Hepatolenticular degeneration clinical phenotype screening and determining method and system based on EHR

Abstract

The invention discloses an EHR-based hepatolenticular degeneration clinical phenotype screening and determining method and system, relates to the technical field of medical data processing, and is used for judging which magnetic resonance imaging technology is adopted to detect a patient by using a selection rule based on clinical data of the patient. A patient is detected based on the adopted magnetic resonance imaging technology, imaging data are obtained, and the imaging data and clinical data are combined to generate a craniocerebral MR phenotype report of the patient. The system automatically judges and can dynamically adjust the image technology according to real-time data (such as disease courses, symptoms, laboratory examination results and the like) of the patient, a more suitable diagnosis tool is provided, an image mode is recommended in a personalized manner based on the condition of the patient, and the diagnosis accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data processing, and in particular to an EHR-based method and system for screening and determining a clinical phenotype of hepatolenticular degeneration. Background Art

[0002] Hepatolenticular degeneration is a common hereditary copper metabolism disorder that mainly affects the liver and brain. Its clinical manifestations are diverse and may include liver damage, neurological symptoms (such as movement disorders, psychiatric symptoms), kidney lesions, etc. The electronic health record system (EHR) is an important part of modern medical information management. It records all medical data of patients from diagnosis, treatment to follow-up. These data include medical records, laboratory test results, medical images, prescription information, etc., which can provide rich clinical information. Clinical phenotypic screening aims to identify individuals with specific disease characteristics from a large amount of patient data. For hepatolenticular degeneration, this means extracting relevant diagnostic codes, laboratory results, clinical symptoms, imaging reports and other information from EHR. As the amount of EHR data continues to grow, how to efficiently store and manage this data has become a challenge;

[0003] The prior art has the following defects:

[0004] 1. In existing technologies, the choice of MRI technology usually depends on the doctor's experience rather than the patient's individual data. The selection of multimodal imaging (such as QSM, T2mapping, and IDEAL-IQ) is complicated. Without the assistance of an automated system, it may lead to unnecessary imaging examinations or miss more valuable tests.

[0005] 2. Existing imaging detection technologies are often based on static decisions and lack flexibility. They cannot adjust the imaging method in time as the patient's condition progresses. Patients with Wilson's disease at different stages require different imaging modes. Existing technologies are difficult to dynamically adapt to this demand, and the detection scheme lacks personalization.

[0006] Based on this, the present invention proposes a method and system for screening and determining the clinical phenotype of Wilson's disease based on EHR. The system automatically determines that it can dynamically adjust the imaging technology according to the patient's real-time data (such as course of disease, symptoms, laboratory test results, etc.), provide more appropriate diagnostic tools, and recommend imaging modes based on the patient's condition to improve diagnostic accuracy. Summary of the invention

[0007] The purpose of the present invention is to provide a method and system for screening and determining the clinical phenotype of Wilson's disease based on EHR to address the deficiencies in the background technology.

[0008] In order to achieve the above object, the present invention provides the following technical solution: a method for screening and determining the clinical phenotype of hepatolenticular degeneration based on EHR, the method comprising the following steps:

[0009] Collect clinical data related to Wilson's disease from the patient's electronic health record system and annotate clinical information related to Wilson's disease using natural language processing technology;

[0010] Based on the clinical data of the patient, using a selection rule to determine which magnetic resonance imaging technology to use to examine the patient, and based on the adopted magnetic resonance imaging technology, examining the patient and acquiring imaging data;

[0011] Imaging data are combined with clinical data to generate a report on the patient's brain MR phenotype.

[0012] In a preferred embodiment, clinical data related to Wilson's disease are collected from the patient's electronic health record system, and the clinical data include serum copper level, urinary copper excretion, serum ferritin level, abnormal values ​​of MRI basal ganglia signals, and abnormal values ​​of liver function indicators.

[0013] In a preferred embodiment, based on the clinical data of the patient, using a selection rule to determine which magnetic resonance imaging technology to use to detect the patient comprises the following steps:

[0014] After obtaining the patient's serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality;

[0015] The Q weights of the serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality of the patients were obtained by QSM technology, and the selection score of the QSM technology was calculated by weighting the serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality;

[0016] Obtain the T weights of the T2mapping technology for the patient's serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality, and weight the serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality to calculate the selection score of the T2mapping technology;

[0017] Obtain the I weights of the IDEAL-IQ technology for the patient's serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality, and weight the serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality to calculate the IDEAL-IQ technology selection score;

[0018] After obtaining the selection scores of the QSM technology, the T2mapping technology, and the IDEAL-IQ technology, the patient was examined using the magnetic resonance imaging technology with the largest selection score.

[0019] In a preferred embodiment, the logic for obtaining the Q weight, T weight and I weight is as follows: obtaining the historical usage frequency and historical accuracy of magnetic resonance imaging technology in clinical data detection, normalizing the historical usage frequency and historical accuracy so that the value range of the historical usage frequency and the historical accuracy is mapped to between [0,1], obtaining the normalized value of the historical usage frequency and the normalized value of the historical accuracy, summing the normalized value of the historical usage frequency and the normalized value of the historical accuracy to obtain the selection probability, for a certain type of clinical data, after obtaining the selection probability of the QSM technology, the selection probability of the T2-mapping technology, and the selection probability of the IDEAL-IQ technology, sum the selection probability of the QSM technology, the selection probability of the T2-mapping technology, and the selection probability of the IDEAL-IQ technology to obtain the total probability, obtain the Q weight corresponding to the clinical data by dividing the selection probability of the QSM technology by the total probability, obtain the T weight corresponding to the clinical data by dividing the selection probability of the T2-mapping technology by the total probability, and obtain the I weight corresponding to the clinical data by dividing the selection probability of the IDEAL-IQ technology by the total probability.

[0020] In a preferred embodiment, the imaging data is combined with the clinical data to generate a cranial MR phenotype report of the patient, comprising the following steps:

[0021] If the QSM technique is used, the imaging data obtained is the magnetic susceptibility of the basal ganglia, if the T2-mapping technique is used, the imaging data obtained is the signal intensity of the tissue structure, and if the IDEAL-IQ technique is used, the imaging data obtained is the fat fraction;

[0022] According to the magnetic resonance technology used, the imaging data is combined with the clinical data to obtain the evaluation coefficient, which is expressed as: , where is the evaluation coefficient, For the imaging data from magnetic resonance imaging techniques, For clinical data, The number of clinical data is shown. Indicates The deviation value of clinical data, , are the adjustment coefficients for imaging data and clinical data, respectively, and , All are greater than 0;

[0023] The evaluation coefficient is compared with the preset evaluation threshold. If the evaluation coefficient is less than or equal to the evaluation threshold, it is judged that the patient's Wilson's disease condition is mild. If the evaluation coefficient is greater than the evaluation threshold, it is judged that the patient's Wilson's disease condition is severe.

[0024] In a preferred embodiment, the magnetic susceptibility of the basal ganglia The calculation expression is: , where is the magnetic field disturbance of the basal ganglia, is the main magnetic field strength of the basal ganglia;

[0025] Signal strength of tissue structure The calculation expression is: , where is the signal strength of the tissue structure, is the initial signal strength of the tissue structure, for Moments of relaxation time, for relaxation time of the moment;

[0026] Fat Fraction The calculation expression is: , where is the fat signal intensity, is the water signal strength.

[0027] The clinical phenotype screening and determination system for hepatolenticular degeneration based on EHR includes a data acquisition module, a technology selection module, and a report display module;

[0028] Data acquisition module: collect clinical data related to Wilson's disease from the patient's electronic health record system, and annotate clinical information related to Wilson's disease through natural language processing technology;

[0029] Technology selection module: Based on the patient's clinical data, the selection rules are used to determine which magnetic resonance imaging technology to use to examine the patient;

[0030] Report presentation module: Based on the magnetic resonance imaging technology used, the patient is tested and imaging data is acquired, and the imaging data is combined with clinical data to generate a cranial MR phenotype report for the patient.

[0031] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0032] The present invention uses selection rules to determine which magnetic resonance imaging technology to use for testing patients based on the patient's clinical data, tests the patient based on the adopted magnetic resonance imaging technology and obtains imaging data, and generates a cranial MR phenotype report for the patient after combining the imaging data with the clinical data. The system automatically determines that it can dynamically adjust the imaging technology based on the patient's real-time data (such as course of disease, symptoms, laboratory test results, etc.), provide more appropriate diagnostic tools, and recommend imaging modes based on the patient's condition to improve diagnostic accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0034] Figure 1 The present invention is a flow chart of the method. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution 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 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.

[0036] Example 1: Please refer to Figure 1 As shown, the clinical phenotype screening and determination method of Wilson's disease based on EHR described in this embodiment comprises the following steps:

[0037] Clinical data related to hepatolenticular degeneration (HLD) were collected from the patient's electronic health record (EHR) system, and the clinical information related to hepatolenticular degeneration was annotated using natural language processing technology. Based on the patient's clinical data, selection rules were used to determine which magnetic resonance imaging technology to use to test the patient. The patient was tested and imaging data was obtained based on the magnetic resonance imaging technology used. The imaging data was combined with the clinical data to generate the patient's cranial MR phenotype report.

[0038] This application uses selection rules to determine which magnetic resonance imaging technology to use for testing patients based on the patient's clinical data, tests patients based on the adopted magnetic resonance imaging technology and obtains imaging data, and generates a cranial MR phenotype report for the patient after combining the imaging data with clinical data. The system automatically determines that it can dynamically adjust the imaging technology based on the patient's real-time data (such as course of disease, symptoms, laboratory test results, etc.), provide more appropriate diagnostic tools, and recommend imaging modes based on the patient's condition to improve diagnostic accuracy.

[0039] Example 2: Collecting clinical data related to hepatolenticular degeneration (HLD) from a patient's electronic health record (EHR) system, comprising the following steps:

[0040] Clinical data related to Wilson's disease were collected from the patients' electronic health record system. Clinical data included serum copper level, urinary copper excretion, serum ferritin level, abnormal values ​​of basal ganglia signals on MRI, and abnormal values ​​of liver function indexes (ALT / AST);

[0041] Key types of data collected from EHR systems:

[0042] Serum copper level: The main indicator of abnormal copper metabolism, usually obtained through a blood test.

[0043] Urinary copper excretion: Detecting copper excretion is an important diagnostic indicator for Wilson's disease.

[0044] Serum ferritin levels: monitor iron metabolism and can help detect copper and iron deposition in the basal ganglia.

[0045] MRI basal ganglia signal abnormalities: Through brain MRI scans, the signal changes of the basal ganglia are observed to detect nerve damage and metal deposition.

[0046] Liver function indicators (ALT / AST): obtained through liver function tests, reflecting liver health and liver cell damage.

[0047] In an EHR system, different types of data may be stored in different modules. They can be collected from the following sources:

[0048] Laboratory test module: stores laboratory test results such as serum copper level, urinary copper excretion, serum ferritin level, ALT / AST, etc.

[0049] Imaging module: MRI images and their reports are usually stored in the radiology module or PACS (Picture-Archiving-and-Communication-System). Abnormal values ​​of basal ganglia signals can be extracted through imaging reports.

[0050] Clinical diagnosis and recording module: Clinical manifestations recorded by doctors, summaries of laboratory tests, and the patient's historical medical records, etc.

[0051] Extract structured data from the EHR through a standardized data interface (FHIR or HL7), combine NLP technology to parse unstructured text, and generate standardized medical data;

[0052] FHIR (Fast-Healthcare-Interoperability-Resources) and HL7 (HealthLevel7) are standard protocols for medical data exchange. Structured data in the EHR can be extracted through these standard interfaces.

[0053] FHIR resources: Common resource types include Observation (for laboratory test results), Diagnostic-Report (for imaging reports), etc. Data such as serum copper and urinary copper excretion can be obtained through these resource types.

[0054] Annotate clinical information related to Wilson's disease through natural language processing technology, including the following steps:

[0055] Imaging reports and diagnostic reports are sometimes in unstructured text form. Natural language processing (NLP) technology can be used to extract useful information from these texts, such as descriptions of abnormal signals in the MRI basal ganglia.

[0056] Tools: Use NLP libraries (such as Spacy, BERT, cTAKES, etc.) or NLP tools dedicated to the medical field to extract specific abnormal signal values or descriptions from radiology reports.

[0057] Extract imaging reports and diagnostic reports from the EHR system. These are usually natural language texts written by doctors or recorded by the system, and the data may exist in multiple formats, including PDF, spreadsheets, plain text, JSON, etc. First, convert these data into plain text format for easy NLP processing, remove irrelevant content, such as report headers, footers, annotations, special characters, etc., and standardize the text, such as unifying case, removing stop words (such as "de", "he"), removing punctuation marks, etc.

[0058] Before starting automated processing, some labeled training data need to be provided to the model, that is, manually extract key medical terms and important information in the text, manually annotate or annotate medical entities in the text through existing medical term dictionaries (such as UMLS, SNOMEDCT), such as "abnormal signal in the basal ganglia" or "abnormal brain MRI".

[0059] Build a dictionary of medical terms related to HLD, including:

[0060] Organs or structures: basal ganglia, lenticular nucleus, red nucleus, etc.;

[0061] Imaging manifestations: signal enhancement, signal reduction, abnormality, lesions, etc.;

[0062] Disease states: metal deposition, tissue necrosis, etc.

[0063] Use NLP libraries (such as Spacy, NLTK, BERT, etc.) to segment the text, decompose the sentences and words in the report into basic units, use the named entity recognition technology in NLP to automatically identify and extract medical terms (such as anatomical structure, pathological status, imaging features, etc.), use ready-made medical NER models, or train through annotated medical data sets to automatically extract key information such as basal ganglia signal abnormalities.

[0064] Identify specific image features using a rule-based approach. For example, you can set the following rules:

[0065] Anatomical structures (e.g., basal ganglia, lenticular nucleus);

[0066] Signal anomaly type (enhanced, weakened, abnormal, etc.);

[0067] Description of severity (mild, moderate, significant, etc.).

[0068] Use pre-trained models specifically for medical imaging (such as BioBERT and ClinicalBERT) to extract features from imaging reports. These models can better understand the terms and relationships in the medical field, and further extract the relationships between the extracted entities. For example, the relationship between "basal ganglia" and "significant signal abnormality" is "anatomical structure-abnormal manifestation". Through dependency syntax parsing, find the grammatical relationship between entities, such as subject-verb-object relationship, modification relationship, etc.

[0069] Convert the extracted entities and relationships into structured data for subsequent analysis or storage in formats such as JSON and CSV for easy integration into databases.

[0070] The extracted structured information is stored in the database to facilitate subsequent query and analysis.

[0071] You can use a relational database (such as MySQL) or a NoSQL database (such as MongoDB) to store this data.

[0072] The extracted imaging features can be combined with other clinical data for more in-depth analysis, such as analyzing the relationship between abnormal MRI signals and serum copper levels to assist in diagnosis and treatment.

[0073] Integrate NLP processing results into the EHR system to automatically update and extract key information from imaging reports.

[0074] Realize the doctor's auxiliary decision-making function and automatically recommend detection technology or diagnosis plan.

[0075] Based on the clinical data of the patient, a selection rule is used to determine which magnetic resonance imaging technology to use to examine the patient, including the following steps:

[0076] The processing logic of the selection rule is as follows: after obtaining the patient's serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality, obtain the Q weight of the QSM technology for the patient's serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality, and weight the serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality to calculate the selection score of the QSM technology, and the expression is:

[0077] , where Rate your selection of QSM technologies, , , , , They are serum copper level, urinary copper excretion, serum ferritin level, abnormal values ​​of MRI basal ganglia signals, and abnormal values ​​of liver function indicators. , , , , is the Q weight of each clinical data, and , , , , All are greater than 0;

[0078] After obtaining the patient's serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality, the T weight of the T2mapping technology for the patient's serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality was obtained. The serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality were weighted to calculate the selection score of the T2mapping technology. The expression is:

[0079] , where Score the selection of T2mapping techniques, , , , , They are serum copper level, urinary copper excretion, serum ferritin level, abnormal values ​​of MRI basal ganglia signals, and abnormal values ​​of liver function indicators. , , , , is the T weight of each clinical data, and , , , , All are greater than 0;

[0080] After obtaining the patient's serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality, the I weights of the IDEAL-IQ technology for the patient's serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality were obtained. The serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality were weighted to calculate the selection score of the IDEAL-IQ technology. The expression is:

[0081] , where Rate your IDEAL-IQ technology selections, , , , , They are serum copper level, urinary copper excretion, serum ferritin level, abnormal values ​​of MRI basal ganglia signals, and abnormal values ​​of liver function indicators. , , , , is the I weight of each clinical data, and , , , , All are greater than 0;

[0082] After obtaining the selection scores of the QSM technology, the T2mapping technology, and the IDEAL-IQ technology, the patient was examined using the magnetic resonance imaging technology with the largest selection score.

[0083] Among them, the acquisition logic of Q weight, T weight and I weight is as follows: obtain the historical usage frequency and historical accuracy of magnetic resonance imaging technology in clinical data detection, normalize the historical usage frequency and historical accuracy, map the value range of the historical usage frequency and historical accuracy to [0,1], obtain the normalized value of the historical usage frequency and the normalized value of the historical accuracy, sum the normalized value of the historical usage frequency and the normalized value of the historical accuracy to obtain the selection probability, for a certain clinical data, after obtaining the selection probability of QSM technology, the selection probability of T2-mapping technology, and the selection probability of IDEAL-IQ technology, sum the selection probability of QSM technology, the selection probability of T2-mapping technology, and the selection probability of IDEAL-IQ technology to obtain the total probability, obtain the Q weight corresponding to the clinical data by dividing the selection probability of QSM technology by the total probability, obtain the T weight corresponding to the clinical data by dividing the selection probability of T2-mapping technology by the total probability, and obtain the I weight corresponding to the clinical data by dividing the selection probability of IDEAL-IQ technology by the total probability;

[0084] For example, if the clinical data is serum copper level, the greater the selection probability of the magnetic resonance technology, the better the performance of the magnetic resonance technology on the serum copper level. The selection probability of the QSM technology for detecting serum copper level is obtained, the selection probability of the T2-mapping technology for detecting serum copper level is obtained, and the selection probability of the IDEAL-IQ technology for detecting serum copper level is obtained. The selection probabilities of the QSM technology, the T2-mapping technology, and the IDEAL-IQ technology are summed to obtain the total probability. The Q weight corresponding to the clinical data is obtained by dividing the selection probability of the QSM technology by the total probability, the T weight corresponding to the clinical data is obtained by dividing the selection probability of the T2-mapping technology by the total probability, and the I weight corresponding to the clinical data is obtained by dividing the selection probability of the IDEAL-IQ technology by the total probability, that is, .

[0085] The patient is examined and imaging data is obtained based on the adopted magnetic resonance imaging technology, including the following steps:

[0086] The patient is positioned in the MRI machine, making sure the area to be scanned (e.g. the brain) is centered in the MRI device.

[0087] Equipment calibration: adjust equipment parameters, such as field strength, gradient coils, etc., to ensure imaging quality.

[0088] Select sequence: Set the scanning sequence according to the selected imaging technique (such as QSM, T2-mapping, IDEAL-IQ dedicated sequence).

[0089] Start scanning: Start acquiring magnetic resonance imaging data according to the preset scanning protocol.

[0090] Collecting image data: The machine acquires imaging data of the brain, basal ganglia and other areas of interest through different sequences. The type of data collected varies for different technologies.

[0091] QSM: Produces high-contrast images related to the concentration of magnetic materials (e.g., iron, copper).

[0092] T2-mapping: Generates images reflecting tissue T2 relaxation time for assessing tissue edema or fibrosis.

[0093] IDEAL-IQ: Acquire images for fat-water separation and quantification.

[0094] Image reconstruction: Convert the original MRI signal data into visual image data through computer algorithms.

[0095] Artifact removal and filtering: Perform post-processing operations such as artifact correction and noise filtering on the image to improve image quality.

[0096] Image registration and calibration: Ensure that images from different imaging sequences have the same spatial resolution and alignment to facilitate subsequent analysis.

[0097] Generate a quantitative parametric plot:

[0098] QSM images provide quantitative data on magnetic susceptibility.

[0099] T2-mapping images can generate quantitative T2 relaxation time maps.

[0100] The IDEAL-IQ image provides fat mass and other metabolic information.

[0101] The imaging data is combined with the clinical data to generate a patient's cranial MR phenotype report, which includes the following steps:

[0102] If the QSM technique is used, the imaging data obtained is the magnetic susceptibility of the basal ganglia, if the T2-mapping technique is used, the imaging data obtained is the signal intensity of the tissue structure, and if the IDEAL-IQ technique is used, the imaging data obtained is the fat fraction;

[0103] Magnetic susceptibility of the basal ganglia The calculation expression is: , where is the magnetic field disturbance of the basal ganglia, is the main magnetic field strength of the basal ganglia;

[0104] Signal strength of tissue structure The calculation expression is: , where is the signal strength of the tissue structure, is the initial signal strength of the tissue structure, is the relaxation time, which reflects the relaxation characteristics within the tissue, The longer the time, the more severe the tissue edema, fibrosis or damage. The basal ganglia area of ​​patients with Wilson's disease may show Extension of time;

[0105] Fat Fraction The calculation expression is: , where is the fat signal intensity, is the water signal strength.

[0106] According to the magnetic resonance technology used, the imaging data is combined with the clinical data to obtain the evaluation coefficient, which is expressed as: , where is the evaluation coefficient, For the The imaging data of magnetic resonance imaging technology, For clinical data, The number of clinical data is shown. Indicates The deviation value of clinical data, , are the adjustment coefficients for imaging data and clinical data, respectively, and , All are greater than 0;

[0107] The larger the evaluation coefficient is, the more severe the patient's hepatolenticular degeneration is. The evaluation coefficient is compared with the preset evaluation threshold. If the evaluation coefficient is less than or equal to the evaluation threshold, the patient's hepatolenticular degeneration is judged to be mild. If the evaluation coefficient is greater than the evaluation threshold, the patient's hepatolenticular degeneration is judged to be severe.

[0108] The calculation logic of the deviation value of clinical data is: subtract the standard clinical data value from the real-time clinical data value to obtain the deviation value of the clinical data.

[0109] The patient's brain MR phenotype report is generated by integrating the judgment results of the patient's hepatolenticular degeneration status, imaging data and clinical data, which specifically includes the following steps:

[0110] The integrated results will be generated into a report, including the following parts:

[0111] Basic patient information: such as age, gender, medical history, etc.

[0112] Imaging analysis results:

[0113] QSM data: reports basal ganglia magnetic susceptibility values ​​and quantification of copper and iron deposition.

[0114] T2 mapping data: Reports T2 relaxation time and edema or damage.

[0115] IDEAL-IQ data: Reports brain fat content and metabolic abnormalities.

[0116] Clinical data analysis: analysis results of serum copper, urinary copper excretion, serum ferritin, liver function indicators, etc.

[0117] Comprehensive analysis and evaluation: Combine clinical and imaging data to provide disease severity assessment and give further treatment recommendations.

[0118] Example 3: The EHR-based hepatolenticular degeneration clinical phenotype screening and determination system described in this example includes a data acquisition module, a technology selection module, and a report display module;

[0119] Data acquisition module: collects clinical data related to hepatolenticular degeneration (HLD) from the patient's electronic health record (EHR) system, annotates clinical information related to hepatolenticular degeneration through natural language processing technology, and sends the clinical data to the technology selection module;

[0120] Technology selection module: Based on the patient's clinical data, the selection rules are used to determine which magnetic resonance imaging technology to use to test the patient, and the magnetic resonance imaging technology used is sent to the report display module;

[0121] Report presentation module: Based on the magnetic resonance imaging technology used, the patient is tested and imaging data is acquired, and the imaging data is combined with clinical data to generate a cranial MR phenotype report for the patient.

[0122] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0123] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0124] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for screening and determining the clinical phenotype of Wilson's disease based on EHR, characterized in that: The determination method comprises the following steps: Collect clinical data related to Wilson's disease from the patient's electronic health record system and annotate clinical information related to Wilson's disease using natural language processing technology; Based on the clinical data of the patient, using a selection rule to determine which magnetic resonance imaging technology to use to examine the patient, and based on the adopted magnetic resonance imaging technology, examining the patient and acquiring imaging data; Imaging data are combined with clinical data to generate a report on the patient's brain MR phenotype.

2. The method for screening and determining the clinical phenotype of Wilson's disease based on EHR according to claim 1, characterized in that: Clinical data related to Wilson's disease were collected from the patients' electronic health record system, including serum copper level, urinary copper excretion, serum ferritin level, abnormal values ​​of basal ganglia signals on MRI, and abnormal values ​​of liver function indicators.

3. The method for screening and determining the clinical phenotype of Wilson's disease based on EHR according to claim 2, characterized in that: Based on the clinical data of the patient, a selection rule is used to determine which magnetic resonance imaging technology to use to examine the patient, including the following steps: After obtaining the patient's serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality; The Q weights of the serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality of the patients were obtained by QSM technology, and the selection score of the QSM technology was calculated by weighting the serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality; Obtain the T weights of the T2mapping technology for the patient's serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality, and weight the serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality to calculate the selection score of the T2mapping technology; Obtain the I weights of the IDEAL-IQ technology for the patient's serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality, and weight the serum copper level, urinary copper excretion, serum ferritin level, MRI basal ganglia signal abnormality, and liver function index abnormality to calculate the IDEAL-IQ technology selection score; After obtaining the selection scores of the QSM technology, the T2mapping technology, and the IDEAL-IQ technology, the patient was examined using the magnetic resonance imaging technology with the largest selection score.

4. The method for screening and determining the clinical phenotype of Wilson's disease based on EHR according to claim 3, characterized in that: The logic for obtaining the Q weight, T weight, and I weight is as follows: obtain the historical frequency of use and the historical accuracy of magnetic resonance imaging technology in clinical data detection, normalize the historical frequency of use and the historical accuracy, map the value range of the historical frequency of use and the historical accuracy to [0,1], obtain the normalized value of the historical frequency of use and the normalized value of the historical accuracy, sum the normalized value of the historical frequency of use and the normalized value of the historical accuracy to obtain the selection probability, for a certain clinical data, after obtaining the selection probability of the QSM technology, the selection probability of the T2-mapping technology, and the selection probability of the IDEAL-IQ technology, sum the selection probability of the QSM technology, the selection probability of the T2-mapping technology, and the selection probability of the IDEAL-IQ technology to obtain the total probability, obtain the Q weight corresponding to the clinical data by dividing the selection probability of the QSM technology by the total probability, obtain the T weight corresponding to the clinical data by dividing the selection probability of the T2-mapping technology by the total probability, and obtain the I weight corresponding to the clinical data by dividing the selection probability of the IDEAL-IQ technology by the total probability.

5. The method for screening and determining the clinical phenotype of Wilson's disease based on EHR according to claim 4, characterized in that: The imaging data is combined with the clinical data to generate a patient's cranial MR phenotype report, which includes the following steps: If the QSM technique is used, the imaging data obtained is the magnetic susceptibility of the basal ganglia, if the T2-mapping technique is used, the imaging data obtained is the signal intensity of the tissue structure, and if the IDEAL-IQ technique is used, the imaging data obtained is the fat fraction; According to the magnetic resonance technology used, the imaging data is combined with the clinical data to obtain the evaluation coefficient, which is expressed as: , where is the evaluation coefficient, For the imaging data from magnetic resonance imaging techniques, For clinical data, The number of clinical data is shown. Indicates The deviation value of clinical data, , are the adjustment coefficients for imaging data and clinical data, respectively, and , All are greater than 0; The evaluation coefficient is compared with the preset evaluation threshold. If the evaluation coefficient is less than or equal to the evaluation threshold, it is judged that the patient's Wilson's disease condition is mild. If the evaluation coefficient is greater than the evaluation threshold, it is judged that the patient's Wilson's disease condition is severe.

6. The method for screening and determining the clinical phenotype of Wilson's disease based on EHR according to claim 5, characterized in that: Magnetic susceptibility of the basal ganglia The calculation expression is: , where is the magnetic field disturbance of the basal ganglia, is the main magnetic field strength of the basal ganglia; Signal strength of tissue structure The calculation expression is: , where is the signal strength of the tissue structure, is the initial signal strength of the tissue structure, for Moments of relaxation time, for relaxation time of the moment; Fat Fraction The calculation expression is: , where is the fat signal intensity, is the water signal strength.

7. An EHR-based system for screening and determining the clinical phenotype of hepatolenticular degeneration, used to implement the determination method according to any one of claims 1 to 6, characterized in that: It includes data acquisition module, technology selection module and report display module; Data acquisition module: collect clinical data related to Wilson's disease from the patient's electronic health record system, and annotate clinical information related to Wilson's disease through natural language processing technology; Technology selection module: Based on the patient's clinical data, the selection rules are used to determine which magnetic resonance imaging technology to use to examine the patient; Report presentation module: Based on the magnetic resonance imaging technology used, the patient is tested and imaging data is acquired, and the imaging data is combined with clinical data to generate a cranial MR phenotype report for the patient.

Images