Automatic hospital guide system and method for chemical examination

Abstract

The invention relates to the technical field of automatic hospital guide, and discloses an automatic hospital guide system and method for laboratory examination, and the method comprises the steps: building a symptom-symptom association rule and department diagnosis and treatment range mapping, and endowing a weight to the typicality of the symptom according to the symptom; converting the patient self-description into standard medical terms, and generating a symptom set; comparing the coincidence degrees of the patient symptoms and the disease database, if the patient symptoms are completely matched with the disease database, triggering a primary selection strategy, and directly locking candidate diseases; if not, executing a maximum matching strategy, and triggering inquiry verification in a descending order according to the symptom overlap ratio; inquiring unmatched symptoms of the candidate diseases, and confirming or eliminating the diseases according to patient feedback; and preferentially recommending departments covering compound diseases, and performing parallel recommendation when multiple departments are matched. The system continuously optimizes the model through patient feedback to form a closed-loop learning mechanism. According to the method, knowledge base quantitative analysis and dynamic inquiry verification are fused, and a patient is helped to be guided to complete laboratory examination in a correct diagnosis and treatment department.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic diagnosis guidance, and in particular to an automatic diagnosis guidance system and method for laboratory tests. Background Art

[0002] In the field of medical health, laboratory tests are an important medical means. To ensure that the laboratory tests performed by patients match their own conditions, patients need to obtain laboratory test items from doctors in matching diagnosis and treatment departments. Therefore, a guidance method is needed to recommend diagnosis and treatment departments that suit patients' conditions.

[0003] As a key tool for improving medical efficiency and optimizing resource allocation, the core goal of the automatic medical guidance system is to quickly match patients to appropriate diagnosis and treatment departments through intelligent analysis of patient symptoms and signs. However, existing medical guidance technologies still face multiple challenges in practical applications, especially in terms of the accuracy of symptom analysis and the intelligence of department matching strategies.

[0004] Traditional guidance methods usually rely on the subjective judgment of doctors or nurses and provide guidance based on the patient's symptom description. However, this method has certain limitations. Factors such as differences in patients' ability to express themselves, the experience level of medical staff, and the degree of standardization of the guidance process will affect the accuracy and efficiency of the guidance.

[0005] Existing automated medical guidance systems are usually based on symptom matching, matching the symptom information entered by the patient with the disease symptom library in the database to recommend possible diseases and treatment departments; however, such systems mainly rely on simple keyword matching and lack in-depth analysis of the complex correlations between symptoms.

[0006] In summary, the existing automatic guidance methods for laboratory tests have problems such as insufficient accuracy in symptom matching and a lack of comprehensive analysis in the process of symptom speculation and in the recommendation of diagnosis and treatment departments. These problems affect the overall performance of the automatic guidance system, causing patients to receive inaccurate recommendation results, affecting their medical experience and diagnostic efficiency. Therefore, a more accurate and efficient automatic guidance method is needed to make up for the shortcomings of existing technologies. Summary of the Invention

[0007] The present invention provides an automatic guidance system and method for laboratory tests, which facilitate solving the problems mentioned in the above background technology.

[0008] The present invention provides the following technical solution: an automatic diagnosis guidance method for laboratory tests, comprising:

[0009] Establish a matching model;

[0010] For each disease, a disease set named after the disease is established, and the symptoms corresponding to each disease are recorded in the set;

[0011] Obtain patient symptom information and enter patient symptom set;

[0012] Obtain the number of elements in each symptom set that are identical to the patient's symptom set, and record it as the identical amount of each symptom set;

[0013] The primary selection strategy and the maximum matching strategy are selected based on the same quantity, and the diagnosis and treatment department is recommended to the patient based on the execution results.

[0014] Optionally, establishing a matching model includes:

[0015] By collecting information about the symptoms corresponding to each symptom, a symptom-symptom relationship database is formed;

[0016] Mark the disease corresponding to each symptom and form the association rules between symptoms and diseases;

[0017] Collect and organize the data on the range of diseases diagnosed and treated by each diagnosis and treatment department, and obtain the types of diseases that each diagnosis and treatment department can diagnose and treat;

[0018] Combine the data of symptoms, diseases and diagnosis and treatment departments, associate the symptoms and diseases with the corresponding diagnosis and treatment departments, and establish a symptom-based diagnosis and treatment department matching model;

[0019] In the matching model, the symptoms corresponding to each disease are obtained, and the symptoms corresponding to each disease are weighted according to the typicality of the symptoms among the diseases.

[0020] Optionally, obtaining patient symptom information and entering a patient symptom set includes:

[0021] Obtain information about patient symptoms;

[0022] A patient symptom set is established for the patient, and the patient's symptom information is entered into the patient symptom set.

[0023] Optionally, the method of selecting and executing a preliminary selection strategy and a maximum matching strategy based on the same quantity, and recommending a diagnosis and treatment department to the patient based on the execution results, includes:

[0024] Get the number of elements in the patient's symptom set and record it as the total number of symptoms;

[0025] If there is a set of symptoms with the same quantity equal to the total quantity of symptoms, the symptoms corresponding to the set of symptoms are marked as primary selected symptoms, and the primary selection strategy is executed;

[0026] If there is no set of symptoms with the same quantity equal to the total quantity of symptoms, the maximum quantity matching strategy is executed;

[0027] The primary selection strategy is specifically as follows:

[0028] Get the number of primary selected diseases;

[0029] If the number of pre-selected symptoms is equal to one, the pre-selected symptoms are marked as selected symptoms, and the consultation strategy is executed, including:

[0030] Obtain the elements in the symptom set corresponding to the selected symptom that are different from the patient's symptom set, use them as the consultation symptoms, and execute the consultation strategy to obtain the patient's selection of the consultation symptoms;

[0031] If the patient's choice of the symptom in the inquiry is the first choice, the currently selected disease will be regarded as the presumed disease;

[0032] If the patient's choice of the symptom in the inquiry is the second choice, the currently selected symptom will be deleted and the maximum matching strategy will be implemented;

[0033] If the number of preliminarily selected symptoms is greater than one, then obtain the elements in the symptom set corresponding to the preliminarily selected symptoms that are identical to the patient's symptom set, obtain the weight of each element in the symptom set, calculate the sum of the weights, and record it as the weight value of each preliminarily selected symptom;

[0034] Comparing the weight values of each of the preliminarily selected symptoms, sorting the preliminarily selected symptoms in descending order according to the weight values, and recording the sorting as the preferred sorting;

[0035] Select each of the primary selected symptoms as a selected symptom in turn according to the order of preference;

[0036] After a primary symptom is selected as a selected symptom, the elements in the symptom set corresponding to the selected symptom that are different from the patient's symptom set are obtained, which are used as the consultation symptom, and the consultation strategy is executed to obtain the patient's selection of the consultation symptom;

[0037] If the patient's choice of the symptom in the inquiry is the first choice, the currently selected disease will be regarded as the presumed disease;

[0038] If the patient's choice of the symptom in the consultation is the second choice, the currently selected symptom will be deleted from the preferred ranking, and the primary selected symptom will be selected as the new selected symptom in the preferred ranking until the inferred symptom is obtained. When all the selected symptoms in the preferred ranking are deleted and no inferred symptom is obtained, the maximum matching strategy will be executed.

[0039] Optionally, the execution of the maximum matching strategy is specifically as follows:

[0040] Sort the symptom sets in descending order according to the size of each symptom set, and record the sorting as matching sorting;

[0041] Selecting each symptom set as a pre-selected symptom set in sequence according to the matching ranking;

[0042] After selecting a preselected symptom set, obtain the elements in the preselected symptom set that are different from the patient set, use them as the consultation symptoms, and execute the consultation strategy to obtain the patient's selection of the consultation symptoms;

[0043] If the patient's choice of the symptom in the inquiry is the first choice, the symptom corresponding to the current pre-selected symptom set is used as the presumed symptom, and a double screening strategy is implemented;

[0044] If the patient's selection of the symptom in the inquiry is the second choice, the current pre-selected symptom set is deleted from the matching sorting, and a symptom set is selected in the matching sorting as a new pre-selected symptom set until the inferred symptom is obtained.

[0045] Optionally, the double screening strategy is performed as follows:

[0046] Obtain the symptom set corresponding to the presumed symptom, and obtain the elements in the patient's symptom set that are different from the symptom set, and enter them into a set, recorded as a double set;

[0047] Compare the binary set with the disease set in the matching set to determine whether there is a disease set in the matching model that contains all elements in the binary set;

[0048] If it does not exist, then the double screening strategy ends;

[0049] If it exists, obtain the existing symptom set containing all elements in the dual set and record it as the screening symptom set;

[0050] If there is only one screening symptom set, obtain the elements in the screening symptom set that are different from those in the dual set, use them as the consultation symptoms, and execute the consultation strategy to obtain the patient's choice of consultation symptoms. If the patient's choice of the consultation symptoms is the first choice, then use the symptoms corresponding to the screening symptom set as the second inferred symptom. If the patient's choice of the consultation symptoms is the second choice, then end the dual screening strategy.

[0051] If the number of existing screening symptom sets is greater than one, then obtain the elements in each screening symptom set that are identical to the elements in the double set, obtain the weight of each element in the screening symptom set, calculate the sum of the weights, and record it as the sum of the screening weights in each screening symptom set;

[0052] Comparing the sum of the screening weights in each screening symptom set, sorting the screening symptom set in descending order according to the size of the screening weight value, and recording the sorting as the screening sorting;

[0053] Select each screening symptom set in turn according to the order of screening sorting, and after each screening symptom set is selected, obtain the elements in the screening symptom set that are different from those in the dual set, use them as the consultation symptoms, and execute the consultation strategy to obtain the patient's selection of the consultation symptoms;

[0054] If the patient's selection of the symptom in the medical consultation is the first choice, the symptom corresponding to the currently selected screening symptom set is used as the second presumed symptom;

[0055] If the patient's selection of the symptom in the medical consultation is the second choice, the currently selected screening symptom set is deleted from the screening order, and the first screening symptom set is reselected in the screening order until the second presumed symptom is obtained, or all screening symptom sets in the screening order are deleted.

[0056] Optionally, executing the consultation strategy to obtain the patient's selection of consultation symptoms includes:

[0057] Present the medical symptoms to the patient and obtain the patient's choice of the medical symptoms, including the first choice and the second choice;

[0058] The first option is: the patient determines that there are symptoms in the questionnaire that match his or her own symptoms;

[0059] The second option is that the patient determines that none of the symptoms in the medical consultation are consistent with his or her own.

[0060] Optionally, recommending a diagnosis and treatment department to the patient based on the matching result with the matching model includes:

[0061] Obtaining the matched inferred symptom and the second inferred symptom;

[0062] If only the presumed symptom or the second presumed symptom exists, the diagnosis and treatment department corresponding to the presumed symptom or the second presumed symptom is obtained from the matching model as the diagnosis and treatment department recommended to the patient;

[0063] If the presumed symptom and the second presumed symptom exist at the same time, the diagnosis and treatment department corresponding to both the presumed symptom and the second presumed symptom is obtained in the matching model as the diagnosis and treatment department recommended to the patient;

[0064] If there is no diagnosis and treatment department corresponding to both the presumed symptom and the second presumed symptom in the matching model, then the two diagnosis and treatment departments corresponding to the presumed symptom and the second presumed symptom are obtained respectively as the diagnosis and treatment departments recommended to the patient;

[0065] When the patient completes the consultation, obtain the patient's evaluation of the consultation and update the data based on the evaluation, including:

[0066] The patient's evaluation of this guided consultation specifically includes: whether the recommended diagnosis and treatment department meets the needs, whether the recommended diagnosis and treatment department does not meet the needs;

[0067] When the patient's evaluation of this medical consultation is that the recommended diagnosis and treatment department meets the needs, the patient's corresponding symptom information and the recommended diagnosis and treatment department are associated and recorded in the matching model.

[0068] A system using the automatic laboratory test guidance method comprises:

[0069] Information collection module: used to collect patients' self-reported information;

[0070] Data matching module: used to match diseases and treatment departments according to patient symptoms;

[0071] Data storage module: used to store matching model data;

[0072] Information interaction module: used to display information to patients and obtain their choices and evaluations of the information.

[0073] The present invention has the following beneficial effects:

[0074] 1. By collecting the disease information corresponding to each symptom and forming a symptom-symptom relationship library, multi-dimensional data integration can be achieved to ensure that the system has comprehensive coverage and high reliability when constructing the knowledge base; marking the corresponding diseases for various symptoms and forming association rules between symptoms and diseases can clarify the characteristics of each disease and improve the accuracy of subsequent symptom matching; collecting and organizing the diagnosis and treatment disease range data of each diagnosis and treatment department can help to accurately define the professional diagnosis and treatment boundaries of different departments, ensure that effective docking between departments and diseases can be achieved when making recommendations, and reduce the inconvenience caused to patients due to mismatched diagnosis departments; combining symptoms, diseases and diagnosis and treatment department data to establish a symptom-based diagnosis and treatment department matching model can utilize the mutual verification of multi-dimensional data to enhance the scientificity and practicality of the model, thereby improving the reliability of diagnosis and treatment recommendations in the inference process; obtaining the symptoms corresponding to each disease in the matching model and weighting them according to their typicality in the disease can achieve quantitative analysis of symptoms, highlight the impact of key symptoms, and optimize the disease screening process.

[0075] 2. By first quantifying the number of elements in the patient's symptom set to determine the total number of symptoms, an objective basis is provided for subsequent matching, so that a scientific judgment can be made between the preliminary selection strategy and the maximum matching strategy; when there is a match in the symptom set that is completely consistent with the total number of symptoms, the system can quickly mark it as a preliminary symptom and directly execute the preliminary selection strategy, thereby greatly improving the matching efficiency; conversely, in the absence of a complete match, the maximum matching strategy is adopted to ensure that when the symptom information is incomplete or there is ambiguity, the most likely symptom can still be screened out according to the maximum matching principle; in the preliminary selection strategy, when the number of preliminary selected symptoms is only one, by obtaining the elements in the symptom set corresponding to the symptom that are different from the patient's symptom set and executing the consultation strategy, both the patient's feedback can be used to confirm the symptom and the deviation caused by single data judgment can be avoided; if the preliminary symptom is If the number of symptoms is greater than one, the weights of the elements in each preliminarily selected symptom that are the same as the patient's symptoms are evaluated, and the weight sum is calculated and sorted to ensure that more typical and more critical symptoms are given priority, thereby achieving the preferred sorting of symptoms. This weight sorting mechanism can objectively reflect the degree of match between each symptom and the patient's symptoms; in the sorting process, the preliminarily selected symptoms are selected as the selected symptoms in turn and the patient's choice of consultation symptoms is confirmed again through the consultation strategy. This not only allows the system to fully consider the patient's subjective feelings, but also can promptly eliminate mismatches when the patient's feedback does not meet expectations, and dynamically adjust the inference direction; overall, this multi-level, multi-strategy design not only improves the accuracy of the recommendation of diagnosis and treatment departments, but also achieves the complementary effect of rapid screening and detailed consultation, ensuring that the diagnosis and treatment departments finally recommended to patients meet their needs.

[0076] 3. Improve the accuracy of symptom matching through a maximum matching strategy, and achieve dynamic optimization based on patient feedback. This strategy first sorts the symptom set in descending order according to the number of matching symptoms, ensuring that the symptom set closest to the patient's symptoms is prioritized, thereby improving the success rate of matching, reducing unnecessary calculations and inquiries, and improving system efficiency. Secondly, the strategy adopts a step-by-step screening and dynamic adjustment method, starting with the most matching symptom set, to reduce the interference of low-matching symptoms, making the screening process more scientific and reasonable. During the specific implementation process, the strategy obtains elements in the pre-selected symptom set that are different from the patient's symptom set and sets them as question symptoms for further verification. The accuracy of symptom matching can be further improved by confirming additional symptoms, avoiding blind matching based solely on the initial matching degree. If the patient's choice of the question symptom is the first choice, that is, they agree that the question symptom is consistent with their own situation, then it indicates that the pre-selected symptom set is more consistent with the patient's situation. At this time, the symptoms corresponding to the set can be determined as presumed symptoms, and a double screening strategy is further implemented to ensure the accuracy of the diagnosis from multiple angles. If the patient's choice for the symptom in the consultation is the second choice, it indicates that the matching degree of the symptom set is not ideal. The system will automatically eliminate the current pre-selected symptom set and continue to select the next possible symptom set from the matching ranking until the symptom that best matches the patient's symptoms is inferred.

[0077] 4. Through the double screening strategy, when the disease that completely covers the patient's main symptoms is not matched, the symptom information that may be missed during the initial matching process is compensated; when the patient's self-reported symptoms cannot be completely classified into a single disease, the system reorganizes these symptoms that are not included in the initial matching disease into a new set, namely the double set, to take into account the situation that the patient's symptoms may correspond to two different diseases; this strategy can not only avoid the risk of missed diagnosis or misdiagnosis due to incomplete coverage of a single disease, but also screen out possible second presumed diseases from candidate diseases through further questioning and weight evaluation; through this double-layer screening mechanism, the system can more comprehensively capture the patient's actual symptom characteristics, dynamically adjust the matching strategy, and ensure that even when the symptoms are more complex or there is overlap, it can take into account multiple possibilities; in addition, the double screening strategy can The additional consultation phase obtains the patient's feedback on the symptoms that are not covered by the initial match, making the inference process more rigorous and personalized; this method not only improves the overall diagnostic accuracy of the system, but also, by comprehensively considering all the patient's symptoms, to a certain extent enhances the patient's trust in the automatic guidance system; at the same time, the strategy can objectively compare the matching degree of different disease candidates through quantitative weighting and sorting processing, ensuring that the final recommended second inferred disease is more in line with the patient's actual situation; therefore, when the double screening strategy fails to fully match all the patient's main symptoms, it effectively reduces the risk of omissions caused by insufficient symptom coverage by constructing a new symptom set and introducing dynamic consultation feedback, and achieves accurate capture of possible second diseases, thereby ensuring that the guidance based on the patient's main symptom information meets the patient's needs.

[0078] 5. Recommending medical departments to patients based on the matching results with the matching model. This recommendation method fully utilizes the inferred symptom data of the matching model to achieve accurate matching and personalized recommendations of medical departments. First, after the system obtains the inferred symptom and the second inferred symptom, it extracts the corresponding medical departments from the matching model according to the different situations of only one symptom or both. This differentiated processing method ensures that when the symptom matching results are relatively clear, a highly professional department can be directly recommended. In the case of coexistence of two symptoms, the medical department that can cover both symptoms is given priority, thereby improving the comprehensiveness and pertinence of diagnosis and treatment. If there is no diagnosis and treatment department corresponding to both diseases in the matching model, the corresponding departments will be obtained separately to provide patients with multiple treatment paths, reducing the treatment deviation caused by the inability of a single department to cover all symptoms; secondly, after the patient completes the consultation, the system will actively obtain the patient's feedback information on the guidance recommendation, and promptly feed back the evaluation of "meets the needs" or "does not meet the needs" to the matching model to achieve dynamic data updating and self-optimization; this closed-loop feedback mechanism based on the patient's actual medical experience not only improves the system's learning ability, but also can gradually improve the accuracy and rationality of future diagnosis and treatment department recommendations with the support of continuously accumulated real data. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 It is a basic flow chart of the present invention. DETAILED DESCRIPTION

[0080] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0081] Example 1, refer to Figure 1 , an automatic guidance method for laboratory examination, comprising:

[0082] Establish a matching model;

[0083] For each disease, a disease set named after the disease is established, and the symptoms corresponding to each disease are recorded in the set;

[0084] Obtain patient symptom information and enter patient symptom set;

[0085] Obtain the number of elements in each symptom set that are identical to the patient's symptom set, and record it as the identical amount of each symptom set;

[0086] The primary selection strategy and the maximum matching strategy are selected based on the same quantity, and the diagnosis and treatment department is recommended to the patient based on the execution results.

[0087] The establishing of the matching model comprises:

[0088] By collecting information about the symptoms corresponding to each symptom, a symptom-symptom relationship database is formed;

[0089] Mark the disease corresponding to each symptom and form the association rules between symptoms and diseases;

[0090] Collect and organize the data on the range of diseases diagnosed and treated by each diagnosis and treatment department, and obtain the types of diseases that each diagnosis and treatment department can diagnose and treat;

[0091] Combine the data of symptoms, diseases and diagnosis and treatment departments, associate the symptoms and diseases with the corresponding diagnosis and treatment departments, and establish a symptom-based diagnosis and treatment department matching model;

[0092] In the matching model, the symptoms corresponding to each disease are obtained, and the corresponding symptoms of each disease are weighted according to their typicality among the diseases. By integrating the multi-dimensional data relationship library of symptoms, diseases, and departments, the comprehensiveness of the medical knowledge system is covered, ensuring the reliability of the model; by clarifying the association rules between symptoms and diseases, the core characteristics of each disease are directly defined, improving the accuracy of subsequent matching; combining the department's diagnosis and treatment scope data, the professional boundaries of each department are accurately delineated to avoid patients from repeating treatment due to incorrect department selection; through symptom weight assessment, the contribution of key symptoms to diseases is quantified, and the model's screening logic for complex symptoms is optimized.

[0093] The step of obtaining patient symptom information and entering a patient symptom set includes:

[0094] Obtain patient symptom information, specifically:

[0095] Obtain patient self-report information through patient self-report;

[0096] The self-reported information specifically includes: the patient's self-reported discomfort symptoms;

[0097] Through natural language processing technology, the system identifies and analyzes the dialects and colloquial expressions in patients' oral symptom information, and converts the patients' oral symptom information into standard Mandarin.

[0098] Perform semantic analysis on the symptom information converted into standard Mandarin, extract the words and sentences containing symptom expressions, and record them as key words and sentences;

[0099] Identify and analyze the meaning of key words and convert them into standard medical terms as patient symptom information;

[0100] A patient symptom set is established for the patient, and the patient's symptom information is entered into the patient symptom set.

[0101] The method of selecting and executing the preliminary selection strategy and the maximum matching strategy based on the same quantity, and recommending a diagnosis and treatment department to the patient based on the execution results, includes:

[0102] Get the number of elements in the patient's symptom set and record it as the total number of symptoms;

[0103] If there is a set of symptoms with the same quantity equal to the total quantity of symptoms, the symptoms corresponding to the set of symptoms are marked as primary selected symptoms, and the primary selection strategy is executed;

[0104] If there is no set of symptoms with the same quantity as the total number of symptoms, the maximum matching strategy is executed; through quantitative judgment of the total number of symptoms, the preliminary strategy (quick locking when fully matched) or the maximum matching strategy (downgrading when ambiguous symptoms) is dynamically selected, taking into account both efficiency and fault tolerance; in the preliminary strategy, if there is a single preliminary symptom, misjudgment is avoided by reverse verification of unmatched symptoms (questioning strategy); if there are multiple preliminary symptoms, highly typical symptoms are verified first based on weight and ranking to ensure that the matching priority is consistent with medical logic.

[0105] The primary selection strategy is specifically as follows:

[0106] Get the number of primary selected diseases;

[0107] If the number of pre-selected symptoms is equal to one, the pre-selected symptoms are marked as selected symptoms, and the consultation strategy is executed, including:

[0108] Obtain the elements in the symptom set corresponding to the selected symptom that are different from the patient's symptom set, use them as the consultation symptoms, and execute the consultation strategy to obtain the patient's selection of the consultation symptoms;

[0109] If the patient's choice of the symptom in the inquiry is the first choice, the currently selected disease will be regarded as the presumed disease;

[0110] If the patient's choice of the symptom in the inquiry is the second choice, the currently selected symptom will be deleted and the maximum matching strategy will be implemented;

[0111] If the number of preliminarily selected symptoms is greater than one, then obtain the elements in the symptom set corresponding to the preliminarily selected symptoms that are identical to the patient's symptom set, obtain the weight of each element in the symptom set, calculate the sum of the weights, and record it as the weight value of each preliminarily selected symptom;

[0112] Comparing the weight values of each of the preliminarily selected symptoms, sorting the preliminarily selected symptoms in descending order according to the weight values, and recording the sorting as the preferred sorting;

[0113] Select each of the primary selected symptoms as a selected symptom in turn according to the order of preference;

[0114] After a primary symptom is selected as a selected symptom, the elements in the symptom set corresponding to the selected symptom that are different from the patient's symptom set are obtained, which are used as the consultation symptom, and the consultation strategy is executed to obtain the patient's selection of the consultation symptom;

[0115] If the patient's choice of the symptom in the inquiry is the first choice, the currently selected disease will be regarded as the presumed disease;

[0116] If the patient's choice of the symptom in the consultation is the second choice, the currently selected symptom will be deleted from the preferred ranking, and the primary selected symptom will be selected as the new selected symptom in the preferred ranking until the inferred symptom is obtained. When all the selected symptoms in the preferred ranking are deleted and no inferred symptom is obtained, the maximum matching strategy will be executed.

[0117] The execution of the maximum matching strategy is specifically as follows:

[0118] Sort the symptom sets in descending order according to the size of each symptom set, and record the sorting as matching sorting;

[0119] Selecting each symptom set as a pre-selected symptom set in sequence according to the matching ranking;

[0120] After selecting a preselected symptom set, obtain the elements in the preselected symptom set that are different from the patient set, use them as the consultation symptoms, and execute the consultation strategy to obtain the patient's selection of the consultation symptoms;

[0121] If the patient's choice of the symptom in the inquiry is the first choice, the symptom corresponding to the current pre-selected symptom set is used as the presumed symptom, and a double screening strategy is implemented;

[0122] If the patient's second choice is the symptom in the inquiry, the current pre-selected symptom set is deleted from the matching sort, and the symptom set in the matching sort is selected as the new pre-selected symptom set until the inferred symptom is obtained. By sorting the matching symptom set in descending order, high-probability symptoms are verified first, reducing invalid calculations; by dynamically eliminating low-matching items (when the patient's feedback is the second choice), the optimal solution is quickly converged; combined with a double screening strategy, potential symptoms are further explored to avoid missed diagnoses due to scattered symptoms.

[0123] The double screening strategy is specifically as follows:

[0124] Obtain the symptom set corresponding to the presumed symptom, and obtain the elements in the patient's symptom set that are different from the symptom set, and enter them into a set, recorded as a double set;

[0125] Compare the binary set with the disease set in the matching set to determine whether there is a disease set in the matching model that contains all elements in the binary set;

[0126] If it does not exist, then the double screening strategy ends;

[0127] If it exists, obtain the existing symptom set containing all elements in the dual set and record it as the screening symptom set;

[0128] If there is only one screening symptom set, obtain the elements in the screening symptom set that are different from those in the dual set, use them as the consultation symptoms, and execute the consultation strategy to obtain the patient's choice of consultation symptoms. If the patient's choice of the consultation symptoms is the first choice, then use the symptoms corresponding to the screening symptom set as the second inferred symptom. If the patient's choice of the consultation symptoms is the second choice, then end the dual screening strategy.

[0129] If the number of existing screening symptom sets is greater than one, then obtain the elements in each screening symptom set that are identical to the elements in the double set, obtain the weight of each element in the screening symptom set, calculate the sum of the weights, and record it as the sum of the screening weights in each screening symptom set;

[0130] Comparing the sum of the screening weights in each screening symptom set, sorting the screening symptom set in descending order according to the size of the screening weight value, and recording the sorting as the screening sorting;

[0131] Select each screening symptom set in turn according to the order of screening sorting, and after each screening symptom set is selected, obtain the elements in the screening symptom set that are different from those in the dual set, use them as the consultation symptoms, and execute the consultation strategy to obtain the patient's selection of the consultation symptoms;

[0132] If the patient's selection of the symptom in the medical consultation is the first choice, the symptom corresponding to the currently selected screening symptom set is used as the second presumed symptom;

[0133] If the patient's selection of a symptom is the second choice, the currently selected screening symptom set is removed from the screening order, and the first screening symptom set is reselected from the screening order until the second presumed symptom is obtained, or all screening symptom sets in the screening order are deleted. By constructing a double set, symptoms not covered by the initial match are re-incorporated into the analysis to address the possibility of complex symptoms; candidate symptoms are screened through weighted sorting to ensure the typicality of the second presumed symptom; combined with dynamic consultation feedback, the system's inference and the patient's subjective feelings are balanced to enhance the rationality of the guidance results.

[0134] The execution of the inquiry strategy to obtain the patient's selection of inquiry symptoms includes:

[0135] Present the medical symptoms to the patient and obtain the patient's choice of the medical symptoms, including the first choice and the second choice;

[0136] The first option is: the patient determines that there are symptoms in the questionnaire that match his or her own symptoms;

[0137] The second option is that the patient determines that none of the symptoms in the medical consultation are consistent with his or her own.

[0138] The method of recommending a diagnosis and treatment department to the patient based on the matching result with the matching model includes:

[0139] Obtaining the matched inferred symptom and the second inferred symptom;

[0140] If only the presumed symptom or the second presumed symptom exists, the diagnosis and treatment department corresponding to the presumed symptom or the second presumed symptom is obtained from the matching model as the diagnosis and treatment department recommended to the patient;

[0141] If the presumed symptom and the second presumed symptom exist at the same time, the diagnosis and treatment department corresponding to both the presumed symptom and the second presumed symptom is obtained in the matching model as the diagnosis and treatment department recommended to the patient;

[0142] If there is no diagnosis and treatment department corresponding to both the presumed symptom and the second presumed symptom in the matching model, then the two diagnosis and treatment departments corresponding to the presumed symptom and the second presumed symptom are obtained respectively as the diagnosis and treatment departments recommended to the patient;

[0143] When the patient completes the consultation, obtain the patient's evaluation of the consultation and update the data based on the evaluation, including:

[0144] The patient's evaluation of this guided consultation specifically includes: whether the recommended diagnosis and treatment department meets the needs, whether the recommended diagnosis and treatment department does not meet the needs;

[0145] When a patient's evaluation of the recommended medical department meets their needs, the patient's symptom information is linked to the recommended medical department and recorded in the matching model. Flexible department recommendations are made based on single or complex symptoms, with priority given to comprehensive departments covering multiple conditions. When departments cannot be covered simultaneously, alternative pathways are provided to reduce the risk of misleading recommendations. A closed-loop optimization model based on patient feedback continuously improves the accuracy of recommendations.

[0146] Embodiment 2, a system for implementing the automatic diagnosis guidance method for laboratory tests, comprising:

[0147] Information collection module: used to collect the patient's height and weight data, as well as the patient's front and side views;

[0148] Data matching module: used to match diseases and treatment departments according to patient symptoms;

[0149] Data storage module: used to store inference model data and matching model data;

[0150] Information interaction module: used to display information to patients and obtain their choices and evaluations of the information.

[0151] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0152] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for automatic diagnosis guidance for laboratory tests, characterized in that: include: Establish a matching model; For each disease, a disease set named after the disease is established, and the symptoms corresponding to each disease are recorded in the set; Obtain patient symptom information and enter patient symptom set; Obtain the number of elements in each symptom set that are identical to the patient's symptom set, and record it as the identical amount of each symptom set; The primary selection strategy and the maximum matching strategy are selected based on the same quantity, and the diagnosis and treatment department is recommended to the patient based on the execution results.

2. The automatic diagnosis guidance method for laboratory tests according to claim 1, characterized in that: The establishing of the matching model comprises: By collecting information about the symptoms corresponding to each symptom, a symptom-symptom relationship database is formed; Mark the disease corresponding to each symptom and form the association rules between symptoms and diseases; Collect and organize the data on the range of diseases diagnosed and treated by each diagnosis and treatment department, and obtain the types of diseases that each diagnosis and treatment department can diagnose and treat; Combine the data of symptoms, diseases and diagnosis and treatment departments, associate the symptoms and diseases with the corresponding diagnosis and treatment departments, and establish a symptom-based diagnosis and treatment department matching model; In the matching model, the symptoms corresponding to each disease are obtained, and the symptoms corresponding to each disease are weighted according to the typicality of the symptoms among the diseases.

3. The automatic diagnosis guidance method for laboratory tests according to claim 1, characterized in that: The step of obtaining patient symptom information and entering a patient symptom set includes: Obtain information about patient symptoms; A patient symptom set is established for the patient, and the patient's symptom information is entered into the patient symptom set.

4. The automatic diagnosis guidance method for laboratory tests according to claim 1, characterized in that: The method of selecting and executing the preliminary selection strategy and the maximum matching strategy based on the same quantity, and recommending a diagnosis and treatment department to the patient based on the execution results, includes: Get the number of elements in the patient's symptom set and record it as the total number of symptoms; If there is a set of symptoms with the same quantity equal to the total quantity of symptoms, the symptoms corresponding to the set of symptoms are marked as primary selected symptoms, and the primary selection strategy is executed; If there is no set of symptoms with the same quantity equal to the total quantity of symptoms, the maximum quantity matching strategy is executed; The primary selection strategy is specifically as follows: Get the number of primary selected diseases; If the number of pre-selected symptoms is equal to one, the pre-selected symptoms are marked as selected symptoms, and the consultation strategy is executed, including: Obtain the elements in the symptom set corresponding to the selected symptom that are different from the patient's symptom set, use them as the consultation symptoms, and execute the consultation strategy to obtain the patient's selection of the consultation symptoms; If the patient's choice of the symptom in the inquiry is the first choice, the currently selected disease will be regarded as the presumed disease; If the patient's choice of the symptom in the inquiry is the second choice, the currently selected symptom will be deleted and the maximum matching strategy will be implemented; If the number of preliminarily selected symptoms is greater than one, then obtain the elements in the symptom set corresponding to the preliminarily selected symptoms that are identical to the patient's symptom set, obtain the weight of each element in the symptom set, calculate the sum of the weights, and record it as the weight value of each preliminarily selected symptom; Comparing the weight values of each of the preliminarily selected symptoms, sorting the preliminarily selected symptoms in descending order according to the weight values, and recording the sorting as the preferred sorting; Select each of the primary selected symptoms as a selected symptom in turn according to the order of preference; After a primary symptom is selected as a selected symptom, the elements in the symptom set corresponding to the selected symptom that are different from the patient's symptom set are obtained, which are used as the consultation symptom, and the consultation strategy is executed to obtain the patient's selection of the consultation symptom; If the patient's choice of the symptom in the inquiry is the first choice, the currently selected disease will be regarded as the presumed disease; If the patient's choice of the symptom in the consultation is the second choice, the currently selected symptom will be deleted from the preferred ranking, and the primary selected symptom will be selected as the new selected symptom in the preferred ranking until the inferred symptom is obtained. When all the selected symptoms in the preferred ranking are deleted and no inferred symptom is obtained, the maximum matching strategy will be executed.

5. The automatic diagnosis guidance method for laboratory tests according to claim 1, characterized in that: The execution of the maximum matching strategy is specifically as follows: Sort the symptom sets in descending order according to the size of each symptom set, and record the sorting as matching sorting; Selecting each symptom set as a pre-selected symptom set in sequence according to the matching ranking; After selecting a preselected symptom set, obtain the elements in the preselected symptom set that are different from the patient set, use them as the consultation symptoms, and execute the consultation strategy to obtain the patient's selection of the consultation symptoms; If the patient's choice of the symptom in the inquiry is the first choice, the symptom corresponding to the current pre-selected symptom set is used as the presumed symptom, and a double screening strategy is implemented; If the patient's selection of the symptom in the inquiry is the second choice, the current pre-selected symptom set is deleted from the matching sorting, and a symptom set is selected in the matching sorting as a new pre-selected symptom set until the inferred symptom is obtained.

6. The automatic diagnosis guidance method for laboratory tests according to claim 5, characterized in that: The double screening strategy is specifically as follows: Obtain the symptom set corresponding to the presumed symptom, and obtain the elements in the patient's symptom set that are different from the symptom set, and enter them into a set, recorded as a double set; Compare the binary set with the disease set in the matching set to determine whether there is a disease set in the matching model that contains all elements in the binary set; If it does not exist, then the double screening strategy ends; If it exists, obtain the existing symptom set containing all elements in the dual set and record it as the screening symptom set; If there is only one screening symptom set, obtain the elements in the screening symptom set that are different from those in the dual set, use them as the consultation symptoms, and execute the consultation strategy to obtain the patient's choice of consultation symptoms. If the patient's choice of the consultation symptoms is the first choice, then use the symptoms corresponding to the screening symptom set as the second inferred symptom. If the patient's choice of the consultation symptoms is the second choice, then end the dual screening strategy. If the number of existing screening symptom sets is greater than one, then obtain the elements in each screening symptom set that are identical to the elements in the double set, obtain the weight of each element in the screening symptom set, calculate the sum of the weights, and record it as the sum of the screening weights in each screening symptom set; Comparing the sum of the screening weights in each screening symptom set, sorting the screening symptom set in descending order according to the size of the screening weight value, and recording the sorting as the screening sorting; Select each screening symptom set in turn according to the order of screening sorting, and after each screening symptom set is selected, obtain the elements in the screening symptom set that are different from those in the dual set, use them as the consultation symptoms, and execute the consultation strategy to obtain the patient's selection of the consultation symptoms; If the patient's selection of the symptom in the medical consultation is the first choice, the symptom corresponding to the currently selected screening symptom set is used as the second presumed symptom; If the patient's selection of the symptom in the medical consultation is the second choice, the currently selected screening symptom set is deleted from the screening order, and the first screening symptom set is reselected in the screening order until the second presumed symptom is obtained, or all screening symptom sets in the screening order are deleted.

7. An automatic diagnosis guidance method for laboratory tests according to any one of claims 4 to 6, characterized in that: The execution of the inquiry strategy to obtain the patient's selection of inquiry symptoms includes: Present the medical symptoms to the patient and obtain the patient's choice of the medical symptoms, including the first choice and the second choice; The first option is: the patient determines that there are symptoms in the questionnaire that match his or her own symptoms; The second option is that the patient determines that none of the symptoms in the medical consultation are consistent with his or her own.

8. The automatic diagnosis guidance method for laboratory tests according to claim 1, characterized in that: The method of recommending a diagnosis and treatment department to the patient based on the matching result with the matching model includes: Obtaining the matched inferred symptom and the second inferred symptom; If only the presumed symptom or the second presumed symptom exists, the diagnosis and treatment department corresponding to the presumed symptom or the second presumed symptom is obtained from the matching model as the diagnosis and treatment department recommended to the patient; If the presumed symptom and the second presumed symptom exist at the same time, the diagnosis and treatment department corresponding to both the presumed symptom and the second presumed symptom is obtained in the matching model as the diagnosis and treatment department recommended to the patient; If there is no diagnosis and treatment department corresponding to both the presumed symptom and the second presumed symptom in the matching model, then the two diagnosis and treatment departments corresponding to the presumed symptom and the second presumed symptom are obtained respectively as the diagnosis and treatment departments recommended to the patient; When the patient completes the consultation, obtain the patient's evaluation of the consultation and update the data based on the evaluation, including: The patient's evaluation of this guided consultation specifically includes: whether the recommended diagnosis and treatment department meets the needs, whether the recommended diagnosis and treatment department does not meet the needs; When the patient's evaluation of this medical consultation is that the recommended diagnosis and treatment department meets the needs, the patient's corresponding symptom information and the recommended diagnosis and treatment department are associated and recorded in the matching model.

9. A system using the automatic guidance method for laboratory tests according to claim 1, characterized in that: include: Information collection module: used to collect patients' self-reported information; Data matching module: used to match diseases and treatment departments according to patient symptoms; Data storage module: used to store matching model data; Information interaction module: used to display information to patients and obtain their choices and evaluations of the information.

Images