Patient cognitive disorder recognition method and system and electronic equipment

Abstract

The invention relates to the technical field of medical treatment, and discloses a patient cognitive impairment recognition method and system and electronic device.The method comprises the steps that patient original data are obtained, and the patient original data comprise clinical biochemical data, image data and cognitive behavior data; preprocessing the original data of the patient to obtain preprocessed data; feature extraction is conducted on the preprocessed data, key features are obtained, and the key features comprise clinical biochemical features, image features and cognitive behavior features; fusing the clinical biochemical features, the image features and the cognitive behavior features to obtain fused features; and determining a cognitive impairment recognition result based on the fusion features and the cognitive impairment recognition function. According to the embodiment of the invention, the early detection rate of cognitive impairment of the pituitary adenoma patient is improved, missed diagnosis and misdiagnosis are reduced, and the early warning cognitive risk when obvious clinical symptoms do not appear in cognitive impairment recognition of the pituitary adenoma patient is reduced.

Description

Technical Field

[0001] The present disclosure relates to the field of medical technology, and in particular to a method, system, and electronic device for identifying cognitive impairment in patients. Background Art

[0002] Pituitary adenomas are common intracranial tumors, accounting for approximately 15%-20% of all intracranial tumors. Although most pituitary adenomas are benign, due to their unique anatomical location and endocrine function, patients often experience a variety of complex clinical symptoms, including abnormal hormone secretion, visual impairment, and cognitive dysfunction. In recent years, with the deepening of neuroscience and clinical medicine research, the issue of cognitive dysfunction in patients with pituitary adenomas has gradually attracted attention. Studies have shown that patients with pituitary adenomas generally have cognitive problems such as memory loss, decreased attention, and executive dysfunction. These problems not only seriously affect the quality of life of patients, but may also interfere with treatment compliance and rehabilitation outcomes. However, the assessment of cognitive function in patients with pituitary adenomas in clinical practice still faces many challenges, including the subjectivity of assessment methods, the lag in detection timing, and the difficulty of integrating multidimensional data.

[0003] In the traditional field of cognitive impairment assessment, neuropsychological scales (such as the MMSE and MoCA) have been the mainstay of clinical diagnosis. While these scales offer the advantages of ease of use and low cost, they have significant limitations: First, scale test results are susceptible to interference from factors such as the patient's education level, the testing environment, and the experience of the tester; second, traditional scales lack sensitivity for early mild cognitive impairment, often detecting abnormalities only after significant functional decline; and third, scale tests typically only provide static assessment results, failing to capture the dynamic characteristics of cognitive function as the disease progresses. For the unique population of patients with pituitary adenomas, traditional scales also lack targeted assessment modules for specific cognitive domains (such as executive function and information processing speed), resulting in a high rate of missed diagnosis.

[0004] In view of this, the present invention is proposed. Summary of the Invention

[0005] In order to solve the above technical problems or at least partially solve the above technical problems, the embodiments of the present disclosure provide a method, system and electronic device for identifying patient cognitive impairment, which can improve the early detection rate of cognitive impairment in patients with pituitary adenomas, reduce missed diagnoses and misdiagnoses, and reduce the early warning cognitive risk of cognitive impairment in patients with pituitary adenomas when obvious clinical symptoms have not yet appeared.

[0006] In a first aspect, an embodiment of the present disclosure provides a method for identifying cognitive impairment in a patient, the method comprising:

[0007] Obtaining original patient data, including clinical biochemical data, imaging data, and cognitive behavioral data;

[0008] Preprocessing the patient's original data to obtain preprocessed data, wherein the preprocessed data includes preprocessed clinical biochemical data, preprocessed imaging data, and preprocessed cognitive behavioral data;

[0009] Performing feature extraction on the preprocessed data to obtain key features, wherein the key features include clinical biochemical features, imaging features, and cognitive behavioral features;

[0010] fusing the clinical-biochemical features, imaging features, and cognitive-behavioral features to obtain a fusion feature;

[0011] A cognitive impairment recognition result is determined based on the fusion feature and the cognitive impairment recognition function.

[0012] In a second aspect, an embodiment of the present disclosure further provides a system for identifying cognitive impairment in patients, the system comprising:

[0013] A data acquisition module is used to acquire original patient data, including clinical biochemical data, imaging data, and cognitive behavioral data;

[0014] A preprocessing module, configured to preprocess the original patient data to obtain preprocessed data, wherein the preprocessed data includes preprocessed clinical biochemical data, preprocessed imaging data, and preprocessed cognitive behavioral data;

[0015] A feature acquisition module is used to extract features from the preprocessed data to obtain key features, wherein the key features include clinical biochemical features, imaging features, and cognitive behavioral features;

[0016] A fusion module, used to fuse the clinical biochemical features, imaging features and cognitive behavioral features to obtain a fusion feature;

[0017] The result acquisition module is used to determine the cognitive impairment recognition result based on the fusion feature and the cognitive impairment recognition function.

[0018] In a third aspect, an embodiment of the present disclosure further provides an electronic device comprising: one or more processing devices; a storage device for storing one or more programs; the one or more programs are executed by the one or more processing devices to implement the method for identifying cognitive impairment in patients as described above.

[0019] Embodiments of the present disclosure provide a method, system, and electronic device for identifying patient cognitive impairment, the method comprising: obtaining raw patient data; preprocessing the raw patient data to obtain preprocessed data; extracting features from the preprocessed data to obtain key features, the key features including clinical biochemical features, imaging features, and cognitive behavioral features; fusing the clinical biochemical features, imaging features, and cognitive behavioral features to obtain fused features; and determining a cognitive impairment identification result based on the fused features and a cognitive impairment identification function. The present disclosure screens and processes missing values ​​from patient data to eliminate noise and inconsistencies, thereby improving the efficiency and accuracy of data processing. Furthermore, by extracting features from preprocessed clinical biochemical data, preprocessed imaging data, and preprocessed cognitive behavioral data, important indicators related to cognitive impairment are identified, cognitive behavioral abnormalities are quantified, and the limitations of traditional subjective scale assessments based on physician experience are overcome, thereby improving the accuracy of subsequent analysis. Then, the clinical biochemical features, imaging features, and cognitive behavioral features from different sources are effectively fused to integrate the advantages of each modality and generate a more comprehensive feature description. The fusion of features from different sources not only improves the interpretability of the data, but also enhances the model's adaptability to various features, making the comprehensive analysis more accurate, able to capture potential signals of cognitive impairment, and improve the sensitivity and accuracy of cognitive impairment signals; finally, the cognitive impairment recognition function significantly improves the early detection rate of cognitive impairment in patients with pituitary adenomas, reduces missed diagnoses and misdiagnoses, and reduces the warning cognitive risk of cognitive impairment in patients with pituitary adenomas before obvious clinical symptoms appear. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and other features, advantages, and aspects of the various embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that the originals and elements are not necessarily drawn to scale.

[0021] Figure 1 Flowchart of a method for identifying cognitive impairment in a patient according to an embodiment of the present disclosure.

[0022] Figure 2 Schematic diagram of the structure of a patient cognitive impairment identification system in an embodiment of the present disclosure.

[0023] Figure 3 Schematic diagram of the structure of an electronic device in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0024] The following describes embodiments of the present disclosure in more detail with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of protection of the present disclosure.

[0025] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0026] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only for illustrative purposes and are not intended to limit the scope of these messages or information.

[0027] With the advancement of medical imaging technology, structural and functional neuroimaging (such as MRI, fMRI, and DTI) has provided new insights into the pathological mechanisms of cognitive impairment. In the field of pituitary adenomas, studies have found a correlation between tumor volume and invasiveness and the degree of cognitive impairment, and that cognitive improvement in some patients after surgical treatment is associated with the restoration of white matter fiber tract integrity. However, existing imaging analyses primarily rely on physician judgment and simple quantitative measurements, lacking the ability to deeply mine massive amounts of imaging data. In particular, traditional analytical methods often struggle to identify the intrinsic links between subtle structural changes or complex functional connectivity changes and cognitive impairment. Furthermore, single imaging markers often have limited explanatory power and require combined analysis with other clinical indicators to improve diagnostic efficacy.

[0028] The development of medical informatization has led to the widespread use of electronic medical record systems, making it easier to access structured data such as clinical biochemical data and treatment records. Abnormal levels of endocrine hormones (such as growth hormone and ACTH) in patients with pituitary adenomas have been shown to be associated with cognitive impairment, but a precise mathematical model has yet to be established to quantify the relationship between these indicators and cognitive impairment. Currently, clinical decision-making is often based on simple thresholds, failing to fully utilize dynamic monitoring data from multiple time points. Furthermore, unstructured text data in electronic medical record systems (such as medical records and imaging reports) contains a wealth of valuable information, but due to limitations in natural language processing technology, this information has yet to be effectively translated into analyzable predictive features for cognitive impairment. In recent years, digital cognitive assessment tools have begun to be used in clinical research. These cognitive tests, administered via computer or tablet, can more accurately record parameters such as reaction time and accuracy. Some systems also capture behavioral indicators such as mouse tracking and touchscreen manipulation. These methods have demonstrated advantages in early screening for neurodegenerative diseases such as Alzheimer's disease and Parkinson's disease, but their application in the field of pituitary adenomas is still in its infancy.

[0029] Existing digital assessment tools mostly use fixed testing paradigms and lack targeted designs for the specific cognitive characteristics of pituitary adenomas. At the same time, the collected behavioral data are often only used for simple statistical analysis and fail to form an effective multi-dimensional fusion analysis with other modal medical data. Breakthroughs in artificial intelligence technology provide new solutions to the above problems. In medical image analysis, deep learning algorithms have demonstrated performance that surpasses traditional methods in tasks such as tumor segmentation and anomaly detection. In the construction of clinical prediction models, machine learning technology can process high-dimensional features and discover complex nonlinear relationships. In the field of behavioral data analysis, time series modeling algorithms can extract subtle changes in behavioral patterns. However, existing AI medical applications mostly focus on a single data type or specific clinical problems, and cross-modal intelligent analysis systems for cognitive impairment in pituitary adenomas have not yet been reported.

[0030] To address the above issues, the embodiments of the present disclosure provide a method for identifying cognitive impairment in patients, which can improve the early detection rate of cognitive impairment in patients with pituitary adenomas, reduce missed diagnoses and misdiagnoses, and reduce the early warning cognitive risk of cognitive impairment in patients with pituitary adenomas before obvious clinical symptoms appear.

[0031] Figure 1 This is a flow chart of a method for identifying a patient's cognitive impairment in an embodiment of the present disclosure. The method can be executed by a patient's cognitive impairment identification system, which can be implemented in software and / or hardware, and can be configured in an electronic device. Figure 1 As shown, the method may specifically include the following steps:

[0032] S110: Acquire original patient data, including clinical biochemical data, imaging data, and cognitive behavioral data.

[0033] Specifically, clinical biochemical data can include hormone levels, which can be measured by regularly collecting blood samples and recording hormone levels, eye tracking, reaction time, and other data (e.g., growth hormone, prolactin, adrenocorticotropic hormone, etc.) using electronic medical records (EMRs), as well as cognitive behavioral data (e.g., eye tracking, reaction time, and performance on virtual reality tasks). Imaging data can include brain imaging data, such as MRI data or CT scan data. In this article, raw data types can include text data, word data, image data, or any other processable data format (e.g., numbers, strings, lists, etc.). Raw data can come from hospitals, medical institutions, medical research institutions, or other legally and compliant sources.

[0034] S120: Preprocessing the original patient data to obtain preprocessed data, wherein the preprocessed data includes preprocessed clinical biochemical data, preprocessed imaging data, and preprocessed cognitive behavioral data.

[0035] In a specific embodiment, preprocessing the original patient data to obtain preprocessed data includes:

[0036] Performing repeated data screening on the original data of the patient to obtain screening data;

[0037] Missing value processing is performed on the screened data to obtain preprocessed data.

[0038] Specifically, clinical biochemical data, imaging data, and cognitive behavioral data were first examined. Duplicate raw data were then removed to obtain screened data to avoid double counting during analysis. Missing value processing was performed on the screened data to obtain preprocessed data. Missing value processing methods for screened data include: filling missing data in the screened data with the mean or median of the screened data, deleting missing data in the screened data, or using interpolation methods to interpolate missing data in the screened data to obtain preprocessed data.

[0039] This embodiment screens patient data and processes missing values ​​to eliminate noise and inconsistency, thereby improving the efficiency and accuracy of data processing.

[0040] In one embodiment, the image data includes first image data and second image data, and after preprocessing the original patient data, the following steps are further included:

[0041] The pixel values ​​of the first image data and the pixel values ​​of the second image data are weightedly fused to obtain preprocessed image data.

[0042] Based on the above embodiment, the image data includes first image data and second image data, wherein the first image data may be MRI data and the second image data may be CT data. After preprocessing the first image data and the second image data, the preprocessed first image data and the second image data are obtained, and then the pixel values ​​of the preprocessed first image data and the pixel values ​​of the second image data are weightedly fused to obtain the preprocessed image data.

[0043] For example, the first image data may be MRI image data, and the second image data may be CT image data. The grayscale value of the pixel at the CT image The pixel grayscale values ​​at are weighted averaged to obtain the preprocessed image data. For example, the pixel values ​​of an MRI image are:

[0044] [100, 150];

[0045] [120, 130];

[0046] Pixel values ​​of CT images:

[0047] [80, 200];

[0048] [90, 140];

[0049] The pixel values ​​of the fused image are as follows:

[0050] [94, 165];

[0051] [111, 133].

[0052] The detailed process is as follows: To fuse the MRI and CT images based on the importance of each modality, weights need to be assigned. Assuming that MRI is better at representing cellular structures, the MRI weight (Wmri) is assigned to 0.7 and the CT weight (Wct) is assigned to 0.3.

[0053] For each pixel position, the fusion value is calculated as follows: fusion value = Wmri × MRI pixel value + Wct × CT pixel value. For example, pixel (0, 0): fusion value = 0.7 × 100 + 0.3 × 80 = 94.

[0054] This embodiment combines the advantages of different modalities and improves data sensitivity by weighted averaging the pixel values ​​of the MRI image and the pixel values ​​of the CT image.

[0055] S130: Extracting features from the pre-processed data to obtain key features, wherein the key features include clinical biochemical features, imaging features, and cognitive features.

[0056] Specifically, based on the above embodiment, key features can be extracted from pre-processed clinical biochemical data, pre-processed imaging data and pre-processed cognitive behavioral data, namely clinical biochemical features, imaging features and cognitive behavioral features, among which clinical biochemical index features Pay attention to the temporal changes in hormone levels. Imaging characteristics Focus on analyzing grayscale changes in images and extracting texture information. Cognitive behavioral characteristics Focus on the temporal sequence of responses and assess behavioral performance.

[0057] In a specific embodiment, a first formula is used to extract features from pre-processed clinical biochemical data to obtain clinical biochemical features, wherein the expression of the first formula is: ,

[0058] in, Indicates clinical and biochemical characteristics, Indicates the starting time point of preprocessing clinical biochemical data, that is, the starting time point of preprocessing data, Indicates the end time point of preprocessing clinical biochemical data, that is, the end time point of preprocessing data, Indicates at a point in time The hormone level at which represents baseline hormone levels, Indicates hormone levels Time point The second derivative of represents the power law adjustment coefficient, Indicates the start time and end time point The first discrete time points, Indicates multiplication.

[0059] The first formula enables nonlinear enhancement and multi-scale analysis of the dynamic changes in hormone levels. This formula, through a combination of integral and second-order derivative terms, effectively captures the transient characteristics and metabolic decay patterns of hormone secretion. Specifically, the integral term, by combining the hormone rate of change and logarithmic ratio operations, significantly amplifies the transient fluctuation characteristics of pulsed hormone secretion, providing a sensitive indicator for the early diagnosis of endocrine diseases. Simultaneously, the second-order derivative term introduces a power-law decay coefficient, which can quantitatively assess the accelerated changes in hormone metabolism at different time scales and effectively detect decreased hormone clearance due to liver and kidney dysfunction. The innovation of this formula lies in integrating the time-varying, nonlinear, and metabolic dynamics of hormones into a unified framework. This not only improves the ability to characterize dynamic changes in hormones, but also provides a quantitative basis for the clinical assessment of endocrine function and metabolic abnormalities. It has important application value in the auxiliary diagnosis of diseases such as pituitary dysfunction and metabolic syndrome.

[0060] In a specific embodiment, a second formula is used to extract features from the pre-processed image data to obtain image features, wherein the expression of the second formula is: ,

[0061] in, Represents image features, represents the integrated area of ​​the brain scan in the preprocessed image data, Represents the coordinates in the preprocessed image data The grayscale value of the pixel at represents the standard deviation of the preprocessed image data, represents the mean of the preprocessed image data, represents the divergence operator, Represents the preprocessed image data Axial gradient phase, Indicates multiplication.

[0062] The second formula achieves accurate extraction of multi-scale features of brain images by fusing grayscale statistical features and spatial gradient information, which can effectively enhance the contrast between lesion areas and normal tissues, especially the sensitivity of detecting small tumors. , ) is combined with the local spatial variation (∇ operator) to effectively enhance the contrast between brain lesion areas (such as tumors or infarcts) and normal tissues. Axial gradient phase analysis is particularly well-suited for extracting inter-layer structural features from three-dimensional images such as brain MRI. This processing method can accurately identify spatial variations in anatomical structures such as cortical folds. By organically combining statistical features with spatial differential features through integral operations, it overcomes the limitations of traditional methods that process grayscale or texture features separately.

[0063] In a specific embodiment, a third formula is used to extract features from the pre-processed cognitive behavior data to obtain cognitive behavior features, wherein the expression of the third formula is: ,

[0064] in, Represents cognitive behavioral characteristics, Represents the preprocessed cognitive behavior data at time point The reaction time series at Represents the preprocessed cognitive behavior data at time point The reaction time series at The periodic frequency, Represents the pre-processed cognitive behavior data at time point in a time period The reaction time series at The average value of the time period is the starting time point and end time point Any time period between represents the time decay window factor, Represents the preprocessed cognitive behavioral data at time point The reaction time series at The speed change influence factor, represents the penalty coefficient factor, Indicates multiplication.

[0065] The third formula achieves accurate quantification of the neurocognitive state by integrating the dynamic fluctuation characteristics of reaction time and behavioral stability analysis. Through joint analysis in the time and frequency domains, the second-order derivative-periodic coupling term in the formula ( ) can effectively capture periodic anomalies with microsecond-level response delays, while the dynamic penalty term, by quantifying behavioral response deviations, can significantly improve the detection rate of executive function decline in patients. This highly accurate, quantifiable assessment method provides a technical means for intelligent diagnosis of cognitive impairment, early screening, and therapeutic efficacy monitoring.

[0066] This embodiment extracts features from pre-processed clinical biochemical data, pre-processed imaging data, and pre-processed cognitive behavioral data to identify important indicators related to cognitive impairment, quantify cognitive behavioral abnormalities, overcome the limitations of traditional subjective scale assessments based on physician experience, and thus improve the accuracy of subsequent analysis.

[0067] S140: Fusing the clinical biochemical features, imaging features, and cognitive behavioral features to obtain fusion features.

[0068] For example, the clinical biochemical features, imaging features, and cognitive behavioral features from different sources can be effectively fused through dynamic weights and variance weighting to form a comprehensive feature matrix, where the expression of the fused features is: ,

[0069] in, represents the fusion feature, represents the hyperbolic characteristic function, Indicates the The temporal dynamic weight of each mode, Indicates the The characteristic variance of the modes, Indicates the The features or feature matrix of each mode, Represents the element-by-element multiplication symbol, represents the feature weight matrix, Indicates the time point Find the gradient, Represents the feature weight matrix at time point The time gradient at Indicates multiplication.

[0070] The uniqueness of the feature fusion method of the present invention lies in the combination of dynamic weight adjustment and variance weighting mechanism. It can not only give full play to the respective advantages of different modalities (clinical biochemistry, imaging, cognitive behavior) and extract the complementary features of multi-source information, but also automatically adjust the importance of each modality during the fusion process, thereby reducing the impact of noise and error. In addition, the use of hyperbolic tangent activation function and time gradient realizes nonlinear mapping and temporal dynamic management of features, enhancing the model's ability to capture potential complex relationships in clinical data. Overall, this fusion method has better anti-interference and self-adaptation capabilities, which helps to extract more robust and diagnostically valuable comprehensive features, thereby significantly improving the recognition accuracy of cognitive impairment-related indicators, which is superior to traditional static or single modality fusion methods.

[0071] By effectively fusing clinical biochemical, imaging, and cognitive behavioral features from different sources, the disclosed embodiments can integrate the strengths of each modality and generate a more comprehensive feature description. Fusion of features from different sources not only improves the interpretability of the data but also enhances the model's adaptability to various features, making the comprehensive analysis more accurate, capable of capturing potential signals of cognitive impairment, and improving the sensitivity and accuracy of cognitive impairment signals.

[0072] It should be noted that the features of each modality (F1, F2, F3) will be a matrix or vector, where the rows represent different samples or observations and the columns represent the extracted features. Therefore, when we refer to the features of the "g-th modality", we are actually referring to the feature representation of all samples under the corresponding modality. Modality here refers to information from different data sources or types. For example, biochemical features come from biochemical measurements (such as hormone levels), imaging features come from imaging technology (such as grayscale values ​​of MRI or CT), and behavioral features are behavioral response data (such as reaction time series). In multimodal learning, information from each source is usually represented in a different way, so we call them different "modalities".

[0073] :With The variance of features within each modality. Variance measures the degree to which a feature changes during the collection process, reflecting the amount of information in the dataset. By incorporating variance into the fusion process, the influence of each feature on the final fusion result can be adjusted. Generally speaking, modalities with larger feature variances contribute more to the fusion result because they provide more information.

[0074] : A time-dynamic weight associated with each modality. It takes into account the importance and availability of each modality at different points in time. The importance of data from different sources may change over time, so the composite weight needs to be adjusted dynamically.

[0075] : Represents the time gradient of the feature weight matrix. This part is used to describe how the weights change over time. By applying the time gradient, the flexibility of the model can be further enhanced, making it able to adapt to the impact of data changes.

[0076] S150: Determine a cognitive impairment recognition result based on the fusion feature and the cognitive impairment recognition function.

[0077] Specifically, the cognitive impairment recognition result may be a cognitive impairment recognition coefficient or a cognitive impairment recognition score, for example, a preset threshold value that is set in advance. When the cognitive impairment recognition score is less than the preset threshold value, it is determined to be a low-level cognitive impairment. When the cognitive impairment recognition score is greater than the preset threshold value, it is determined to be a high-level cognitive impairment. The specific preset threshold values ​​may include a first threshold value and a second threshold value. When the cognitive impairment recognition score is less than the first threshold value, it is determined to be a low-level cognitive impairment; when the cognitive impairment recognition score is greater than the first threshold value and less than the second threshold value, it is determined to be a medium-level cognitive impairment; when the cognitive impairment recognition score is greater than the second threshold value, it is determined to be a high-level cognitive impairment. For example, the first threshold value may be 50, and the second threshold value may be 80. >80, there is significant cognitive impairment; when 50< When <80, cognitive impairment requires further observation; When <50, cognitive ability is normal, that is, there is no cognitive impairment.

[0078] In a specific embodiment, the expression of the cognitive impairment identification function is: ,

[0079] in, represents the cognitive impairment score, Indicates the starting time point of preprocessing data, Indicates the end time point of preprocessing data, Indicates the Fusion features At the time point The value at represents the power law adjustment coefficient, Indicates the cognitive behavioral characteristics At the time point The measured value at Indicates the cognitive behavioral characteristics At the start time The baseline value at Indicates a time point changes, Indicates the cognitive behavioral characteristics The weight tilt angle, represents the inverse tangent function, represents the fusion feature, represents the integrated area of ​​the brain scan in the preprocessed image data, represents the gradient field of the preprocessed image features, represents multiplication, represents the norm, Represents cognitive behavioral characteristics The number of .

[0080] The present invention combines the triple features of imaging, behavioral, and biochemical data to comprehensively reflect an individual's cognitive state. Its complex integral and product structures enable deep interaction between these diverse information across time and space. The introduction of a power-law adjustment coefficient enhances the model's adaptability to individual differences and sensitivity changes, enabling it to capture subtle cognitive changes and highlight early abnormalities. By continuously integrating the integral interval and introducing the image gradient field, it takes into account both the temporal evolution and structural spatial characteristics of cognitive indicators, enhancing the sensitivity of identifying complex dynamic changes. While extracting cognitive scores, this function also considers the spatial structure of the image. Through gradient field integration and behavioral dynamics, time-derivative and exponential adjustment, it achieves multi-angle cross-validation of indicators, reducing the risk of misdiagnosis associated with a single indicator. This cognitive impairment identification function effectively integrates key features of imaging, behavioral, and biochemical data through multi-source, multi-scale, and nonlinear dynamic modeling. This not only enhances sensitivity to subtle cognitive changes but also suppresses noise interference, enabling accurate capture and dynamic tracking of early abnormalities, significantly improving the diagnostic accuracy and reliability of early screening for cognitive impairment.

[0081] For example, the cognitive impairment recognition function is constructed using fusion features. In the process of solving the cognitive impairment score, the value of the fusion feature needs to be input into the cognitive impairment recognition function to obtain the cognitive impairment score, which is used to evaluate whether the cognitive impairment is normal. The assessment plan is made by combining the cognitive impairment score with the patient's medical history and symptoms. First, the medical history assessment is performed:

[0082] 1) Detailed medical history collection: Medical staff should conduct a detailed investigation of the patient's medical history, including family history, past medical history, lifestyle, timeline of cognitive function changes, etc.

[0083] 2) Identify comorbidities: Look for accompanying mental health disorders (e.g., anxiety, depression), physical illnesses (e.g., diabetes, hypertension), and other factors (e.g., medication use and environmental influences) that may affect cognitive states.

[0084] Then, based on the cognitive impairment score and the patient's specific situation, an appropriate intervention and treatment plan can be developed, including:

[0085] 1) Cognitive training: Recommend cognitive training programs to patients to stimulate the brain through appropriate activities and delay cognitive decline.

[0086] 2) Drug treatment: If the evaluation indicates the possibility of Alzheimer's disease or another type of dementia, the doctor may consider prescribing relevant medications (such as donepezil).

[0087] The disclosed embodiments significantly improve the early detection rate of cognitive impairment in pituitary adenoma patients through the cognitive impairment recognition function, reduce missed diagnoses and misdiagnoses, and reduce the early warning cognitive risk of cognitive impairment in pituitary adenoma patients before obvious clinical symptoms appear.

[0088] In summary, the embodiments of the present disclosure provide a method for identifying cognitive impairment in patients, which includes: obtaining original patient data; preprocessing the original patient data to obtain preprocessed data; extracting features from the preprocessed data to obtain key features, wherein the key features include clinical biochemical features, imaging features, and cognitive behavioral features; fusing the clinical biochemical features, imaging features, and cognitive behavioral features to obtain fused features; and determining cognitive impairment identification results based on the fused features and the cognitive impairment identification function. The present disclosure eliminates noise and inconsistencies by screening and processing missing values ​​of patient data, thereby improving the efficiency and accuracy of data processing; further, by extracting features from preprocessed clinical biochemical data, preprocessed imaging data, and preprocessed cognitive behavioral data, important indicators related to cognitive impairment are identified, cognitive behavioral abnormalities are quantified, and the limitations of traditional subjective scale assessments based on doctor experience are overcome, thereby improving the accuracy of subsequent analysis; then, the clinical biochemical features, imaging features, and cognitive behavioral features from different sources are effectively fused to integrate the advantages of each modality and generate a more comprehensive feature description. The fusion of features from different sources not only improves the interpretability of the data, but also enhances the model's adaptability to various features, making the comprehensive analysis more accurate, able to capture potential signals of cognitive impairment, and improve the sensitivity and accuracy of cognitive impairment signals; finally, the cognitive impairment recognition function significantly improves the early detection rate of cognitive impairment in patients with pituitary adenomas, reduces missed diagnoses and misdiagnoses, and reduces the warning cognitive risk of cognitive impairment in patients with pituitary adenomas before obvious clinical symptoms appear.

[0089] Figure 2 FIG. 1 is a schematic diagram of the structure of the patient cognitive impairment identification system in the embodiment of the present disclosure. Figure 2 As shown, the system includes a data acquisition module S210, a preprocessing module S220, a feature acquisition module S230, a fusion module S240 and a result acquisition module S250.

[0090] The data acquisition module S210 is used to acquire original patient data, which includes clinical biochemical data, imaging data, and cognitive behavioral data.

[0091] The preprocessing module S220 is used to preprocess the original patient data to obtain preprocessed data, which includes preprocessed clinical biochemical data, preprocessed imaging data, and preprocessed cognitive behavioral data.

[0092] The feature acquisition module S230 is used to extract features from the pre-processed data to obtain key features, which include clinical biochemical features, imaging features, and cognitive behavioral features.

[0093] The fusion module S240 is used to fuse the clinical biochemical features, imaging features and cognitive behavioral features to obtain fusion features.

[0094] The result acquisition module S250 is used to determine a cognitive impairment recognition result based on the fusion feature and the cognitive impairment recognition function.

[0095] In a specific embodiment, the pre-processing module S220 is further configured to perform duplicate data screening on the original patient data to obtain screened data; and perform missing value processing on the screened data to obtain pre-processed data.

[0096] In a specific embodiment, the pre-processing module S220 is further configured to perform weighted fusion on the pixel values ​​of the first image data and the second image data to obtain pre-processed image data, where the image data includes the first image data and the second image data.

[0097] In a specific embodiment, the feature acquisition module S230 is used to extract features from the pre-processed clinical biochemical data using a first formula to obtain clinical biochemical features, wherein the expression of the first formula is: ,

[0098] in, Indicates clinical and biochemical characteristics, Indicates the starting time point of preprocessing clinical biochemical data, that is, the starting time point of preprocessing data, Indicates the end time point of preprocessing clinical biochemical data, that is, the end time point of preprocessing data, Indicates at a point in time The hormone level at which represents baseline hormone levels, Indicates hormone levels Time point The second derivative of represents the power law adjustment coefficient, Indicates the start time and end time point The first discrete time points, Indicates multiplication.

[0099] In a specific embodiment, the feature acquisition module S230 is used to extract features from the pre-processed image data using a second formula to obtain image features, wherein the expression of the second formula is: ,

[0100] in, Represents image features, represents the integrated area of ​​the brain scan in the preprocessed image data, Represents the coordinates in the preprocessed image data The grayscale value of the pixel at represents the standard deviation of the preprocessed image data, represents the mean of the preprocessed image data, represents the divergence operator, Represents the preprocessed image data Axial gradient phase, Indicates multiplication.

[0101] In a specific embodiment, the feature acquisition module S230 is configured to extract features from the pre-processed cognitive behavior data using a third formula to obtain cognitive behavior features, wherein the expression of the third formula is: ,

[0102] in, Represents cognitive behavioral characteristics, Represents the preprocessed cognitive behavioral data at time point The reaction time series at Represents the preprocessed cognitive behavioral data at time point The reaction time series at The periodic frequency, Represents the pre-processed cognitive behavior data at time point in a time period The reaction time series at The average value of the time period is the starting time point and end time point Any time period between represents the time decay window factor, Represents the preprocessed cognitive behavioral data at time point The reaction time series at The speed change influence factor, represents the penalty coefficient factor, Indicates multiplication.

[0103] In a specific embodiment, the expression of the fusion feature is: ,

[0104] in, represents the fusion feature, represents the hyperbolic characteristic function, Indicates the The temporal dynamic weight of each mode, Indicates the The characteristic variance of the modes, Indicates the The features or feature matrix of each mode, Represents the element-by-element multiplication symbol, represents the feature weight matrix, Indicates the time point Find the gradient, Represents the feature weight matrix at time point The time gradient at Indicates multiplication.

[0105] In a specific embodiment, the expression of the cognitive impairment identification function is: ,

[0106] in, represents the cognitive impairment score, Indicates the starting time point of preprocessing data, Indicates the end time point of preprocessing data, Indicates the Fusion features At the time point The value at represents the power law adjustment coefficient, Indicates the cognitive behavioral characteristics At the time point The measured value at Indicates the cognitive behavioral characteristics At the start time The baseline value at Indicates a time point changes, Indicates the cognitive behavioral characteristics The weight tilt angle, represents the inverse tangent function, represents the fusion feature, represents the integrated area of ​​the brain scan in the preprocessed image data, represents the gradient field of the preprocessed image features, represents multiplication, represents the norm, Represents cognitive behavioral characteristics The number of .

[0107] The patient cognitive impairment identification system provided by the embodiment of the present disclosure can execute the steps of the patient cognitive impairment identification method provided by the method embodiment of the present disclosure. The execution steps and beneficial effects are not repeated here.

[0108] Figure 3This is a schematic diagram of the structure of an electronic device in an embodiment of the present disclosure. Figure 3 , which shows a structural diagram of an electronic device 500 suitable for implementing the embodiments of the present disclosure. Figure 3 The electronic device shown is only an example and should not limit the functions and scope of use of the embodiments of the present disclosure.

[0109] like Figure 3 As shown, electronic device 500 may include a processing device 501, ROM 502, RAM 503, a bus 504, an input / output (I / O) interface 505, an input device 506, an output device 507, a storage device 508, and a communication device 509. The processing device (e.g., a central processing unit, a graphics processing unit, etc.) 501 can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 502 or a program loaded from storage device 508 into random access memory (RAM) 503 to implement the method for identifying cognitive impairment in patients according to the embodiments of the present disclosure. RAM 503 also stores various programs and data required for the operation of electronic device 500. The processing device 501, ROM 502, and RAM 503 are interconnected via bus 504. An input / output (I / O) interface 505 is also connected to bus 504.

[0110] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart, thereby implementing the patient cognitive impairment identification method as described above. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 509, or installed from the storage device 508, or installed from the ROM 502. When the computer program is executed by the processing device 501, the above-mentioned functions defined in the method of the embodiment of the present disclosure are performed.

[0111] It should be noted that the computer-readable medium described above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. Computer-readable storage media may include, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or components, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such a propagated data signal may take a variety of forms, including, but not limited to, electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wire, optical cable, RF (radio frequency), or any suitable combination thereof.

[0112] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device. The computer-readable medium carries one or more programs. When executed by the electronic device, the electronic device: obtains raw patient data, including clinical biochemical data, imaging data, and cognitive behavioral data; preprocesses the raw patient data to obtain preprocessed data, including preprocessed clinical biochemical data, preprocessed imaging data, and preprocessed cognitive behavioral data; extracts features from the preprocessed data to obtain key features, including clinical biochemical features, imaging features, and cognitive behavioral features; fuses the clinical biochemical features, imaging features, and cognitive behavioral features to obtain fused features; and determines a cognitive impairment recognition result based on the fused features and a cognitive impairment recognition function.

[0113] Optionally, when the above one or more programs are executed by the electronic device, the electronic device may also execute other steps described in the above embodiments.

[0114] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fibers, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0115] The above description is merely a preferred embodiment of the present disclosure and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of the present disclosure is not limited to technical solutions formed by specific combinations of the aforementioned technical features. It also encompasses other technical solutions formed by any combination of the aforementioned technical features or their equivalents, without departing from the scope of the above disclosure. For example, a technical solution formed by replacing the aforementioned features with (but not limited to) technical features with similar functions disclosed in this disclosure.

Claims

1. A method for identifying cognitive impairment in a patient, characterized in that: The method for identifying cognitive impairment in patients comprises: Obtaining original patient data, including clinical biochemical data, imaging data, and cognitive behavioral data; Preprocessing the patient's original data to obtain preprocessed data, wherein the preprocessed data includes preprocessed clinical biochemical data, preprocessed imaging data, and preprocessed cognitive behavioral data; Performing feature extraction on the preprocessed data to obtain key features, wherein the key features include clinical biochemical features, imaging features, and cognitive behavioral features; fusing the clinical-biochemical features, imaging features, and cognitive-behavioral features to obtain a fusion feature; A cognitive impairment recognition result is determined based on the fusion feature and the cognitive impairment recognition function.

2. The method for identifying cognitive impairment in patients according to claim 1, characterized in that: Preprocessing the original patient data to obtain preprocessed data includes: Performing repeated data screening on the original data of the patient to obtain screening data; Missing value processing is performed on the screened data to obtain preprocessed data.

3. The method for identifying cognitive impairment in patients according to claim 1, wherein: The image data includes first image data and second image data. After preprocessing the original patient data, the method further includes: The pixel values ​​of the first image data and the pixel values ​​of the second image data are weightedly fused to obtain preprocessed image data.

4. The method for identifying cognitive impairment in patients according to claim 1, wherein: Feature extraction is performed on the preprocessed data to obtain key features, including: The first formula is used to extract features from the preprocessed clinical biochemical data to obtain clinical biochemical features, wherein the expression of the first formula is: ， in, Indicates clinical and biochemical characteristics, Indicates the starting time point of preprocessing clinical biochemical data, that is, the starting time point of preprocessing data, Indicates the end time point of preprocessing clinical biochemical data, that is, the end time point of preprocessing data, Indicates at a point in time The hormone level at which represents baseline hormone levels, Indicates hormone levels Time point The second derivative of represents the power law adjustment coefficient, Indicates the start time and end time point The first discrete time points, Indicates multiplication.

5. The method for identifying cognitive impairment in patients according to claim 4, characterized in that: Feature extraction is performed on the preprocessed data to obtain key features, including: The second formula is used to extract features from the preprocessed image data to obtain image features, wherein the expression of the second formula is: ， in, Represents image features, represents the integrated area of ​​the brain scan in the preprocessed image data, Represents the coordinates in the preprocessed image data The grayscale value of the pixel at represents the standard deviation of the preprocessed image data, represents the mean of the preprocessed image data, represents the divergence operator, Represents the preprocessed image data Axial gradient phase, Indicates multiplication.

6. The method for identifying cognitive impairment in patients according to claim 5, characterized in that: Feature extraction is performed on the preprocessed data to obtain key features, including: The third formula is used to extract features from the pre-processed cognitive behavior data to obtain cognitive behavior features, wherein the expression of the third formula is: ， in, Represents cognitive behavioral characteristics, Represents the preprocessed cognitive behavioral data at time point The reaction time series at Represents the preprocessed cognitive behavioral data at time point The reaction time series at The periodic frequency, Represents the pre-processed cognitive behavior data at time point in a time period The reaction time series at The average value of the time period is the starting time point and end time point Any time period between represents the time decay window factor, Represents the preprocessed cognitive behavioral data at time point The reaction time series at The speed change influence factor, represents the penalty coefficient factor, Indicates multiplication.

7. The method for identifying cognitive impairment in patients according to claim 6, characterized in that: The expression of the fusion feature is: ， in, represents the fusion feature, represents the hyperbolic characteristic function, Indicates the The temporal dynamic weight of each mode, Indicates the The characteristic variance of the modes, Indicates the The features or feature matrix of each mode, Represents the element-by-element multiplication symbol, represents the feature weight matrix, Indicates the time point Find the gradient, Represents the feature weight matrix at time point The time gradient at Indicates multiplication.

8. The method for identifying cognitive impairment in patients according to claim 7, characterized in that: The expression of the cognitive impairment identification function is: ， in, represents the cognitive impairment score, Indicates the starting time point of preprocessing data, Indicates the end time point of preprocessing data, Indicates the Fusion features At the time point The value at represents the power law adjustment coefficient, Indicates the cognitive behavioral characteristics At the time point The measured value at Indicates the cognitive behavioral characteristics At the start time The baseline value at Indicates a time point changes, Indicates the cognitive behavioral characteristics The weight tilt angle, represents the inverse tangent function, represents the fusion feature, represents the integrated area of ​​the brain scan in the preprocessed image data, represents the gradient field of the preprocessed image features, represents multiplication, represents the norm, Represents cognitive behavioral characteristics The number of .

9. A system for identifying cognitive impairment in patients, characterized in that: The patient cognitive impairment identification system comprises: A data acquisition module is used to acquire original patient data, including clinical biochemical data, imaging data, and cognitive behavioral data; A preprocessing module, configured to preprocess the original patient data to obtain preprocessed data, wherein the preprocessed data includes preprocessed clinical biochemical data, preprocessed imaging data, and preprocessed cognitive behavioral data; A feature acquisition module is used to extract features from the preprocessed data to obtain key features, wherein the key features include clinical biochemical features, imaging features, and cognitive behavioral features; A fusion module, used to fuse the clinical biochemical features, imaging features and cognitive behavioral features to obtain a fusion feature; The result acquisition module is used to determine the cognitive impairment recognition result based on the fusion feature and the cognitive impairment recognition function.

10. An electronic device, characterized in that The electronic device comprises: one or more processing devices; a storage device for storing one or more programs, Wherein, when the one or more programs are executed by the one or more processing devices, the method for identifying cognitive impairment of a patient according to any one of claims 1-8 is implemented.

Images