponding low perfusion intensity ratio of the brain tissue of the target object are determined according to the maximum residual function time image.

[0009] Optionally, the magnetic resonance imaging data includes perfusion weighted imaging data and diffusion weighted imaging data; Determine the imaging features of the target subject's brain tissue based on magnetic resonance imaging data, including: determining a maximum time image of residual function of brain tissue of the target subject based on the perfusion-weighted imaging data; Determining an apparent diffusion coefficient image of the target subject's brain tissue based on the diffusion-weighted imaging data; Based on at least the residual function maximum time image and the apparent diffusion coefficient image, the imaging characteristics of the brain tissue of the target subject are determined.

[0010] Further, based at least on the residual function maximum time image and the apparent diffusion coefficient image, the imaging characteristics of the brain tissue of the target object are determined, including: determining the volume of the infarct area in the brain tissue of the target subject according to the apparent diffusion coefficient image; Based at least on the residual function maximum time image, the volume of the target object's brain tissue having a TMAX value greater than each of at least one preset threshold, the volume of the perfusion-diffusion mismatch area, and the corresponding low perfusion intensity ratio of the target object's brain tissue are determined respectively.

[0011] Furthermore, clinical baseline data included NIHSS scores, attribute data, and disease history data; Determine the clinical baseline characteristics of the target subjects based on clinical baseline data, including: The attribute data and the disease history data are encoded and / or standardized respectively to obtain the attribute characteristics and the disease history characteristics of the target object.

[0012] Furthermore, based on the imaging features and clinical baseline characteristics, the hemorrhagic transformation risk prediction results of the target subject are determined, including: Obtaining a low perfusion intensity ratio corresponding to the brain tissue of the target object; Determine the target subject's risk level of hemorrhagic transformation based on the low perfusion intensity ratio; The hemorrhagic transformation risk prediction model corresponding to the hemorrhagic transformation risk level is used to process the imaging characteristics and clinical baseline characteristics to obtain the hemorrhagic transformation risk prediction results of the target object.

[0013] Another aspect of the present invention provides a device for predicting the risk of hemorrhagic transformation, the device comprising: A data acquisition module, used to acquire magnetic resonance imaging data of the target subject's brain tissue and clinical baseline data of the target subject; A first feature determination module, used to determine the imaging features of the target object's brain tissue according to the magnetic resonance imaging data; A second feature determination module is used to determine the clinical baseline features of the target subject based on the clinical baseline data; The risk prediction module is used to determine the hemorrhagic transformation risk prediction results of the target object based on imaging characteristics and clinical baseline characteristics.

[0014] Another aspect of the present invention provides an electronic device, including a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the hemorrhagic transformation risk prediction method as described above.

[0015] On the other hand, the present invention provides a computer-readable storage medium, which stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by a processor to implement the hemorrhagic transformation risk prediction method as described above.

[0016] Another aspect of the present invention provides a computer program product, comprising a computer program / instruction, which implements the above-mentioned hemorrhagic transformation risk prediction method when executed by a processor.

[0017] Due to the above technical solution, the present invention has the following beneficial effects: According to the hemorrhagic transformation risk prediction method of the embodiment of the present invention, HT risk prediction is performed on the target object based on the imaging characteristics of the brain tissue of the target object and the clinical baseline characteristics of the target object. Therefore, the risk of hemorrhagic transformation can be quickly and accurately predicted during the treatment of AIS patients, and the accuracy of HT risk prediction is further improved, thereby providing doctors with reliable decision support, improving diagnostic efficiency, and improving the treatment effect of AIS patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments or prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 is a schematic diagram of a HT risk prediction system provided by an embodiment of the present invention; Figure 2 is a flow chart of a method for predicting risk of hemorrhagic transformation provided by one embodiment of the present invention; Figure 3 is a schematic diagram of the structure of a fusion model based on prior knowledge provided by an embodiment of the present invention; Figure 4 is a flow chart of a method for predicting risk of hemorrhagic transformation provided by another embodiment of the present invention; Figure 5 It is a schematic diagram of the structure of a device for predicting the risk of hemorrhagic transformation provided by one embodiment of the present invention; Figure 6 is a block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, device, product or equipment that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0022] The method provided in the embodiment of the present invention can be applied to the scenario of predicting the HT risk of AIS patients. At present, the HT risk of AIS patients is mainly predicted by using a prediction model based on MRI data (such as PWI images and DWI images). Since a large number of imaging markers related to HT have not been fully studied, the prediction model based on existing imaging methods has certain limitations.

[0023] Existing prediction models based on a single variable (such as clinical baseline characteristics or MRI parameters of patients) show low sensitivity and specificity in predicting HT, and the prediction accuracy of the model is difficult to meet clinical needs. Since the occurrence of HT is the result of the combined action of multiple factors, the prediction model based on a single variable cannot fully consider the complex interactions between different factors.

[0024] In addition, although machine learning technology has been introduced into the medical field in recent years to explore HT risk prediction, existing models often rely on only a single algorithm and lack the integration and comprehensive use of multiple algorithms, resulting in instability and deviation in practical applications.

[0025] The embodiment of the present invention proposes a method for predicting the risk of hemorrhagic transformation. By integrating multiple machine learning algorithms and combining the patient's clinical characteristics and multiple imaging parameters, the HT risk of AIS patients can be accurately predicted, the deficiencies in the existing technology can be overcome, and the accuracy and reliability in clinical applications can be improved, thereby providing doctors with a more intelligent and efficient decision support tool to help timely assess the patient's HT risk, thereby optimizing the treatment strategy and improving the patient's prognosis.

[0026] In order to make the purpose, technical solution and advantages disclosed in the embodiments of the present invention more clearly understood, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the embodiments of the present invention and are not used to limit the embodiments of the present invention.

[0027] Reference Manual Attached Figure 1 , which shows a schematic diagram of a HT risk prediction system provided by an embodiment of the present invention. Figure 1 As shown, the HT risk prediction system may include at least one medical scanning device 110 and a computer device 120. The computer device 120 and each medical scanning device 110 may be directly or indirectly connected via wired or wireless communication, which is not limited in this embodiment of the present invention.

[0028] The medical scanning device 110 may be, but is not limited to, a magnetic resonance imaging device, etc. The computer device 120 may be, but is not limited to, various servers, personal computers, laptops, smart phones, tablet computers, and portable wearable devices. The server may be an independent server or a server cluster or distributed system composed of multiple servers. It may also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0029] The computer device 120 can obtain medical imaging data (such as perfusion weighted imaging data, diffusion weighted imaging data, etc.) of the target object's brain tissue scanned by the medical scanning device 110, as well as the target object's clinical baseline data, and determine the target object's hemorrhagic transformation risk prediction result through the hemorrhagic transformation risk prediction method provided in an embodiment of the present invention.

[0030] In practical applications, the computer device 120 can provide accurate HT risk prediction results within a short period of time after the MRI scan, helping doctors make clinical decisions within a critical time window. The computer device 120 can be integrated into the hospital's imaging diagnosis platform as a real-time decision support tool to improve diagnostic efficiency and improve patient treatment outcomes.

[0031] It should be noted that Figure 1 It is just an example. Those skilled in the art will understand that although Figure 1 Only one medical scanning device 110 is shown, but this does not constitute a limitation on the embodiment of the present invention, and more or less medical scanning devices 110 than shown in the figure may be included.

[0032] Reference Manual Attached Figure 2 , which shows the process of a method for predicting the risk of hemorrhagic transformation provided by an embodiment of the present invention, which can be applied to Figure 1 In the computer device 120. Specifically, Figure 2 As shown, the method may include the following steps: S210: Acquire magnetic resonance imaging data of the target subject's brain tissue and clinical baseline data of the target subject.

[0033] In the embodiment of the present invention, the image of the target brain tissue can be collected by MRI technology to obtain MRI data, wherein the target brain tissue may be a patient who may suffer from AIS.

[0034] In an embodiment of the present invention, clinical baseline data of the target object may also be collected. The clinical baseline data may include, but is not limited to, one or more of the National Institutes of Health Stroke Scale (NIHSS) score of the target object upon admission, attribute data of the target object, and disease history data (e.g., hypertension, diabetes, atrial fibrillation, etc.).

[0035] It should be noted that the source of MRI data and clinical baseline data can be directly imported related data, or obtained from other resource libraries through real-time configuration connections, or obtained from stored image databases after searching based on information such as the patient's name, etc. The embodiment of the present invention does not limit this.

[0036] In some embodiments, the MRI data may include but is not limited to perfusion weighted imaging (PWI) data, and the clinical baseline data may include but is not limited to NIHSS score.

[0037] In some embodiments, the MRI data may include, but are not limited to, PWI data and DWI data, and the clinical baseline data may include, but are not limited to, NIHSS score, attribute data, and disease history data.

[0038] S220: Determine imaging features of the target object's brain tissue according to the magnetic resonance imaging data.

[0039] In the embodiment of the present invention, the acquired MRI data may be processed to obtain imaging features of the brain tissue of the target object.

[0040] Specifically, when different MRI data are acquired, different imaging features can be processed.

[0041] In some embodiments, when the MRI data only includes PWI data, determining the imaging characteristics of the target object's brain tissue based on the magnetic resonance imaging data may include: determining the residual function maximum time (Time to Maximum Enhancement, TMAX) image of the target object's brain tissue based on the perfusion weighted imaging data; determining the imaging characteristics of the target object's brain tissue based on the residual function maximum time image. Specifically, the imaging characteristics of the target object's brain tissue may include, but are not limited to, the volume of the hypoperfusion area and the hypoperfusion intensity ratio (HIR).

[0042] In the embodiment of the present invention, the PWI data can be processed by an existing processing method to obtain a TMAX image, for example, the PWI data can be processed by RAPID software to obtain a TMAX image, which will not be described in detail in the embodiment of the present invention. The TMAX image can include the time when the blood storage function of each voxel in the target brain tissue reaches the maximum value.

[0043] In an embodiment of the present invention, determining the imaging characteristics of the target object's brain tissue based on the maximum time image of residual function may include: determining the volume of the low perfusion area in the target object's brain tissue and the corresponding low perfusion intensity ratio of the target object's brain tissue based on the maximum time image of residual function.

[0044] Specifically, the brain tissue area with a TMAX value > 6s can be determined as a low perfusion area based on the TMAX image, and the volume of the area (i.e., the low perfusion area volume) can be determined. The brain tissue area with a TMAX value > 10s can also be determined based on the TMAX image, and the ratio of the volume of the brain tissue area with a TMAX value > 10s to the volume of the low perfusion area can be calculated as the HIR.

[0045] In some embodiments, when the MRI data includes PWI data and DWI data, determining the imaging characteristics of the target object's brain tissue based on the magnetic resonance imaging data may include: determining the residual function maximum time image of the target object's brain tissue based on the perfusion weighted imaging data; determining the apparent diffusion coefficient (Apparent Diffusion Coefficient, ADC) image of the target object's brain tissue based on the diffusion weighted imaging data; and determining the imaging characteristics of the target object's brain tissue based at least on the residual function maximum time image and the apparent diffusion coefficient image.

[0046] Specifically, the imaging features of the brain tissue of the target object may include, but are not limited to, the volume of the infarct area, the volume of each preset threshold value of the TMAX value greater than at least one preset threshold value, the volume of the perfusion-diffusion mismatch area, and the low perfusion intensity ratio. Among them, at least one preset threshold value can be preset according to actual conditions, for example, it can be set to 4s, 6s, 8s, 10s, etc., and the embodiment of the present invention does not limit this.

[0047] In the embodiment of the present invention, the method for determining the TMAX image is the same as that in the case where the MRI data only includes PWI data, and the embodiment of the present invention will not be described in detail herein.

[0048] In the embodiment of the present invention, the DWI data can be processed by an existing processing method to obtain an ADC image, for example, the DWI data can be processed by RAPID software to obtain an ADC image, which will not be described in detail in the embodiment of the present invention. The ADC image can include the diffusion speed of water molecules in each voxel in the target brain tissue.

[0049] In an embodiment of the present invention, determining the imaging characteristics of the target object's brain tissue based at least on the residual function maximum time image and the apparent diffusion coefficient image may include: determining the volume of the infarct area in the target object's brain tissue according to the apparent diffusion coefficient image; and determining, based at least on the residual function maximum time image, the volume of the target object's brain tissue having a TMAX value greater than each preset threshold value of at least one preset threshold value, the volume of the perfusion-diffusion mismatch area, and the low perfusion intensity ratio corresponding to the target object's brain tissue.

[0050] Specifically, it can be determined based on the ADC image that ADC<620×10 -6 mm 2 / s as the infarct area, and determine the volume of the area (i.e., the volume of the infarct area). According to the TMAX image, the brain tissue areas with TMAX values ​​greater than each preset threshold value in at least one preset threshold value can be determined, and the volumes of these areas can be determined. For example, when at least one preset threshold value is set to 4s, 6s, 8s, and 10s, the brain tissue areas with TMAX values>4s, TMAX values>6s, TMAX values>8s, and TMAX values>10s can be determined respectively, and the volumes of these areas can be determined.

[0051] Specifically, the difference between the volume of the brain tissue region with a TMAX value > 6s (i.e., the volume of the low perfusion region) and the volume of the infarct region can be calculated as the volume of the perfusion-diffusion mismatch region. The ratio of the volume of the brain tissue region with a TMAX value > 10s to the volume of the brain tissue region with a TMAX value > 6s can be calculated as the HIR.

[0052] S230: Determine clinical baseline characteristics of the target subject based on the clinical baseline data.

[0053] In the embodiment of the present invention, the acquired clinical baseline data may be processed to obtain the clinical baseline characteristics of the target object.

[0054] Specifically, when different clinical baseline data are obtained, different clinical baseline characteristics can be processed, wherein the clinical baseline characteristics may include but are not limited to NIHSS scores, attribute characteristics, and disease history characteristics.

[0055] In some embodiments, when the acquired clinical baseline data only includes the NIHSS score, the NIHSS score can be directly used as the clinical baseline feature of the target object.

[0056] In some embodiments, when the acquired clinical baseline data includes not only the NIHSS score but also attribute data and disease history data, determining the clinical baseline characteristics of the target object based on the clinical baseline data may include: encoding and / or standardizing the attribute data and disease history data respectively to obtain the attribute characteristics and disease history characteristics of the target object.

[0057] Specifically, the attribute data of the target object may include the gender of the target object, and the disease history data may include whether the target object suffers from hypertension, diabetes, atrial fibrillation, and other diseases. The gender of the target object may be binarized, with females coded as 0 and males coded as 1. The disease history of the target object may also be binarized, with no disease history coded as 0 and disease history coded as 1. The NIHSS score may be directly used as the clinical baseline feature of the target object.

[0058] In some embodiments, the attribute data and the disease history data may also be standardized to eliminate the dimension and magnitude differences of the data so that different features have the same scale. The specific standardization method may refer to the prior art, and the embodiments of the present invention will not be described in detail here.

[0059] It can be understood that the embodiments of the present invention can convert discrete attributes into numerical representations that can be used by the model by encoding the acquired clinical baseline data, thereby facilitating the subsequent use of machine learning algorithms to predict HT risks.

[0060] S240: Determine the hemorrhagic transformation risk prediction results of the target subject based on imaging features and clinical baseline characteristics.

[0061] In an embodiment of the present invention, imaging features and clinical baseline features can be combined to predict the HI risk of target brain tissue using a machine learning algorithm. Specifically, the obtained imaging features and clinical baseline features can be directly input into a pre-trained hemorrhagic transformation risk prediction model, which can process the input imaging features and clinical baseline features to obtain and output the hemorrhagic transformation risk prediction result of the target object.

[0062] It can be understood that the hemorrhagic transformation risk prediction method according to the embodiment of the present invention predicts the HT risk of the target object based on the imaging characteristics of the brain tissue of the target object and the clinical baseline characteristics of the target object, which can quickly and accurately predict the risk of hemorrhagic transformation during the treatment of AIS patients, further improving the accuracy of HT risk prediction, thereby providing doctors with reliable decision support, improving diagnostic efficiency, and improving the treatment effect of AIS patients.

[0063] In some embodiments, when different MRI data and clinical baseline data are obtained, different hemorrhagic transformation risk prediction models may be used. For example, when the acquired MRI data only includes PWI data, and / or the acquired clinical baseline data only includes NIHSS score, the hemorrhagic transformation risk prediction model may be a logistic regression model. When the acquired MRI data includes PWI data and DWI data, and the acquired clinical baseline data includes NIHSS score, attribute data, and disease history data, the hemorrhagic transformation risk prediction model may be a fusion model based on prior knowledge.

[0064] In some embodiments, when the acquired MRI data only includes PWI data, and / or the acquired clinical baseline data only includes NIHSS score, the obtained imaging features (including low perfusion area volume and HIR) and clinical baseline features (including NIHSS score) can be input into a pre-trained Logistic regression model to obtain the hemorrhagic transformation risk prediction result output by the model.

[0065] In practical applications, since the Logistic regression model has the advantages of simplicity and efficiency in clinical operations, and the model structure is simple and suitable for rapid deployment and application in clinical environments, the Logistic regression model can be selected for HT risk prediction. Specifically, the Logistic regression model uses only three features, namely, the volume of the low perfusion area (the volume of the brain tissue area with a TMAX value>6s), HIR, and NIHSS score. That is, the input data of the Logistic regression model may include the volume of the low perfusion area, HIR, and NIHSS score. The output data of the model may be an HT risk score, with a score range of 0 to 1. When the score output by the Logistic regression model is greater than or equal to the preset score, it indicates that the HT risk is high, and when it is less than the preset score, it indicates that the HT risk is low. Among them, the preset score can be pre-set according to the actual situation, for example, it can be set to 0.5, and the embodiment of the present invention does not limit this.

[0066] Specifically, the Logistic regression model can be obtained by using the PWI sample data determined by the MRI data of multiple patients with AIS and the corresponding NIHSS score samples as training sample data, training by the maximum likelihood estimation method, and optimizing the model parameters using the 5-fold cross validation technology. Alternatively, the PWI sample data determined by the MRI data of multiple patients with AIS and the corresponding NIHSS score samples, and the PWI sample data determined by the MRI data of multiple normal brain tissues (i.e., brain tissues of healthy people) and the corresponding NIHSS score samples can be used as training sample data, training by the maximum likelihood estimation method, and optimizing the model parameters using the 5-fold cross validation technology. The specific model training method can refer to the prior art, and the embodiments of the present invention will not be repeated here.

[0067] It can be understood that the embodiment of the present invention can achieve high prediction accuracy by using a Logistic regression model, using only a small number of key features (such as low perfusion area volume, HIR and NIHSS score), avoiding dependence on too much data, and reducing the complexity of clinical data collection and processing. Thus, the model simplifies the operation process while ensuring high prediction accuracy, improves the real-time and usability of the overall system, and facilitates rapid clinical deployment and application. Doctors can quickly assess the HT risk of patients in clinical scenarios, thereby optimizing the treatment decision-making process, reducing waiting time, and improving medical efficiency.

[0068] In some embodiments, the fusion model based on prior knowledge can be composed of two sub-models and a truncation feature, and the truncation feature can be an image feature and has a corresponding truncation threshold. The fusion model based on prior knowledge can divide the object to be tested into different hemorrhagic transformation risk levels based on the truncation feature, and use different models to predict the risk of hemorrhagic transformation.

[0069] It can be understood that the embodiments of the present invention can combine the advantages of multiple machine learning algorithms by developing a fusion model based on prior knowledge, classifying based on the patient's imaging features, and then applying different sub-models to predict HT risk. Compared with a single model, it can further improve the accuracy, sensitivity and specificity of HT risk prediction, improve the prediction ability of the overall system in complex situations, significantly reduce the deviation and misjudgment in single model prediction, enhance the robustness of the model, and is suitable for deployment and application in complex clinical scenarios.

[0070] Specifically, in conjunction with the reference manual Figure 3, the cutoff feature of the fusion model based on prior knowledge can be the HIR value, and the two sub-models can be the Light Gradient Boosting Machine (LightGBM) model and the Extreme Gradient Boosting (XGBoost) model. In other words, the fusion model based on prior knowledge can divide the object to be tested into different hemorrhagic transformation risk levels based on the HIR value, and use the LightGBM model and the XGBoost model to predict the risk of hemorrhagic transformation.

[0071] Specifically, in conjunction with the reference manual Figure 4 When the acquired MRI data includes PWI data and DWI data, and the acquired clinical baseline data includes NIHSS score, attribute data, and disease history data, the hemorrhagic transformation risk prediction result of the target object is determined based on the imaging characteristics and clinical baseline characteristics, which may include: S241: Obtaining the low perfusion intensity ratio corresponding to the brain tissue of the target object.

[0072] Specifically, the HIR corresponding to the brain tissue of the target object calculated in the above step S220 may be directly obtained.

[0073] S242: Determine the hemorrhagic transformation risk level of the target object according to the low perfusion intensity ratio.

[0074] Specifically, after obtaining the HIR corresponding to the brain tissue of the target object, it can be determined whether the HIR is greater than the cutoff threshold. When the HIR is greater than or equal to the cutoff threshold, the hemorrhagic transformation risk level of the target object is determined to be high risk, and when the HIR is less than the cutoff threshold, the hemorrhagic transformation risk level of the target object is determined to be low risk. The cutoff threshold can be preset according to the actual situation of the cutoff feature, for example, when the cutoff feature is the HIR value, it can be set to 0.5, which is not limited in the embodiment of the present invention.

[0075] S243: Using a hemorrhagic transformation risk prediction model corresponding to the hemorrhagic transformation risk level, the imaging features and clinical baseline characteristics are processed to obtain a hemorrhagic transformation risk prediction result for the target object.

[0076] Specifically, when the hemorrhagic transformation risk level of the target object is high risk, the XGBoost model can be used to predict the hemorrhagic transformation risk. When the hemorrhagic transformation risk level of the target object is low risk, the LightGBM model can be used to predict the hemorrhagic transformation risk.

[0077] Specifically, the obtained imaging features (including the volume of the infarct area, the volume of each preset threshold with a TMAX value greater than at least one preset threshold, the volume of the perfusion-diffusion mismatch area, and HIR) and clinical baseline characteristics (including NIHSS score, attribute characteristics, and disease history characteristics) can be input into a pre-trained XGBoost model or LightGBM model to obtain the hemorrhagic transformation risk prediction results output by the model.

[0078] Specifically, the XGBoost model and the LightGBM model can be trained with PWI sample data and DWI sample data determined by MRI data of multiple patients with AIS, as well as the NIHSS score samples, attribute sample data and disease history sample data corresponding to the patients as training sample data, and trained using the 5-fold cross-validation technology, and the hyperparameters are adjusted by random grid search to ensure the optimal performance of the model. Alternatively, the PWI sample data and DWI sample data determined by MRI data of multiple patients with AIS, as well as the corresponding NIHSS score samples, attribute sample data and disease history sample data, and the PWI sample data and DWI sample data determined by MRI data of multiple normal brain tissues (i.e., brain tissues of healthy people), as well as the corresponding NIHSS score samples, attribute sample data and disease history sample data can be used as training sample data, and trained using the 5-fold cross-validation technology, and the hyperparameters are adjusted by random grid search to ensure the optimal performance of the model. The specific model training method can refer to the prior art, and the embodiments of the present invention will not be repeated here.

[0079] According to the experiment, the fusion model based on prior knowledge significantly improved the accuracy of HT risk prediction for AIS patients by combining two efficient machine learning algorithms, LightGBM and XGBoost, and integrating HIR as the classification standard. The area under the curve (AUC) value of this model on the test set reached 0.9161, which is better than the prediction effect of a single model. This model helps doctors better identify high-risk patients and take timely treatment measures.

[0080] It can be understood that the fusion model based on prior knowledge classifies patients based on the HIR index to flexibly select the appropriate sub-model (LightGBM model or XGBoost model) for HT risk prediction. This design not only improves the prediction flexibility of the model, enabling the model to better cope with different types of AIS patients, thereby obtaining more personalized prediction results, further improving the accuracy of HT risk prediction, but also greatly enhancing the adaptability of the model in actual clinical applications.

[0081] In addition, the multi-model fusion technology reduces the prediction errors and deviations that may be caused by a single model, further improving the stability and robustness of the model, ensuring that high predictive performance can be maintained in complex clinical scenarios.

[0082] It should be noted that when the acquired MRI data includes PWI data and DWI data, and the acquired clinical baseline data includes NIHSS score, attribute data and disease history data, the Logistic regression model can also be used for HT risk prediction according to the user's choice, and the embodiment of the present invention is not limited to this.

[0083] In practical applications, when a higher prediction real-time performance is required, the Logistic regression model can be used to predict HT risk. When a higher prediction accuracy is required, a fusion model based on prior knowledge can be used to predict HT risk. Through the combination of these two models, the embodiment of the present invention can provide a fast and easy clinical prediction tool (based on the Logistic regression model) and provide a more accurate prediction result in complex scenarios (based on the fusion model), which effectively improves the efficiency and accuracy of HT risk prediction for AIS patients.

[0084] In some embodiments, the obtained hemorrhagic transformation risk prediction results can also be displayed through a visual interface, for example, the HT risk score of the target object and the importance of key features can be displayed, and the key features may include but are not limited to TMAX volume, NIHSS score, HIR and other features, so that doctors can intuitively understand the HT risk situation of each patient, thereby assisting clinical decision-making.

[0085] In some embodiments, the hemorrhagic transformation risk prediction results of the target object can also be updated in real time. Once the MRI data and clinical baseline data of the target object's brain tissue are updated, the above-mentioned risk prediction process can be automatically executed to generate and display the latest prediction results, thereby helping doctors to quickly determine the patient's HT risk.

[0086] It can be understood that the embodiment of the present invention combines efficient machine learning algorithms with clinical operability design, and by integrating multiple machine learning algorithms (such as Logistic regression, XGBoost and LightGBM), it greatly improves the sensitivity and specificity of the hemorrhagic transformation risk prediction model, and solves the problem of unstable performance of the existing single model. This not only helps to improve the treatment prognosis of AIS patients, but also can be more widely promoted in clinical applications.

[0087] Reference Manual Attached Figure 5 , which shows the structure of a hemorrhagic transformation risk prediction device 500 provided by an embodiment of the present invention. Figure 5As shown, the apparatus 500 may include: A data acquisition module 510 is used to acquire magnetic resonance imaging data of the target subject's brain tissue and clinical baseline data of the target subject; A first feature determination module 520, for determining imaging features of the target object's brain tissue according to magnetic resonance imaging data; A second feature determination module 530, for determining clinical baseline features of the target subject based on the clinical baseline data; The risk prediction module 540 is used to determine the hemorrhagic transformation risk prediction result of the target object based on the imaging characteristics and clinical baseline characteristics.

[0088] In some embodiments, the risk prediction module 540 may include: A HIR acquisition unit, used for acquiring a low perfusion intensity ratio corresponding to the brain tissue of the target object; a risk level determination unit, configured to determine a hemorrhagic transformation risk level of a target object according to the low perfusion intensity ratio; The risk prediction unit is used to process the imaging features and clinical baseline features using a hemorrhagic transformation risk prediction model corresponding to the hemorrhagic transformation risk level to obtain a hemorrhagic transformation risk prediction result for the target object.

[0089] It should be noted that the device provided in the above embodiment, when implementing its functions, only uses the division of the above functional modules as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device provided in the above embodiment and the corresponding method embodiment belong to the same concept, and the specific implementation process is detailed in the corresponding method embodiment, which will not be repeated here.

[0090] An embodiment of the present invention also provides an electronic device, which includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the hemorrhagic transformation risk prediction method provided in the above method embodiment.

[0091] The memory can be used to store software programs and modules. The processor executes various functional applications and data processing by running the software programs and modules 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, application programs required for functions, etc.; the data storage area can store data created according to the use of the device, etc. In addition, the memory can include a high-speed random access memory and can also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory can also include a memory controller to provide the processor with access to the memory.

[0092] In conjunction with the reference manual Figure 6 , which is a block diagram of an electronic device 600 according to an embodiment of the present invention. The electronic device 600 may include one or more processors 602, a system control logic 608 connected to at least one of the processors 602, a system memory 604 connected to the system control logic 608, a non-volatile memory (NVM) 606 connected to the system control logic 608, and a network interface 610 connected to the system control logic 608.

[0093] The processor 602 may include one or more single-core or multi-core processors. The processor 602 may include any combination of general-purpose processors and special-purpose processors (e.g., graphics processors, application processors, baseband processors, etc.). In the embodiments of this document, the processor 602 may be configured to execute the following: Figures 2 to 4 One or more of the various embodiments shown.

[0094] In some embodiments, system control logic 608 may include any suitable interface controller to provide any suitable interface to at least one of processors 602 and / or any suitable device or component in communication with system control logic 608 .

[0095] In some embodiments, the system control logic 608 may include one or more memory controllers to provide an interface to the system memory 604. The system memory 604 may be used to load and store data and / or instructions. In some embodiments, the memory 604 of the device 600 may include any suitable volatile memory, such as a suitable dynamic random access memory (DRAM).

[0096] NVM / memory 606 may include one or more tangible, non-transitory computer-readable media for storing data and / or instructions. In some embodiments, NVM / memory 606 may include any suitable non-volatile memory such as flash memory and / or any suitable non-volatile storage device, such as at least one of a HDD (Hard Disk Drive), a CD (Compact Disc) drive, and a DVD (Digital Versatile Disc) drive.

[0097] NVM / storage 606 may include a portion of storage resources installed on a device of apparatus 600, or it may be accessible to the apparatus but not necessarily a part of the apparatus. For example, NVM / storage 606 may be accessed over a network via network interface 610.

[0098] In particular, system memory 604 and NVM / storage 606 may include, respectively, a temporary copy and a permanent copy of instructions 620. Instructions 620 may include instructions that, when executed by at least one of processors 602, cause device 600 to perform the following operations: Figures 2 to 4 Instructions of the hemorrhagic transformation risk prediction method shown. In some embodiments, the instructions 620, hardware, firmware and / or software components thereof may be additionally / alternatively placed in the system control logic 608, the network interface 610 and / or the processor 602.

[0099] The network interface 610 may include a transceiver for providing a radio interface for the device 600 to communicate with any other suitable device (such as a front-end module, an antenna, etc.) through one or more networks. In some embodiments, the network interface 610 may be integrated with other components of the device 600. For example, the network interface 610 may be integrated with at least one of a communication module of the processor 602, a system memory 604, an NVM / storage 606, and a firmware device (not shown) having instructions. When at least one of the processors 602 executes the instructions, the device 600 implements Figures 2 to 4 One or more of the various embodiments shown.

[0100] The network interface 610 may further include any suitable hardware and / or firmware to provide a multiple-input multiple-output radio interface. For example, the network interface 610 may be a network adapter, a wireless network adapter, a telephone modem and / or a wireless modem.

[0101] In one embodiment, at least one of the processors 602 may be packaged together with logic for one or more controllers of the system control logic 608 to form a system in package (SiP). In one embodiment, at least one of the processors 602 may be integrated on the same die with logic for one or more controllers of the system control logic 608 to form a system on chip (SoC).

[0102] The device 600 may further include an input / output (I / O) device 612. The I / O device 612 may include a user interface to enable a user to interact with the device 600; the design of the peripheral component interface enables the peripheral components to interact with the device 600. In some embodiments, the device 600 further includes a sensor for determining at least one of an environmental condition and location information related to the device 600.

[0103] In some embodiments, the user interface may include, but is not limited to, a display (e.g., an LCD display, a touch screen display, etc.), a speaker, a microphone, one or more cameras (e.g., a still image camera and / or a video camera), a flashlight (e.g., an LED flash), and a keyboard.

[0104] In some embodiments, the peripheral component interface may include, but is not limited to, a non-volatile memory port, an audio jack, and a power interface.

[0105] In some embodiments, the sensors may include, but are not limited to, gyroscope sensors, accelerometers, proximity sensors, ambient light sensors, and positioning units. The positioning unit may also be part of or interact with the network interface 610 to communicate with components of a positioning network (e.g., global positioning system (GPS) satellites).

[0106] It is to be understood that the structure illustrated in the embodiment of the present invention does not constitute a specific limitation on the electronic device 600. In other embodiments of the present invention, the electronic device 600 may include more or fewer components than shown in the figure, or combine some components, or separate some components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.

[0107] An embodiment of the present invention also provides a computer-readable storage medium, which can be set in an electronic device to store at least one instruction or at least one program related to implementing a hemorrhagic transformation risk prediction method. The at least one instruction or the at least one program is loaded and executed by the processor to implement the hemorrhagic transformation risk prediction method provided by the above method embodiment.

[0108] Optionally, in an embodiment of the present invention, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store program codes.

[0109] One embodiment of the present invention further provides a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. A processor of a computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the hemorrhagic transformation risk prediction method provided in the above various optional implementation examples.

[0110] It should be noted that the sequence of the embodiments of the present invention described above is for description only and does not represent the advantages and disadvantages of the embodiments. The above describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0111] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0112] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0113] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for predicting the risk of hemorrhagic transformation, characterized in that: include: Acquiring magnetic resonance imaging data of brain tissue of a target subject, and clinical baseline data of the target subject, wherein the clinical baseline data includes NIHSS score, attribute data, and disease history data; determining imaging features of the target object's brain tissue according to the magnetic resonance imaging data; Determining clinical baseline characteristics of the target subject according to the clinical baseline data; Determining the hemorrhagic transformation risk prediction result of the target subject based on the imaging features and the clinical baseline features includes: Acquiring a low perfusion intensity ratio corresponding to the brain tissue of the target object; determining a hemorrhagic transformation risk level of the target subject according to the low perfusion intensity ratio; The imaging features and the clinical baseline features are processed using a hemorrhagic transformation risk prediction model corresponding to the hemorrhagic transformation risk level to obtain a hemorrhagic transformation risk prediction result for the target object.

2. The method according to claim 1, characterized in that The magnetic resonance imaging data include perfusion weighted imaging data, and the clinical baseline data include NIHSS score; Determining the imaging features of the target object's brain tissue according to the magnetic resonance imaging data includes: Determining a residual function maximum time image of the target subject's brain tissue according to the perfusion weighted imaging data; The imaging characteristics of the brain tissue of the target object are determined according to the maximum residual function time image.

3. The method according to claim 2, characterized in that Determining the imaging characteristics of the target object's brain tissue according to the residual function maximum time image includes: The volume of the low perfusion area in the brain tissue of the target object and the low perfusion intensity ratio corresponding to the brain tissue of the target object are determined according to the maximum residual function time image.

4. The method according to claim 1, characterized in that The magnetic resonance imaging data includes perfusion weighted imaging data and diffusion weighted imaging data; Determining the imaging features of the target object's brain tissue according to the magnetic resonance imaging data includes: Determining a residual function maximum time image of the target subject's brain tissue according to the perfusion weighted imaging data; Determining an apparent diffusion coefficient image of the target object's brain tissue according to the diffusion weighted imaging data; The imaging characteristics of the brain tissue of the target object are determined based at least on the maximum residual function time image and the apparent diffusion coefficient image.

5. The method according to claim 4, characterized in that The step of determining the imaging characteristics of the brain tissue of the target object based at least on the residual function maximum time image and the apparent diffusion coefficient image comprises: determining the volume of the infarct area in the brain tissue of the target object according to the apparent diffusion coefficient image; Based at least on the residual function maximum time image, the volume of the target object's brain tissue having a TMAX value greater than each preset threshold value in at least one preset threshold value, the volume of the perfusion-diffusion mismatch area, and the low perfusion intensity ratio corresponding to the target object's brain tissue are determined respectively.

6. The method according to claim 4, characterized in that Determining the clinical baseline characteristics of the target subject according to the clinical baseline data includes: The attribute data and the disease history data are respectively encoded and / or standardized to obtain the attribute characteristics and disease history characteristics of the target object.

7. A device for predicting the risk of hemorrhagic transformation, characterized in that: include: A data acquisition module, used to acquire magnetic resonance imaging data of the target object's brain tissue and clinical baseline data of the target object, wherein the clinical baseline data includes NIHSS score, attribute data and disease history data; A first feature determination module, configured to determine imaging features of the target object's brain tissue according to the magnetic resonance imaging data; A second feature determination module, used to determine the clinical baseline features of the target subject according to the clinical baseline data; A risk prediction module, used to determine the hemorrhagic transformation risk prediction result of the target subject based on the imaging characteristics and the clinical baseline characteristics, including: Acquiring a low perfusion intensity ratio corresponding to the brain tissue of the target object; determining a hemorrhagic transformation risk level of the target subject according to the low perfusion intensity ratio; The imaging features and the clinical baseline features are processed using a hemorrhagic transformation risk prediction model corresponding to the hemorrhagic transformation risk level to obtain a hemorrhagic transformation risk prediction result for the target object.

8. An electronic device, characterized in that: The electronic device includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the hemorrhagic transformation risk prediction method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement the hemorrhagic transformation risk prediction method as described in any one of claims 1-6.

Images