Artificial intelligence diagnosis system and method for nervous system diseases

Abstract

The invention discloses an artificial intelligence diagnosis system and method for nervous system diseases, and the system comprises a data collection unit, a feature extraction unit, a feature information library, a data analysis unit, a data processing unit, a diagnosis analysis unit, and a result display unit, and relates to the technical field of intelligent diagnosis. According to the method, related feature information of the target patient can be comprehensively obtained, analysis is carried out in combination with feature vectors of healthy personnel and diagnosed patients, the accuracy and reliability of diagnosis are improved, accurate illness state assessment can be carried out according to individual differences, and a one-step diagnosis mode is avoided; through the preset judgment threshold set, the patients can be clearly distinguished into mild patients, moderate patients and severe patients, and a clear guidance direction is provided for subsequent treatment and nursing; through comprehensive arrangement of diagnosis analysis values of diagnosed patients, the judgment threshold set is dynamically optimized, so that the judgment threshold set is more in line with actual clinical conditions, and the diagnosis accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent diagnosis technology, and specifically to an artificial intelligence diagnosis system and method for nervous system diseases. Background Art

[0002] Neurological diseases have always been a major challenge facing clinical medicine due to their complex causes, diverse symptoms, and difficult diagnosis. Traditional methods of diagnosing neurological diseases often rely on doctors' clinical experience, various imaging examinations, and laboratory tests. However, these methods have many disadvantages, such as strong subjectivity, long testing cycles, and high costs. Taking Parkinson's disease, a common neurodegenerative disease, as an example, traditional diagnosis mainly relies on doctors' clinical experience and observation. Subjectivity is difficult to avoid and has limitations, making it impossible to achieve early and accurate diagnosis, which is extremely unfavorable for timely treatment and improving patients' quality of life.

[0003] With the continuous advancement of medical technology and the rapid development of artificial intelligence, it has become possible to use big data and intelligent algorithms to assist in the diagnosis of neurological diseases. However, the existing diagnostic system is insufficient in terms of the breadth and comprehensiveness of data, and it is difficult to fully integrate a large amount of patient data from different medical institutions, resulting in the need to improve the accuracy and reliability of the diagnostic results; at the same time, there is a lack of clear and detailed standards and methods for determining the severity of the target patient's condition, which makes it difficult to meet the clinical needs for accurate graded diagnosis, which may affect the formulation and implementation of subsequent treatment plans.

[0004] Therefore, there is an urgent need for a more efficient, accurate and objective diagnostic method to address the challenges of diagnosing neurological diseases. Summary of the invention

[0005] In view of the deficiencies of the prior art, the present invention provides an artificial intelligence diagnosis system and method for nervous system diseases, which solves the problems raised in the background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: an artificial intelligence diagnosis system for nervous system diseases, comprising:

[0007] A data collection unit, used to collect various relevant characteristic data of the target patient;

[0008] A feature extraction unit, used to extract quantized values ​​corresponding to a plurality of specified feature vectors from a plurality of related feature data;

[0009] A feature information database, used to store multiple specified feature vectors corresponding to healthy persons and confirmed patients respectively;

[0010] A data analysis unit is used to perform a diagnostic analysis on a target patient based on the result extracted by the feature extraction unit and in combination with a plurality of specified feature vectors corresponding to healthy persons in a feature information database, and obtain a diagnostic analysis value corresponding to the target patient;

[0011] The diagnostic analysis unit is used to compare the diagnostic analysis value corresponding to the target patient with the preset judgment threshold set, and determine whether the target patient suffers from a nervous system disease based on the comparison result:

[0012] The result display unit is used to display the comparison result of the diagnosis and analysis unit to relevant personnel.

[0013] Preferably, the storage result in the feature information library is to collect a variety of relevant feature data of healthy persons and confirmed patients from multiple medical institutions through the data acquisition unit, and extract a plurality of specified feature vectors from the various relevant feature data according to the feature extraction unit.

[0014] Preferably, the diagnostic analysis method for the target patient is as follows:

[0015] StepX1, select a feature vector;

[0016] From the feature vector corresponding to the target patient, obtain n corresponding quantized values ​​and mark them as Li, i=1, 2, ...n;

[0017] Obtain n corresponding quantized values ​​from the healthy persons corresponding to the feature vector in the feature information database, and mark them as Ki, where i=1, 2, ...n;

[0018] StepX2, through Calculate the feature score value D of the target patient corresponding to the feature vector;

[0019] StepX3, calculate the feature score value of each feature vector through the method from StepX1 to StepX2, and mark the feature score value of each feature vector as Dj, j = 1, 2, ... m, m represents the number of all feature vectors;

[0020] StepX4, through Calculate the diagnostic analysis value G of the target patient;

[0021] In the formula, αj is the weight coefficient preset corresponding to Dj, and λ is the bias term.

[0022] Preferably, the comparison method of the diagnostic analysis unit is as follows:

[0023] Among them, the preset judgment threshold set is G0∈[G1min,G2min,G3min];

[0024] If G < G1min, it is determined that the target patient does not have a corresponding neurological disease;

[0025] If G1min≤G<G2min, the target patient is judged to be a mild patient of the corresponding nervous system disease;

[0026] If G2min≤G<G3min, the target patient is judged as a moderate patient of the corresponding nervous system disease;

[0027] If G≥G3min, the target patient is determined to be a severe patient of the corresponding neurological disease.

[0028] Preferably, the data analysis unit is further used to perform diagnostic analysis on all confirmed patients in the feature information database according to the method of performing diagnostic analysis on the target patient, and obtain diagnostic analysis values ​​corresponding to all confirmed patients respectively;

[0029] The system also includes:

[0030] The data processing unit is used to comprehensively organize the diagnostic analysis values ​​corresponding to all confirmed patients and obtain a determination threshold set.

[0031] Preferably, the comprehensive arrangement method is as follows:

[0032] StepY1. First, according to the severity of the disease, the confirmed patients are divided into mild patients, moderate patients, and severe patients;

[0033] Step Y2, select all moderate patients as an example, and mark the diagnostic analysis values ​​corresponding to all confirmed patients as Gt, t = 1, 2, ... e, e represents the number of all confirmed patients;

[0034] Step Y3, according to the formula Calculate the standard deviation value B corresponding to Gt;

[0035] Step Y4, then compare the standard deviation value B with the preset standard deviation threshold By:

[0036] When B>By, the corresponding Gt values ​​are deleted in descending order according to |Gt-Gp| until B≤By is established;

[0037] Among them, Gp is ​​the average value of m tyj, and the deleted tyj value does not participate in the subsequent calculation of Gp;

[0038] Step Y5. When B≤By, select a diagnostic analysis value with the largest value and the smallest value from all Gt involved in calculating the corresponding B value, and form the diagnostic judgment interval corresponding to the moderate patient;

[0039] Step Y6. According to the method of Step Y2 to Step Y5, the diagnostic judgment interval corresponding to the mild patients and the diagnostic judgment interval corresponding to the severe patients are calculated;

[0040] Step Y7, then obtain the corresponding minimum value from the diagnostic judgment intervals corresponding to mild patients, moderate patients and severe patients respectively;

[0041] Then the decision threshold set G0∈[G1min,G2min,G3min] is formed;

[0042] in,

[0043] G1min is the minimum value of the diagnostic judgment interval corresponding to mild patients;

[0044] G2min is the minimum value of the diagnostic judgment interval corresponding to moderate patients;

[0045] G3min is the minimum value of the diagnostic judgment interval corresponding to severe patients.

[0046] Preferably, the multiple related feature data include motion feature data, voice feature data, facial expression feature data and neuroimaging feature data;

[0047] in,

[0048] The motion feature data is obtained by using wearable devices or special motion monitoring instruments to obtain the target patient’s gait information and hand movement information;

[0049] The voice feature data is recorded by a microphone or voice acquisition device to record the target patient's speech content;

[0050] Facial expression feature data is captured by a camera or facial recognition technology to capture changes in facial expressions of target patients;

[0051] Neuroimaging feature data is obtained by using magnetic resonance imaging (MRI) and positron emission tomography (PET) medical imaging equipment to obtain structural and functional information of the target patient's brain;

[0052] Preferably, the plurality of specified feature vectors are as follows:

[0053] For motion data, the relevant features of walking speed and stride are extracted from the gait information; and the quantitative values ​​corresponding to the amplitude, frequency and coordination of hand movements are extracted from the hand movement information;

[0054] For speech data, the quantitative values ​​corresponding to the fundamental frequency, speech rate, intonation, speech clarity and fluency extracted from the speech content by the speech processing algorithm;

[0055] For facial expression data, the quantified values ​​corresponding to the frequency, duration, and amplitude of the corresponding expressions are extracted through image processing technology;

[0056] For neuroimaging data, image analysis algorithms are used to calculate and extract quantitative values ​​corresponding to the volume and metabolic rate of specific brain regions.

[0057] An artificial intelligence diagnosis method for nervous system diseases, an artificial intelligence diagnosis system for nervous system diseases described in the method is implemented, and the method comprises the following steps:

[0058] Step 1: Data Collection

[0059] Collect various relevant characteristic data of target patients;

[0060] Step 2: Feature extraction

[0061] Extracting quantized values ​​corresponding to a plurality of specified feature vectors from a plurality of related feature data;

[0062] Step 3: Build a feature information database

[0063] Storing multiple designated feature vectors corresponding to healthy persons and confirmed patients respectively, wherein the storage result is obtained by collecting multiple relevant feature data of healthy persons and confirmed patients from multiple medical institutions, and extracting multiple designated feature vectors from the multiple relevant feature data according to a feature extraction step;

[0064] Step 4: Data Analysis

[0065] According to the result of feature extraction and in combination with multiple specified feature vectors corresponding to healthy persons in the feature information database, a diagnostic analysis is performed on the target patient to obtain a diagnostic analysis value corresponding to the target patient;

[0066] Step 5: Data processing

[0067] According to the method of performing diagnostic analysis on the target patient, all the confirmed patients in the feature information database are diagnosed and analyzed, and the diagnostic analysis values ​​corresponding to all the confirmed patients are obtained; then the diagnostic analysis values ​​corresponding to all the confirmed patients are comprehensively sorted out to obtain a determination threshold set;

[0068] Step 6: Diagnostic Analysis

[0069] Comparing the diagnostic analysis value corresponding to the target patient with a preset judgment threshold set, and determining whether the target patient suffers from a neurological disease based on the comparison result;

[0070] Step 7: Results display

[0071] Present the comparative results of the diagnostic analysis to relevant personnel.

[0072] Beneficial Effects

[0073] The present invention provides an artificial intelligence diagnosis system and method for nervous system diseases. Compared with the prior art, it has the following beneficial effects:

[0074] Accurate diagnosis: Through multi-dimensional data collection and feature extraction, we can comprehensively obtain relevant characteristic information of target patients, combine the characteristic vectors of healthy people and confirmed patients for analysis, and improve the accuracy and reliability of diagnosis.

[0075] Personalized assessment: Calculate a unique diagnostic analysis value for each target patient, enabling accurate condition assessment based on individual differences, avoiding a one-size-fits-all diagnosis approach.

[0076] Clear disease classification: Through the preset judgment threshold set, patients can be clearly distinguished into mild, moderate and severe patients, providing clear guidance for subsequent treatment and care.

[0077] Rich and reliable data: The data in the feature information database comes from multiple medical institutions, which increases the diversity and representativeness of the data and makes the diagnosis results more universal and authoritative.

[0078] Optimize threshold judgment: Through the comprehensive collation of the diagnostic analysis values ​​of confirmed patients, dynamically optimize the judgment threshold set to make it more in line with the actual clinical situation and further improve the accuracy of diagnosis.

[0079] Intuitive display of results: The result display unit can show the diagnosis results to relevant personnel in a timely and clear manner, making it easier for the medical team to make decisions and develop treatment plans quickly.

[0080] Improve medical efficiency: Use artificial intelligence technology for rapid diagnosis and analysis, saving diagnosis time and helping to improve the efficiency of medical resource utilization.

[0081] Promote medical progress: It provides new technical means and methods for the diagnosis of neurological diseases, which helps to promote the development and innovation of the medical field in the diagnosis of such diseases. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 It is a system block diagram of an artificial intelligence diagnosis system and method for nervous system diseases of the present invention. DETAILED DESCRIPTION

[0083] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0084] As embodiment 1 of the present invention

[0085] See also Figure 1 The present invention provides a technical solution: an artificial intelligence diagnosis system for nervous system diseases, comprising:

[0086] A data acquisition unit, used to collect various relevant characteristic data of the target patient, wherein the various relevant characteristic data include motion characteristic data, voice characteristic data, facial expression characteristic data and neuroimaging characteristic data;

[0087] In this embodiment,

[0088] The motion feature data is obtained by using wearable devices or special motion monitoring instruments to obtain the target patient’s gait information and hand movement information;

[0089] The voice feature data is recorded by a microphone or voice acquisition device to record the target patient's speech content;

[0090] Facial expression feature data is captured by a camera or facial recognition technology to capture changes in facial expressions of target patients;

[0091] Neuroimaging feature data is obtained by using magnetic resonance imaging (MRI) and positron emission tomography (PET) medical imaging equipment to obtain structural and functional information of the target patient's brain;

[0092] A feature extraction unit, used to extract quantized values ​​corresponding to a plurality of specified feature vectors from a plurality of related feature data;

[0093] For motion data, the relevant features of walking speed and stride are extracted from the gait information; and the quantitative values ​​corresponding to the amplitude, frequency and coordination of hand movements are extracted from the hand movement information;

[0094] For speech data, the quantitative values ​​corresponding to the fundamental frequency, speech rate, intonation, speech clarity and fluency extracted from the speech content by the speech processing algorithm;

[0095] For facial expression data, the quantified values ​​corresponding to the frequency, duration, and amplitude of the corresponding expressions are extracted through image processing technology;

[0096] For neuroimaging data, image analysis algorithms are used to calculate and extract the quantitative values ​​corresponding to the volume and metabolic rate of specific brain regions;

[0097] A feature information database, used to store multiple specified feature vectors corresponding to healthy persons and confirmed patients respectively;

[0098] It collects various relevant feature data of healthy people and confirmed patients from multiple medical institutions through a data collection unit, and extracts multiple specified feature vectors from the various relevant feature data according to a feature extraction unit;

[0099] A data analysis unit is used to perform a diagnostic analysis on a target patient based on the result extracted by the feature extraction unit and in combination with a plurality of specified feature vectors corresponding to healthy persons in a feature information database, and obtain a diagnostic analysis value corresponding to the target patient;

[0100] The specific method is as follows:

[0101] StepX1, select a feature vector;

[0102] Obtain n corresponding quantized values ​​from the feature vector corresponding to the target patient and mark them as L i, i=1, 2, ...n;

[0103] Obtain n corresponding quantized values ​​from the healthy persons corresponding to the feature vector in the feature information database, and mark them as Ki, where i=1, 2, ...n;

[0104] StepX2, through and Calculate the mean values ​​Lp and Kp of all Li and Ki respectively;

[0105] StepX3, through

[0106] and

[0107] Calculate the variances FL and FK of all Li and Ki respectively;

[0108] StepX4, through Calculate the feature score value D of the target patient corresponding to the feature vector;

[0109] StepX5, calculate the feature score value of each feature vector through the method of StepX1 to StepX4, and mark the feature score value of each feature vector as Dj, j = 1, 2, ... m, m represents the number of all feature vectors;

[0110] StepX6, pass Calculate the diagnostic analysis value G of the target patient;

[0111] In the formula, αj is the weight coefficient preset corresponding to Dj, and λ is the bias term;

[0112] The diagnostic analysis unit is used to compare the diagnostic analysis value G corresponding to the target patient with the preset judgment threshold set G0∈[G1min,G2min,G3min]:

[0113] Among them, G1min<G2min<G3min;

[0114] If G<G1min, it is determined that the target patient does not have the corresponding neurological disease

[0115] If G1min≤G<G2min, the target patient is judged to be a mild patient of the corresponding nervous system disease;

[0116] If G2min≤G<G3min, the target patient is judged as a moderate patient of the corresponding nervous system disease;

[0117] If G≥G3min, the target patient is determined to be a severe patient of the corresponding neurological disease.

[0118] The result display unit is used to display the comparison result of the diagnosis and analysis unit to relevant personnel.

[0119] This embodiment can comprehensively collect a variety of relevant feature data of the target patient, including movement, voice, facial expression and neuroimaging, etc., thereby improving the comprehensiveness and accuracy of the diagnosis; by extracting and quantifying features of different types of data, it can more accurately analyze the patient's condition; diagnostic analysis is performed based on feature extraction results and feature vectors of healthy people, providing an objective basis for diagnosis; clear diagnostic analysis value calculation methods and diagnostic judgment thresholds enable diagnostic results to have clear grading, which helps doctors formulate corresponding treatment plans; the result display unit can intuitively display the diagnostic results to relevant personnel, thereby improving the efficiency of information transmission.

[0120] Embodiment 2

[0121] As the second embodiment of the present invention, when the present application is implemented, compared with the first embodiment, the technical solution of the present embodiment is different from that of the first embodiment only in that:

[0122] The data analysis unit is also used to perform diagnostic analysis on all confirmed patients in the feature information database according to the method of performing diagnostic analysis on the target patient, and obtain diagnostic analysis values ​​corresponding to all confirmed patients respectively;

[0123] This embodiment also includes:

[0124] A data processing unit is used to comprehensively sort out the diagnostic analysis values ​​corresponding to all confirmed patients and obtain a determination threshold set;

[0125] The specific method is as follows:

[0126] StepY1. First, according to the severity of the disease, the confirmed patients are divided into mild patients, moderate patients, and severe patients;

[0127] Step Y2, select all moderate patients as an example, and mark the diagnostic analysis values ​​corresponding to all confirmed patients as Gt, t = 1, 2, ... e, e represents the number of all confirmed patients;

[0128] Step Y3, according to the formula Calculate the standard deviation value B corresponding to Gt;

[0129] Step Y4, then compare the standard deviation value B with the preset standard deviation threshold By:

[0130] When B>By, the corresponding Gt values ​​are deleted in descending order according to |Gt-Gp| until B≤By is established;

[0131] Among them, Gp is ​​the average value of m tyj, and the deleted tyj value does not participate in the subsequent calculation of Gp;

[0132] Step Y5. When B≤By, select a diagnostic analysis value with the largest value and the smallest value from all Gt involved in calculating the corresponding B value, and form the diagnostic judgment interval corresponding to the moderate patient;

[0133] Step Y6. According to the method of Step Y2 to Step Y5, the diagnostic judgment interval corresponding to the mild patients and the diagnostic judgment interval corresponding to the severe patients are calculated;

[0134] Step Y7, then obtain the corresponding minimum value from the diagnostic judgment intervals corresponding to mild patients, moderate patients and severe patients respectively;

[0135] Then the decision threshold set G0∈[G1min,G2min,G3min] is formed;

[0136] in,

[0137] G1min is the minimum value of the diagnostic judgment interval corresponding to mild patients;

[0138] G2min is the minimum value of the diagnostic judgment interval corresponding to moderate patients;

[0139] G3min is the minimum value of the diagnostic judgment interval corresponding to severe patients.

[0140] In addition to performing diagnostic analysis on the target patient, this embodiment also analyzes all confirmed patients in the feature information library, making the diagnostic analysis more comprehensive and representative; a judgment threshold set is obtained by comprehensively sorting out the diagnostic analysis values ​​of confirmed patients, making the diagnostic judgment more scientific and accurate, and better adapting to the differences in the conditions of different patients; the screening and interval determination method based on the standard deviation improves the reliability and stability of the diagnostic judgment interval.

[0141] Embodiment 3

[0142] As the third embodiment of the present invention, when the present application is specifically implemented, compared with the first and second embodiments, the technical solution of this embodiment is to combine the solutions of the first and second embodiments mentioned above for implementation.

[0143] This embodiment combines the advantages of embodiment one and embodiment two, and not only realizes accurate diagnosis and analysis of target patients, but also optimizes the judgment threshold by processing a large amount of confirmed patient data, thereby improving the accuracy and reliability of the entire diagnostic system; the comprehensive solution can better adapt to different clinical situations and provide more complete technical support for the diagnosis of neurological diseases.

[0144] The present invention also provides an artificial intelligence diagnosis method for nervous system diseases, wherein the artificial intelligence diagnosis system for nervous system diseases described in the method is implemented, and the method comprises the following steps:

[0145] Step 1: Data Collection

[0146] Collect various relevant characteristic data of target patients;

[0147] Step 2: Feature extraction

[0148] Extracting quantized values ​​corresponding to a plurality of specified feature vectors from a plurality of related feature data;

[0149] Step 3: Build a feature information database

[0150] Storing multiple designated feature vectors corresponding to healthy persons and confirmed patients respectively, wherein the storage result is obtained by collecting multiple relevant feature data of healthy persons and confirmed patients from multiple medical institutions, and extracting multiple designated feature vectors from the multiple relevant feature data according to a feature extraction step;

[0151] Step 4: Data Analysis

[0152] According to the result of feature extraction and in combination with multiple specified feature vectors corresponding to healthy persons in the feature information database, a diagnostic analysis is performed on the target patient to obtain a diagnostic analysis value corresponding to the target patient;

[0153] Step 5: Data processing

[0154] According to the method of performing diagnostic analysis on the target patient, all the confirmed patients in the feature information database are diagnosed and analyzed, and the diagnostic analysis values ​​corresponding to all the confirmed patients are obtained; then the diagnostic analysis values ​​corresponding to all the confirmed patients are comprehensively sorted out to obtain a determination threshold set;

[0155] Step 6: Diagnostic Analysis

[0156] Comparing the diagnostic analysis value corresponding to the target patient with a preset judgment threshold set, and determining whether the target patient suffers from a neurological disease based on the comparison result;

[0157] Step 7: Results display

[0158] Present the comparative results of the diagnostic analysis to relevant personnel.

[0159] Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0160] 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 and thresholds in the formula are set by technicians in this field according to actual conditions.

[0161] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0162] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An artificial intelligence diagnostic system for neurological diseases, It is characterized in that include: A data collection unit, used to collect various relevant characteristic data of the target patient; A feature extraction unit, used to extract quantized values ​​corresponding to a plurality of specified feature vectors from a plurality of related feature data; A feature information database, used to store multiple specified feature vectors corresponding to healthy persons and confirmed patients respectively; A data analysis unit is used to perform a diagnostic analysis on a target patient based on the result extracted by the feature extraction unit and in combination with a plurality of specified feature vectors corresponding to healthy persons in a feature information database, and obtain a diagnostic analysis value corresponding to the target patient; The diagnostic analysis unit is used to compare the diagnostic analysis value corresponding to the target patient with the preset judgment threshold set, and determine whether the target patient suffers from a nervous system disease based on the comparison result: The result display unit is used to display the comparison result of the diagnosis and analysis unit to relevant personnel.

2. An artificial intelligence diagnostic system for nervous system diseases according to claim 1, Features: The storage results in the feature information library are a variety of relevant feature data of healthy people and confirmed patients from multiple medical institutions collected by the data acquisition unit, and a plurality of specified feature vectors are extracted from the various relevant feature data according to the feature extraction unit.

3. An artificial intelligence diagnostic system for nervous system diseases according to claim 1, Features: The diagnostic analysis of the target patients is as follows: StepX1, select a feature vector; From the feature vector corresponding to the target patient, obtain n corresponding quantized values ​​and mark them as Li, i=1, 2, ...n; Obtain n corresponding quantized values ​​from the healthy persons corresponding to the feature vector in the feature information database, and mark them as Ki, where i=1, 2, ...n; StepX2, through Calculate the feature score value D of the target patient corresponding to the feature vector; StepX3, calculate the feature score value of each feature vector through the method from StepX1 to StepX2, and mark the feature score value of each feature vector as Dj, j = 1, 2, ... m, m represents the number of all feature vectors; StepX4, through Calculate the diagnostic analysis value G of the target patient; In the formula, αj is the weight coefficient preset corresponding to Dj, and λ is the bias term.

4. An artificial intelligence diagnosis system for nervous system diseases according to claim 3, Features: in, The preset decision threshold set is G0∈[G1min,G2min,G3min], and G1min<G2min<G3min.

5. An artificial intelligence diagnosis system for nervous system diseases according to claim 4, Features: The diagnostic analysis units are compared as follows: If G < G1min, it is determined that the target patient does not have a corresponding neurological disease; If G1min≤G<G2min, the target patient is judged to be a mild patient of the corresponding nervous system disease; If G2min≤G<G3min, the target patient is judged as a moderate patient of the corresponding nervous system disease; If G≥G3min, the target patient is determined to be a severe patient of the corresponding neurological disease.

6. An artificial intelligence diagnosis system for nervous system diseases according to claim 3, Features: The data analysis unit is also used to perform diagnostic analysis on all confirmed patients in the feature information library according to the method of performing diagnostic analysis on the target patient, and obtain the diagnostic analysis values ​​corresponding to all confirmed patients.

7. An artificial intelligence diagnosis system for nervous system diseases according to claim 6, Features: Also includes: The data processing unit is used to comprehensively organize the diagnostic analysis values ​​corresponding to all confirmed patients and obtain a set of judgment thresholds.

8. An artificial intelligence diagnosis system for nervous system diseases according to claim 7, Features: The comprehensive arrangement is as follows: StepY1. First, according to the severity of the disease, the confirmed patients are divided into mild patients, moderate patients, and severe patients; Step Y2, taking all moderate patients as an example, marking the diagnostic analysis values ​​corresponding to all confirmed patients as Gt, t = 1, 2, ... e, e represents the number of all confirmed patients; Step Y3, according to the formula Calculate the standard deviation value B corresponding to Gt; Step Y4, then compare the standard deviation value B with the preset standard deviation threshold By: When B>By, the corresponding Gt values ​​are deleted in descending order according to |Gt-Gp| until B≤By is established; Among them, Gp is ​​the average value of m tyj, and the deleted tyj value does not participate in the subsequent calculation of Gp; Step Y5. When B≤By, select a diagnostic analysis value with the largest value and the smallest value from all Gt involved in calculating the corresponding B value, and form the diagnostic judgment interval corresponding to the moderate patient; Step Y6. According to the method of Step Y2 to Step Y5, the diagnostic judgment interval corresponding to the mild patients and the diagnostic judgment interval corresponding to the severe patients are calculated; Step Y7, then obtain the corresponding minimum value from the diagnostic judgment intervals corresponding to mild patients, moderate patients and severe patients respectively; Then the decision threshold set G0∈[G1min,G2min,G3min] is formed.

9. An artificial intelligence diagnosis system for nervous system diseases according to claim 8, Features: in, G1min is the minimum value of the diagnostic judgment interval corresponding to mild patients; G2min is the minimum value of the diagnostic judgment interval corresponding to moderate patients; G3min is the minimum value of the diagnostic judgment interval corresponding to severe patients.

10. An artificial intelligence diagnosis method for nervous system diseases, the method being implemented by an artificial intelligence diagnosis system for nervous system diseases according to any one of claims 1 to 9, It is characterized in that The method comprises the following steps: Step 1: Data Collection Collect various relevant characteristic data of target patients; Step 2: Feature extraction Extracting quantized values ​​corresponding to a plurality of specified feature vectors from a plurality of related feature data; Step 3: Build a feature information database Storing multiple designated feature vectors corresponding to healthy persons and confirmed patients respectively, wherein the storage result is obtained by collecting multiple relevant feature data of healthy persons and confirmed patients from multiple medical institutions, and extracting multiple designated feature vectors from the multiple relevant feature data according to a feature extraction step; Step 4: Data Analysis According to the result of feature extraction and in combination with multiple specified feature vectors corresponding to healthy persons in the feature information database, a diagnostic analysis is performed on the target patient to obtain a diagnostic analysis value corresponding to the target patient; Step 5: Data processing According to the method of performing diagnostic analysis on the target patient, all the confirmed patients in the feature information database are diagnosed and analyzed, and the diagnostic analysis values ​​corresponding to all the confirmed patients are obtained; then the diagnostic analysis values ​​corresponding to all the confirmed patients are comprehensively sorted out to obtain a determination threshold set; Step 6: Diagnostic Analysis Comparing the diagnostic analysis value corresponding to the target patient with a preset judgment threshold set, and determining whether the target patient suffers from a neurological disease based on the comparison result; Step 7: Results display Present the comparative results of the diagnostic analysis to relevant personnel.

Images