Disease automatic screening auxiliary system based on big data

Abstract

The invention relates to the technical field of disease screening assistance, and discloses an automatic disease screening assistance system based on big data, which comprises a data acquisition unit, a data preprocessing unit, a disease information base, a feature analysis unit and a result output unit, and multi-source comprehensive data is collected, so that one-sidedness is avoided. The data preprocessing unit is used for removing repeated data and standardizing age, symptom severity and inspection data, so that the data is more scientific and comparable, and the screening accuracy is improved; based on a disease information base, patient characteristic values are extracted for all diseases, probability scores are calculated, disease differences are considered, and accurate evaluation is achieved. Whether a patient suffers from illness or not and the degree are judged according to a score threshold value, illness state information better fitting the reality is provided for doctors and patients, and the defects of a traditional method are overcome. And the result output unit generates a report containing basic information, diseases, probability scores and disease degrees of the patients, so that doctors can diagnose and formulate schemes conveniently, the patients can keep and understand the schemes conveniently, and the health management participation degree of the patients is improved.

Description

Technical Field

[0001] The present invention relates to the field of disease screening auxiliary technology, and in particular to an automatic disease screening auxiliary system based on big data. Background Art

[0002] In the field of modern medicine, disease screening and diagnosis are crucial links. With the growth and aging of the population, medical needs are increasing, and traditional disease screening methods are facing huge challenges.

[0003] On the one hand, doctors need to consider a large amount of information when diagnosing a disease. This information includes the patient's basic information, symptom data, examination data, and imaging examination data. However, faced with such complex data, the efficiency and accuracy of doctors' manual analysis and diagnosis may be affected. Different doctors may have different interpretations of the data, and integrating these data from different data sources requires a lot of time and effort.

[0004] On the other hand, with the development of medical technology, the amount of data generated by various examination methods has exploded, forming a huge amount of medical data, namely big data. These big data contain rich information, but if they cannot be effectively used, they will become a burden for medical decision-making. In the process of disease screening, how to quickly and accurately extract valuable information from these massive and complex data to assist doctors in early detection and diagnosis of diseases has become an urgent problem to be solved.

[0005] In addition, the complexity and diversity of diseases also increase the difficulty of screening. Each disease has its own specific disease characteristics, which may manifest differently in different patients, and the relationship between disease characteristics is also relatively complex. Traditional screening methods often find it difficult to fully and accurately consider these factors, which may lead to missed diagnosis or misdiagnosis of the disease.

[0006] Therefore, in order to improve the efficiency and accuracy of disease screening and meet the growing medical needs, there is an urgent need for an automated disease screening auxiliary system based on big data. Summary of the invention

[0007] The purpose of the present invention is to provide an automatic disease screening auxiliary system based on big data, which solves the technical problems raised in the background technology.

[0008] The purpose of the present invention can be achieved through the following technical solutions: The disease automatic screening auxiliary system based on big data includes: A data collection unit, used to collect patient-related data from multiple data sources; wherein the related data includes patient basic information, symptom data, examination data and imaging examination data; A data preprocessing unit, used to preprocess the collected related data to obtain standardized related data through preprocessing; Disease information database, used to store various diseases and their corresponding disease characteristics; The feature analysis unit is used to extract the disease feature values ​​of all disease features in the corresponding disease from the pre-processed associated data based on each disease in the disease information database, and calculate the probability score of the corresponding disease based on the disease feature values, and then determine whether the patient has the corresponding disease and the degree of the corresponding disease based on the probability score; The result output unit is used to generate the corresponding disease and the corresponding disease degree of the patient determined by the feature analysis unit in the form of a report.

[0009] As a further solution of the present invention: basic information includes but is not limited to the patient's age, gender, and family medical history; Symptom data include, but are not limited to, symptom type, duration, and severity; Examination data include but are not limited to blood indicators and biochemical indicators; Imaging examination data include but are not limited to key descriptions and quantitative indicators in the corresponding reports of X-ray, CT, and MRI.

[0010] As a further solution of the present invention, the pretreatment method is as follows: StepA1, remove duplicate data; By comparing all the data fields in the associated data to see if they are completely equal, determine whether the data fields are completely identical. If there are at least two groups of completely identical data fields, only one group is retained; StepA2: Data standardization; Standardize different types of data; StepA2.1, standardize the age; Map age data to the interval [0,1]; Extract the predetermined age range [0,L max ], extract the patient's age value NL; Among them, L max Expressed as the maximum age value in the age range; Then through NL new =NL / L Max , get the standardized age value NL new , and NL new is in the interval [0,1]; StepA2.2, standardize the severity of symptoms; Among all symptoms, there is a corresponding severity scale pre-set; Among them, the severity scale corresponds to a measurement level of 0 to 5; StepA2.3, standardize the inspection data; Extracting various inspection index values ​​from the inspection data and extracting the corresponding preset index threshold intervals; Select an inspection indicator; The standardized inspection indicators are determined as follows: Z new =(ZZ min ) / (Z max -Z min ); In the formula, Z is the value detected by the inspection indicator, Z new is the standardized value of the inspection indicator, Z min is the minimum value in the corresponding preset indicator threshold range, Z max is the maximum value in the corresponding preset indicator threshold range.

[0011] As a further solution of the present invention, a corresponding severity scale is preset for all symptoms; wherein the severity scale corresponds to a measurement level of 0 to 5, and each level has a clear definition, as follows: Level 0: No relevant symptoms, completely normal; Level 1: Symptoms are extremely mild and hardly affect normal life and activities. They are only noticed with careful perception or under certain conditions. Grade 2: Mild symptoms that occasionally interfere with daily activities but do not affect overall function and quality of life, and no special treatment is required; Level 3: Symptoms are moderate and will have a certain impact on daily activities, such as reduced efficiency of some work or leisure activities, and some simple relief measures may be needed; Level 4: Symptoms are more severe, significantly interfering with daily activities and having a significant impact on quality of life, requiring appropriate countermeasures or medical intervention; Level 5: Symptoms are extremely severe, seriously affecting or even making it impossible to carry out normal life and activities, and professional medical treatment is urgently needed; Relevant medical staff fill in the measurement level based on the user's description of each symptom and the corresponding severity scale.

[0012] As a further solution of the present invention, the specific method of the feature analysis unit is as follows: Step G1. Select a disease in the disease information database and mark it as T. w , w=1, 2, …, e, w represents the number of diseases in the disease information database, and e represents the number of all diseases in the disease information database; Step G2: Extract all disease features corresponding to the disease, and then obtain the disease feature value corresponding to the patient based on all the corresponding disease features, and mark it as M i , i∈T w , which means that the disease characteristics corresponding to i are all disease characteristics within the wth disease; Step G3, based on all corresponding disease characteristics, extract the corresponding predetermined feature value interval [M i,min ,M i,max ], where M i,min It is expressed as the minimum eigenvalue of a disease feature when suffering from the corresponding disease, M i,max It is expressed as the maximum eigenvalue of a disease feature when suffering from the corresponding disease; Step G4, then pass , calculate the probability score S of the patient suffering from the wth disease w ; In the formula, α i A preset weight coefficient is set according to the corresponding disease feature i; Step G5: Extract the score threshold SY1 predetermined according to the corresponding disease w SY2 w SY3 w , and then the probability score S of the corresponding disease w The corresponding scoring threshold SY1 w SY2 w SY3 w , make comparisons, and determine the corresponding diseases the patient suffers from and the severity of the disease based on the comparison results.

[0013] As a further embodiment of the present invention, SY1 w >SY2 w >SY3 w .

[0014] As a further solution of the present invention, in Step G5, If S w <SY3 w , then the patient is judged not to have the corresponding disease; If SY2 w >S w ≥SY3 w , then the patient is judged to have the corresponding disease and the degree of the disease is mild; If SY1 w >S w ≥SY2 w , then the patient is judged to have a moderate degree of the corresponding disease; If S w ≥SY1 w, then the patient's corresponding disease is judged to be serious.

[0015] As a further solution of the present invention, the report content includes the patient's basic information, the disease he suffers from, the probability score and the severity of the disease.

[0016] Beneficial effects of the present invention: Comprehensive data utilization: Through the data acquisition unit, comprehensive related data including basic patient information, symptom data, examination data and imaging examination data are collected from multiple data sources, which can comprehensively consider multiple factors for disease screening, avoid the one-sidedness caused by a single data source, and greatly improve the accuracy of disease judgment.

[0017] Standardization processing advantages: The data preprocessing unit preprocesses the collected data, removes duplicate data, and reduces the interference of data redundancy on the analysis results. At the same time, different types of data such as age, symptom severity, and examination data are standardized. For example, age is standardized to the [0,1] interval, so that different age groups have a unified quantitative standard in disease assessment; symptom severity is quantified according to a preset measurement scale, and examination data is standardized through a specific formula. These standardized operations make the data more comparable and scientific in subsequent analysis, further improving the accuracy of screening.

[0018] Accuracy of feature analysis: Based on a huge disease information database, the feature analysis unit extracts disease feature values ​​from patient-related data for each disease and calculates probability scores based on preset weights. This personalized analysis method fully considers the differences in features of different diseases, and can more accurately reflect the possibility that a patient has a specific disease, avoiding the one-size-fits-all problem in traditional screening methods and providing patients with assessment results that are more in line with their actual conditions.

[0019] Reasonableness of degree determination: By comparing the probability score with the predetermined scoring threshold, it can accurately determine whether the patient is ill and the degree of illness (mild, moderate, severe). This not only allows doctors to understand the patient's condition more clearly, but also helps patients understand their own health risks, achieving refined and personalized disease screening.

[0020] Convenient result presentation: The result output unit generates a report with the patient's basic information, diseases, probability score and disease severity, which allows doctors to quickly obtain key information and provides a clear and intuitive basis for diagnosis and treatment plan formulation. It also makes it easier for patients to retain and understand their own health assessment results, helping to increase patients' participation in their own health management. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The present invention will be further described below in conjunction with the accompanying drawings.

[0022] Figure 1 It is a system block diagram of the automatic disease screening auxiliary system based on big data of the present invention.

[0023] Figure 2 It is a module flow diagram of the data preprocessing unit in the disease automatic screening auxiliary system based on big data of the present invention. DETAILED DESCRIPTION

[0024] 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.

[0025] See also Figure 1 and Figure 2 As shown, the present invention is an automatic disease screening auxiliary system based on big data, comprising: A data collection unit, used for collecting relevant data of the patient from multiple data sources; Among them, the associated data include basic patient information, symptom data, examination data, and imaging examination data; Basic information includes but is not limited to the patient's age, gender, and family medical history; for example, the incidence of certain diseases is higher in people of a certain age group, gender, or with a family genetic predisposition; Symptom data include but are not limited to symptom type, duration, and severity; for example, persistent fever and short-term fever may indicate different diseases, and severe headache and mild headache are also different in diagnostic significance; Examination data include but are not limited to blood indicators and biochemical indicators; for example, abnormal white blood cell count in the blood may indicate infectious diseases; Imaging examination data include but are not limited to key descriptions and quantitative indicators in the corresponding reports of X-ray, CT, and MRI; In this embodiment, the data source can be connected to the hospital information system, laboratory information system, etc. to obtain data, and manually input data is also accepted; This comprehensive data collection method provides rich materials for subsequent accurate disease screening. At the same time, it can be connected with hospital information systems and laboratory information systems and accept manually input data, which enhances the flexibility of data acquisition.

[0026] A data preprocessing unit, used for preprocessing the collected related data; The preprocessing method is as follows: StepA1, remove duplicate data; By comparing whether all data fields in the associated data are completely equal, it is determined whether the data fields are exactly the same. If there are at least two groups of completely identical data fields, only one group is retained. This processing method can effectively reduce data redundancy, improve data processing efficiency, and avoid analysis errors and resource waste caused by duplicate data.

[0027] StepA2: Data standardization; Standardize different types of data; StepA2.1, standardize the age; In this embodiment, the age data is mapped to the interval [0,1] so that the data of different age groups have a unified quantitative standard, which is convenient for comprehensively considering the influence of age factors on diseases in disease analysis; Extract the predetermined age range [0,L max ], in this embodiment, its L max The value of is 100 years old; Extract the patient's age value NL; Then through NL new =NL / L Max , get the standardized age value NL new , and NL new is in the interval [0,1]; StepA2.2, standardize the severity of symptoms; Among all symptoms, there is a corresponding severity scale pre-set; The severity scale corresponds to a measurement level of 0 to 5, and each level has a clear definition, for example: Level 0: No relevant symptoms, completely normal; Level 1: Symptoms are extremely mild and hardly affect normal life and activities. They are only noticed with careful perception or under certain conditions. Grade 2: Mild symptoms that occasionally interfere with daily activities but do not affect overall function and quality of life, and no special treatment is required; Level 3: Symptoms are moderate and will have a certain impact on daily activities, such as reduced efficiency of some work or leisure activities, and some simple relief measures may be needed; Level 4: Symptoms are more severe, significantly interfering with daily activities and having a significant impact on quality of life, requiring appropriate countermeasures or medical intervention; Level 5: Symptoms are extremely severe, seriously affecting or even making it impossible to carry out normal life and activities, and professional medical treatment is urgently needed; Combine the user's description of each symptom with the corresponding severity scale and fill in the measurement level; Quantifying the severity of different symptoms provides a unified measurement scale for symptom data in analysis, providing a basis for accurate disease diagnosis; In this embodiment, this standardized approach can uniformly quantify the severity of different symptoms, for example, a mild cough can correspond to a lower level, while severe chest pain can correspond to a higher level, providing standardized symptom data for subsequent analysis; StepA2.3: Standardize the inspection data Extracting various inspection index values ​​from the inspection data and extracting the corresponding preset index threshold intervals; Select an inspection indicator; The standardized inspection indicators are determined as follows: Z new =(ZZ min ) / (Z max -Z min ); In the formula, Z is the value detected by the inspection indicator, Z new is the standardized value of the inspection indicator, Z min is the minimum value in the corresponding preset indicator threshold range, Z max is the maximum value in the corresponding preset indicator threshold range; Disease information database, used to store various diseases and their corresponding disease characteristics; In this embodiment, the disease information database organizes a professional medical team to conduct in-depth research and sorting of a large number of authoritative medical documents, research reports, medical monographs, etc. These documents cover the research results of various medical fields around the world, including classic medical texts and the latest cutting-edge research. Information on various diseases is extracted from these materials, including disease names, symptoms, pathogenesis, pathological characteristics, etc. For example, for cardiovascular diseases, related disease characteristics such as hypertension, hyperlipidemia, and atherosclerosis are sorted out from the literature. For example, patients with hypertension may be accompanied by symptoms such as dizziness and headaches, and atherosclerosis is characterized by thickening of the blood vessel walls in imaging examinations, and these contents are integrated into the disease information database; At the same time, we have established cooperative relationships with major hospitals to collect massive amounts of clinical medical records data. These medical records contain information about patients of different ages, genders, and regions. We extract key information such as the patient's diagnosis, symptom description, examination data, and treatment process from the medical records. For each confirmed disease, we analyze the common characteristics of the patient group. For example, after collecting the medical records of diabetic patients, we found that patients generally have symptoms such as polydipsia, polyphagia, polyuria, and weight loss, as well as abnormal examination indicators such as blood sugar and glycosylated hemoglobin. We associate these characteristics with the disease of diabetes and enter them into the information database.

[0028] Experienced clinicians and medical experts are also invited to participate. They have accumulated rich diagnostic experience in long-term medical practice and have unique insights into some rare or complex diseases. Based on their clinical experience, experts provide disease characteristics that may not be fully reflected in literature and general medical records. For example, for some special types of autoimmune diseases, experts can point out subtle symptoms or special signs that may appear in the early stages of the disease. This information is sorted and included in the disease information database to further improve the description of disease characteristics.

[0029] In this embodiment, these disease characteristics are summarized through medical research and practice, and are the key basis for determining whether a patient has a certain disease; the establishment of a disease information database needs to be continuously updated and improved to adapt to new research results in the medical field and changes in disease types, thereby ensuring the scientificity and timeliness of disease screening.

[0030] The feature analysis unit is used to extract the disease feature values ​​of all disease features in the corresponding disease from the pre-processed associated data based on each disease in the disease information database, and calculate the probability score of the corresponding disease based on the disease feature values, and then determine whether the patient has the corresponding disease and the degree of the corresponding disease based on the probability score; The extraction method is as follows: Step G1. Select a disease in the disease information database and mark it as T. w , w=1, 2, …, e, w represents the number of diseases in the disease information database, and e represents the number of all diseases in the disease information database; Step G2: Extract all disease features corresponding to the disease, and then obtain the disease feature value corresponding to the patient based on all the corresponding disease features, and mark it as M i , i∈T w , which means that the disease characteristics corresponding to i are all disease characteristics within the wth disease; Step G3, based on all corresponding disease characteristics, extract the corresponding predetermined feature value interval [M i,min ,M i,max ], where M i,min It is expressed as the minimum eigenvalue of a disease feature when suffering from the corresponding disease, M i,max It is expressed as the maximum eigenvalue of a disease feature when suffering from the corresponding disease; Step G4, then pass , calculate the probability score S of the patient suffering from the wth disease w ; In the formula, α i A preset weight coefficient is set according to the corresponding disease feature i; Step G5: Extract the score threshold SY1 predetermined according to the corresponding diseasew SY2 w SY3 w , and then the probability score S of the corresponding disease w The corresponding scoring threshold SY1 w SY2 w SY3 w , for comparison: Among them, SY1 w >SY2 w >SY3 w ; If S w <SY3 w , then the patient is judged not to have the corresponding disease; If SY2 w >S w ≥SY3 w , then the patient is judged to have the corresponding disease and the degree of the disease is mild; If SY1 w >S w ≥SY2 w , then the patient is judged to have a moderate degree of the corresponding disease; If S w ≥SY1 w , then the patient's corresponding disease is judged to be severe; In this embodiment, the feature analysis unit extracts disease feature values ​​from the preprocessed associated data based on the disease information database, and calculates the disease probability score according to the preset weight coefficient; by comparing with the predetermined score threshold, it accurately determines whether the patient suffers from the corresponding disease and the extent of the disease. This quantitative analysis method improves the accuracy of disease screening.

[0031] A result output unit, used to generate the corresponding disease and the corresponding disease degree of the patient determined by the feature analysis unit in the form of a report; The report includes basic information about the patient, the disease they suffer from, probability score, and the extent of the disease; This embodiment generates a report in the form of a patient's basic information, the disease he suffers from, the probability score and the degree of the disease, with standardized and clear content, which is convenient for doctors and patients to intuitively understand the disease screening results and provide a favorable reference for subsequent diagnosis and treatment; Example 1: Cardiovascular disease screening for patients The first step is data collection; The data of a 55-year-old male patient was obtained from the hospital information system. His family history showed that his father had coronary heart disease. Symptom data showed that the patient occasionally had palpitations for about 3 months, and the severity was assessed as level 3 on a 0-5 scale. In the examination data, the blood pressure value was 140 / 90mmHg, and the total cholesterol value in the blood lipid index was 6.2mmol / L; Imaging (electrocardiogram) report showed mild ST-T changes; At the same time, the doctor manually entered the patient’s recent chest tightness after exercise; Step 2: Data preprocessing; Duplicate data removal: There is no duplicate data in this collection.

[0032] Data standardization: Age standardization: L max =100,NL=55,NL new =55 / 100=0.55.

[0033] Symptom severity was standardized: palpitations severity was recorded as grade 3.

[0034] Check data normalization: For systolic blood pressure, assuming that the normal range of systolic blood pressure is [90,120] mmHg, Z=140, Z min =90, Z max =120, Z new = (140-90) / (120-90) ≈ 1.67 (outside the normal range, the standardized value is higher); For total cholesterol, assuming the normal range is [2.8-5.2] mmol / L, Z = 6.2, Z min =2.8, Z max =5.2, Z new = (6.2-2.8) / (5.2-2.8)≈1.42; Step 3: Feature analysis; Disease selection and labeling: Select cardiovascular disease and label it as T1.

[0035] Disease feature value extraction: The age-standardized value M1=0.55, the severity of palpitations M2=3, the systolic blood pressure-standardized value M3=1.67, the total cholesterol-standardized value M4=1.42, and the electrocardiogram ST-T change characteristic value M5, which is set to 0.5 according to the preset quantification, because it is a slight change; Eigenvalue interval extraction: M 1,min =0.4, which is the minimum age characteristic of cardiovascular disease onset, M 1,max =0.7; M 2,min =2,M 2,max =6; M 3,min=1.2, which is the lower limit of the characteristic value of the assumed abnormal systolic blood pressure, M 3,max =2; M 4,min =1.2,M 4,max =2; M 5,min =0.3,M 5,max =0.8; Probability score calculation: Assume α1=0.2, α2=0.2, α3=0.3, α4=0.2, α5=0.1; S w =0.2×(0.55-0.4) / (0.7-0.4)+0.2×(3-2) / (6-2)+0.3×(1.67-1.2) / (2-1.2)+0.2×(1.42-1.2) / (2-1.2)+0.1×(0.5-0.3) / (0.8-0.3)≈0.62; Disease determination: Assuming SY11=0.7, SY21=0.5, SY31=0.3, 0.5<0.62<0.7, the patient is determined to have cardiovascular disease, the degree of which is moderate; Step 4: Output the results; Generate a report, which includes: Patient basic information: 55-year-old male, with family history of disease; Disease: cardiovascular disease; Probability score: 0.62; Disease severity: moderate; Example 2: Screening of patients for lung infection Step 1: Data collection; Data from a 30-year-old female patient were collected through the laboratory information system and hospital information system; The patient's basic information showed no special family medical history; Symptom data showed that the cough symptom lasted for 2 weeks, with a severity of 5 on a 0-5 scale, accompanied by fever with a maximum temperature of 38.5°C; In the examination data, the white blood cell count was 12×10 9 / L; The X-ray report during the imaging examination showed small patchy, blurred shadows in the lungs; Step 2: Data preprocessing; Duplicate data removal: no duplicate data; Data standardization: Age standardization: L max =100,NL=30,NL new =30 / 100=0.3; Symptom severity was standardized: cough severity level 5, fever temperature 38.5°C; Assuming the normal body temperature range is [36-37.2]℃, then Z=38.5, Z min =36, Z max =37.2, the normalized value of fever degree is about 0.69; Normalization of examination data: Assuming the normal range of white blood cell count is [4-10] × 10 9 / L,Z=12,Z min =4, Z max =10, the normalized value of white blood cell count is about 0.5; Step 3: Feature analysis; Disease selection and labeling: Select lung infection disease and label it as T2; Disease feature value extraction: The standardized value of age is M1=0.3, the cough severity is M2=5, the fever degree is M3=0.69, the white blood cell count is M4=0.5, and the lung shadow feature value M5 is set to 0.7 according to the preset quantification because there is a clear shadow; Eigenvalue interval extraction: M 1,min =0.2,M 1,max =0.5;M 2,min =3,M 2,max =7;M 3,min =0.5,M 3,max =1;M 4,min =0.4,M 4,max =0.8;M 5,min =0.5,M 5,max =1; Probability score calculation: Assume α1=0.1, α2=0.2, α3=0.2, α4=0.2, α5=0.3; S w =0.1×(0.3-0.2) / (0.5-0.2)+0.2×(5-3) / (7-3)+0.2×(0.69-0.5) / (1-0.5)+0.2×(0.5-0.4) / (0.8-0.4)+0.3×(0.7-0.5) / (1-0.5)≈0.57; Disease determination: Assuming SY12=0.7, SY22=0.5, SY32=0.3, 0.5<0.57<0.7, it is determined that the patient has a lung infection disease with a moderate degree; Step 4: Output the results; Generate a report, including: basic information of the patient: 30-year-old female; disease: lung infection; probability score: 0.57; disease severity: moderate.

[0036] 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.

[0037] 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. The disease automatic screening auxiliary system based on big data is characterized by: include: A data preprocessing unit, used to preprocess the patient's associated data to obtain standardized associated data, the associated data including the patient's basic information, symptom data, examination data and imaging examination data; The feature analysis unit is used to extract the disease feature values ​​of all disease features in the corresponding disease from the preprocessed associated data based on each disease in the disease information database, and calculate the probability score of the corresponding disease based on it, and then determine whether the patient has the corresponding disease and the degree to which he suffers from the corresponding disease based on the probability score.

2. The disease automatic screening auxiliary system based on big data according to claim 1 is characterized in that: The preprocessing method is as follows: StepA1, remove duplicate data; By comparing all the data fields in the associated data to see if they are completely equal, determine whether the data fields are completely identical. If there are at least two groups of completely identical data fields, only one group is retained; StepA2: Data standardization; Standardize different types of data; Standardize for age; Extract the predetermined age range [0,L max ], extract the patient's age value NL; Among them, L max Expressed as the maximum age value in the age range; Then through NL new =NL / L Max , get the standardized age value NL new ; Standardize inspection data; Extract the values ​​of various inspection indicators in the inspection data, and extract the corresponding preset indicator threshold ranges, and then obtain the standardized inspection indicators through the predetermined standardized inspection indicator calculation formula.

3. The disease automatic screening auxiliary system based on big data according to claim 2 is characterized in that: In the process of standardizing the inspection data, an inspection indicator is selected; The calculation formula for standardized inspection indicators is as follows: WITH new =(ZZ min ) / (WITH max -WITH min ); In the formula, Z is the value detected by the inspection indicator, Z new is the standardized value of the inspection indicator, Z min is the minimum value in the corresponding preset indicator threshold range, Z max is the maximum value in the corresponding preset indicator threshold range.

4. The disease automatic screening auxiliary system based on big data according to claim 1 is characterized in that: The specific method of the feature analysis unit is as follows: Step G1. Select a disease in the disease information database and mark it as T. w , w=1, 2, …, e, w represents the number of diseases in the disease information database, and e represents the number of all diseases in the disease information database; Step G2: Extract all disease features corresponding to the disease, and then obtain the disease feature value corresponding to the patient based on all the corresponding disease features, and mark it as M i , i∈T w , which means that the disease characteristics corresponding to i are all disease characteristics within the wth disease; Step G3, based on all corresponding disease characteristics, extract the corresponding predetermined feature value interval [M i,min ,M i,max ], where M i,min It is expressed as the minimum eigenvalue of a disease feature when suffering from the corresponding disease, M i,max It is expressed as the maximum eigenvalue of a disease feature when suffering from the corresponding disease; Step G4, then calculate the probability score of the patient suffering from the wth disease through the preset disease probability score formula; Step G5: Extract the score threshold SY1 predetermined according to the corresponding disease w SY2 w SY3 w ,Then the probability score of the corresponding disease is compared with the corresponding ,score threshold, and the corresponding disease and the severity of the patient are determined based on the ,comparison result.

5. The disease automatic screening auxiliary system based on big data according to claim 4 is characterized in that: In StepG4, the disease probability score formula is as follows: ； In the formula, S w is the probability score of the patient suffering from the wth disease, α i The preset weight coefficient is determined according to the corresponding disease feature i.

6. The disease automatic screening auxiliary system based on big data according to claim 5 is characterized in that: In StepG5, the scoring threshold includes SY1 w SY2 w SY3 w , and SY1 w >SY2 w >SY3 w .

7. The disease automatic screening auxiliary system based on big data according to claim 6 is characterized in that: In StepG5, If S w <SY3 w , then the patient is judged not to have the corresponding disease; If SY2 w >S w ≥SY3 w , then the patient is judged to have the corresponding disease and the degree of the disease is mild; If SY1 w >S w ≥SY2 w , then the patient is judged to have a moderate degree of the corresponding disease; If S w ≥SY1 w , then the patient's corresponding disease is judged to be serious.

8. The disease automatic screening auxiliary system based on big data according to claim 1 is characterized in that: Also includes: A data collection unit, used for collecting relevant data of the patient from multiple data sources; Disease information database, used to store various diseases and their corresponding disease characteristics; The result output unit is used to generate the corresponding disease and the corresponding disease degree of the patient determined by the feature analysis unit in the form of a report.

9. The disease automatic screening auxiliary system based on big data according to claim 8 is characterized in that: Basic information includes but is not limited to the patient's age, gender, and family medical history; Symptom data include, but are not limited to, symptom type and duration; Examination data include but are not limited to blood indicators and biochemical indicators; Imaging examination data include but are not limited to key descriptions and quantitative indicators in the corresponding reports of X-ray, CT, and MRI.

10. The disease automatic screening auxiliary system based on big data according to claim 8 is characterized in that: The report includes the patient's basic information, the disease he suffers from, the probability score and the extent of the disease.

Images