Pain identification method, related method, electronic equipment and storage medium

Abstract

The invention discloses a pain recognition method, a related method, electronic equipment and a storage medium, and the method comprises the steps: obtaining the configuration data of a plurality of preset pain types, and obtaining the medical data related to a pain part of a target patient; wherein the configuration data of the preset pain type comprises a pain configuration condition and a score calculation mode of the preset pain type; based on a matching result between the medical data and the pain configuration condition of each preset pain type, selecting the preset pain type as a candidate pain type; calculating index values of the candidate pain types based on the medical data and the configuration data of the candidate pain types; and selecting the candidate pain type as a target pain type of the target patient based on the index value of each candidate pain type. According to the scheme, the objectivity and accuracy of pain type identification can be improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent medical technology, and in particular to a pain recognition method and related methods, electronic equipment and storage medium. Background Art

[0002] With the continuous advancement of intelligent medical care, the importance of pain detection in medical diagnosis is becoming increasingly important. Accurate pain detection provides a key basis for clinical treatment and helps improve the quality of medical services.

[0003] Existing pain detection and classification techniques typically rely on medical professionals analyzing patients' subjective descriptions and single-dimensional examination data. However, subjective descriptions are susceptible to individual differences, and medicine often presents the same disease with different symptoms, leading to biased analysis and affecting the accuracy of pain detection. Therefore, improving the objectivity and accuracy of pain classification has become an urgent issue. Summary of the Invention

[0004] The main technical problem solved by this application is to provide a pain identification method and related methods, electronic equipment and storage medium, which can improve the objectivity and accuracy of pain type identification.

[0005] In order to solve the above-mentioned technical problems, the first aspect of the present application provides a pain identification method, comprising: obtaining configuration data of several preset pain types, and obtaining medical data related to the pain site of a target patient; wherein the configuration data of the preset pain type includes the pain configuration conditions and score calculation method of the preset pain type; based on the matching results between the medical data and the pain configuration conditions of each preset pain type, selecting the preset pain type as the candidate pain type; based on the medical data and the configuration data of the candidate pain type, calculating the index value of the candidate pain type; based on the index value of each candidate pain type, selecting the candidate pain type as the target pain type of the target patient.

[0006] In order to solve the above technical problems, the second aspect of the present application provides an index value prediction method, including: obtaining an index value prediction model based on sample index values ​​of several sample patients for the same preset pain type and sample medical data of preset types of sample patients; wherein the sample index value is calculated based on the pain recognition method of the first aspect mentioned above; based on the index value prediction model, the target medical data of the preset type of the patient to be tested is predicted to obtain the predicted index value of the preset pain type of the patient to be tested.

[0007] In order to solve the above technical problems, the third aspect of the present application provides a drug efficacy response method, including: constructing sample pairs based on the first historical index value of each sample patient of the same preset pain type who did not use the preset drug, the second historical index value after using the preset drug for a preset period, and the preset period; wherein the first historical index value and the second historical index value are calculated based on the pain recognition method of the first aspect mentioned above; based on several sample pairs, a drug efficacy response model is obtained; based on the drug efficacy response model, the current index value of the preset pain type of the patient to be tested and the usage data of the preset drug are predicted to obtain the drug efficacy response trend of the patient to be tested using the preset drug; wherein the current index value is calculated based on the pain recognition method of the first aspect mentioned above.

[0008] In order to solve the above technical problems, the fourth aspect of the present application provides a disease prediction method, including: obtaining a disease prediction model based on the sample index values ​​and sample symptoms of each sample patient of the same preset pain type; wherein the sample index value is calculated based on the pain recognition method of the above-mentioned first aspect; based on the disease prediction model, predicting the current index value of the patient to be tested to obtain a disease prediction result of the patient to be tested; wherein the current index value is calculated based on the pain recognition method of the above-mentioned first aspect.

[0009] In order to solve the above technical problems, the fifth aspect of the present application provides an electronic device, including a memory and a processor coupled to each other, wherein the memory stores at least program instructions, and the processor is used to execute the program instructions to implement the pain recognition method in the above first aspect, or, to implement the indicator value prediction method in the above second aspect, or, to implement the drug efficacy response method in the above third aspect, or, to implement the disease prediction method in the above fourth aspect.

[0010] In order to solve the above technical problems, the sixth aspect of the present application provides a computer-readable storage medium storing program instructions that can be executed by a processor, wherein the program instructions are used to implement the pain recognition method of the first aspect, or to implement the indicator value prediction method of the second aspect, or to implement the drug efficacy response method of the third aspect, or to implement the disease prediction method of the fourth aspect.

[0011] The above scheme obtains configuration data for a number of preset pain types and medical data related to the pain location of a target patient. The configuration data for the preset pain types includes pain configuration conditions for the preset pain types and a score calculation method. Based on the matching results between the medical data and the pain configuration conditions of each preset pain type, the preset pain type is selected as a candidate pain type. Based on the medical data and the configuration data for the candidate pain types, an index value for the candidate pain type is calculated. Based on the index value of each candidate pain type, the candidate pain type is selected as the target pain type for the target patient. On the one hand, screening the candidate pain types based on the matching results between the pain configuration conditions of each preset pain type and the medical data of the target patient helps to screen out preset pain types that match the cause of the target patient while taking into account "same disease, different symptoms, different diseases, same symptoms" and select them as candidate pain types for further analysis. On the other hand, calculating the index value for each candidate pain type based on the score calculation formula can convert the medical data of the target patient into an objective numerical value for the candidate pain type. The calculation of the index value can explain the mechanism of pain occurrence under the cause corresponding to the candidate pain type, thereby improving the accuracy of the target pain type of the target patient determined based on the index value of each candidate pain type. Therefore, the objectivity and accuracy of pain type identification can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 This is a flowchart of an embodiment of the pain recognition method of the present application; Figure 2 This is a flowchart of another embodiment of the pain recognition method of the present application; Figure 3 This is a flow chart of an embodiment of the index value prediction method of the present application; Figure 4 This is a flow chart of an embodiment of the drug efficacy response method of the present application; Figure 5 This is a flow chart of an embodiment of the disease prediction method of the present application; Figure 6 This is a schematic diagram of the framework of an embodiment of the pain recognition device of the present application; Figure 7 This is a schematic diagram of the framework of an embodiment of the index value prediction device of the present application; Figure 8 This is a schematic diagram of the framework of an embodiment of the drug efficacy response device of the present application; Figure 9 This is a schematic diagram of the framework of an embodiment of the disease prediction device of the present application; Figure 10 This is a schematic diagram of the framework of an embodiment of the electronic device of the present application; Figure 11 It is a schematic diagram of a framework of an embodiment of the computer-readable storage medium of the present application. DETAILED DESCRIPTION

[0013] The following describes the embodiments of the present application in detail with reference to the accompanying drawings.

[0014] In the following description, for the purpose of explanation rather than limitation, specific details such as specific system structures, interfaces, and technologies are provided to facilitate a thorough understanding of the present application.

[0015] The terms "system" and "network" are often used interchangeably in this document. The term "and / or" is simply a description of an association between related objects, indicating that three possible relationships exist. For example, "A and / or B" can mean: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the fragment " / " generally indicates that the related objects are in an "or" relationship. Furthermore, "multiple" in this document refers to two or more than two.

[0016] See also Figure 1 , Figure 1 This is a flowchart of an embodiment of the pain recognition method of the present application. Specifically, it may include the following steps: Step S11: Acquire configuration data of several preset pain types, and acquire medical data related to the pain site of the target patient.

[0017] In the disclosed embodiments, medical data include but are not limited to physiological indicators of the painful area, past medical history, pain description, etc. Specifically, medical data is collected and integrated through an electronic medical record system, medical equipment or other data sources. For example, imaging data of the painful area is collected with the help of medical imaging equipment, and wearable devices are used to monitor the patient's heart rate, galvanic skin response and other physiological indicators in real time. The physiological indicators of the painful area may include heart rate, blood pressure, body temperature, etc. These indicators can reflect the patient's physiological reactions in a painful state. Past medical history records the patient's past disease conditions, treatment process and effects, which helps to identify which diseases the pain may be related to. The pain description is the patient's subjective description of his or her own pain experience, including the nature, degree, duration, etc. of the pain.

[0018] In one implementation scenario, the configuration data of the preset pain type includes the pain configuration conditions and score calculation method of the preset pain type. Specifically, the pain configuration conditions involve multiple dimensions such as the location, nature, duration, and accompanying symptoms of pain, and specific judgment criteria and thresholds are set under each dimension. The score calculation method is based on these configuration conditions to quantitatively score the collected medical data, and through comprehensive calculation, a total score that can reflect the degree and type of pain is obtained. Therefore, the indicator value not only helps to quickly and accurately identify the pain type of the target patient, but also provides a strong basis for the subsequent formulation of treatment plans to ensure the accuracy and effectiveness of pain identification.

[0019] In one implementation scenario, the configuration data for several preset pain types includes pain-requisite conditions, pain-exclusion conditions, and pain-support conditions corresponding to each preset pain type. Pain-requisite conditions represent the conditions that must be met under the preset pain type and are the minimum threshold for establishing a diagnosis. Pain-exclusion conditions represent the contradictory conditions of the preset pain type and set a safety boundary for diagnosis. Pain-support conditions supplement the pain-requisite conditions and are used to construct a probability gradient for diagnosis, providing relevant evidence for further confirming the pain type. Specifically, pain-requisite conditions cover typical characteristics of pain types, such as specific pain locations, duration, and nature, while pain-exclusion conditions are used to exclude symptoms that do not match the preset pain type, such as the absence of certain accompanying symptoms. Pain-support conditions include some atypical symptoms or signs that are helpful in confirming the pain type and can provide additional supporting information to enhance the accuracy of identification.

[0020] In a specific implementation scenario, configuration data for a number of preset pain types may be retrieved based on a preset medical database. The preset medical database stores configuration data corresponding to a number of preset pain types. For example, see Table 1 below:

[0021]

[0022] Table 1 It should be noted that the above-mentioned preset several pain type configuration data only represent one possible implementation method, and this application does not limit other implementation methods.

[0023] In a specific implementation scenario, the pain configuration conditions include pain requirements. For example, a preset pain type may require that pain be located in the lower back, last for more than one hour, and be accompanied by lower back stiffness. Only when these conditions are met will the pain type be considered as a possible identification result.

[0024] In another specific implementation scenario, the pain configuration conditions include pain-required conditions and pain-exclusion conditions. The pain-required conditions are used to preliminarily determine the pain type, and the pain-exclusion conditions are used to further screen and exclude non-compliant pain types. For example, if a preset pain type requires that the pain must be paroxysmal, but the patient describes persistent pain, then this pain type can be excluded. The above scheme can more accurately identify the patient's pain type and avoid misdiagnosis and missed diagnosis.

[0025] Step S12: Based on the matching results between the medical data and the pain configuration conditions of each preset pain type, the preset pain type is selected as a candidate pain type.

[0026] In one implementation scenario, the medical data provided by the patient, such as pain location, duration, and nature, is compared in detail with the pre-set pain configuration conditions for various pain types. For example, if the target patient's magnetic resonance imaging shows a 5mm lumbar disc herniation, it is considered that the pain conditions for "mechanical back pain" are met, and "mechanical back pain" is selected as a candidate pain type. The above scheme can preliminarily screen out preset pain types that match the patient's description as candidate pain types, providing effective data support for subsequent medical decision-making.

[0027] In one implementation scenario, the pain configuration condition includes a necessary pain condition, a first matching result between the medical data and the necessary pain condition is obtained, and in response to the first matching result indicating that the medical data meets the necessary pain condition, a preset pain type is selected as a candidate pain type.

[0028] In another implementation scenario, the pain configuration condition includes a necessary pain condition and a pain exclusion condition. A first matching result is obtained between the medical data and the necessary pain condition. In response to the first matching result indicating that the medical data meets the necessary pain condition, a second matching result is obtained between the medical data and the pain exclusion condition. In response to the second matching result indicating that the medical data does not trigger the pain exclusion condition, a preset pain type is selected as a candidate pain type. For example, if a patient's medical data shows that the pain is located in the knee, lasts for more than one month, and is accompanied by swelling, this may meet the necessary pain condition of "knee arthritis". However, if further analysis of the medical data reveals that the patient has a recent history of severe trauma, this may trigger the exclusion condition of "traumatic pain", thereby excluding "knee arthritis" as a candidate pain type. The above scheme, through such double verification, adopts the progressive matching logic of "necessary condition-exclusion condition" to quickly screen and filter a large number of preset pain types, thereby improving the efficiency of pain type candidate screening.

[0029] In a specific implementation scenario, first medical data of a certain type of target patient is matched with the pain configuration conditions of each of the preset pain types. When a preset pain type meets the necessary pain conditions, the pain exclusion conditions are further matched. If the first medical data successfully matches the pain exclusion conditions, the above-mentioned preset pain type is selected as the candidate pain type. If the first medical data fails to match the pain exclusion conditions, the matching can be continued based on other types of data in the target patient's medical data. If the match is successful, the above-mentioned preset pain type is selected as the candidate pain type. If the matching still fails, the second medical data in the target patient's medical data that is different from the first medical data is used as the new first medical data to match the necessary pain conditions of other preset pain types. For example, based on the magnetic resonance imaging of the target patient showing a 5mm lumbar disc herniation, it is considered that the necessary pain conditions of "mechanical back pain" are met, but it is found that the target patient has decreased muscle strength and a normal electromyogram, which triggers the pain exclusion conditions of "mechanical back pain". In the subsequent matching process, the "psychogenic pain" branch is transferred to. The specific subsequent matching process can refer to the specific steps of the aforementioned embodiment. For the sake of brevity, it will not be repeated here.

[0030] Step S13: Based on the medical data and the configuration data of the candidate pain type, an index value of the candidate pain type is calculated.

[0031] In one implementation scenario, the index value reflects the correlation and possibility between the candidate pain type and the target patient's pain condition. Specifically, the higher the index value, the more closely the candidate pain type matches the target patient's pain condition.

[0032] In one implementation scenario, a pain support condition that matches a score calculation method for a candidate pain type is determined based on medical data consistent with the data type of the pain support condition, first data corresponding to the pain support condition is obtained, and an index value of the candidate pain type is calculated based on the score calculation method and the first data corresponding to the pain support condition.

[0033] In a specific implementation scenario, the score calculation configuration information corresponding to each candidate pain type is retrieved from the knowledge base, which includes a set of pain support conditions and a weight distribution model. The pain support conditions include diagnostic support factors such as symptom duration, laboratory indicators, and imaging parameters. The weight distribution model sets a diagnostic contribution weight for each condition. Then, the original medical data of the target patients from electronic medical records, imaging materials and other channels are cleaned. Through structured extraction, natural language processing, computer vision and other technologies, unstructured data such as text and images are converted into standardized data that matches the pain support conditions to form a first data set. According to the preset threshold, interval, logical and other scoring functions, a single score is calculated for each support condition, and then the single score is multiplied by the corresponding weight and accumulated to obtain the index value of the candidate pain type. For example, based on the glycated hemoglobin of 9.6%, the nerve conduction velocity of 12m / s, the normal value of 50-60m / s, and the delay value of 38m / s in the medical data, the index value for diabetic peripheral neuropathy is 9.6×0.4+38m / s×0.6=26.64. The multimodal data in the target patient's medical data are fused to obtain an objective index value. The above solution processes the target patient's multimodal medical data and performs a one-to-one mapping between the processed medical data and the corresponding supporting conditions. The calculated index values ​​can explain the pain mechanism under the corresponding etiology of the candidate pain type, thereby improving the accuracy of the target patient's target pain type determined based on the index values ​​of each candidate pain type. This improves the objectivity and accuracy of pain type identification.

[0034] In a specific implementation scenario, the configuration data for the preset pain type also includes a pain exclusion condition. After calculating an index value for the candidate pain type based on the score calculation method and the first data corresponding to the pain support condition, in response to the medical data including second data matching the pain exclusion condition, the second data is weighted based on the target weight corresponding to the pain exclusion condition to obtain a conflicting item score. An updated index value is obtained based on the difference between the index value for the candidate pain type and the conflicting item score. For example, if a patient complains of "radiating pain in the lower limbs" but has a normal electromyography, the conflicting item score for the normal electromyography is deducted when calculating the index value corresponding to nerve root compression pain.

[0035] Step S14: Based on the index values ​​of the candidate pain types, select the candidate pain type as the target pain type of the target patient.

[0036] In one implementation scenario, after determining the updated index value, the index values ​​of all candidate pain types are compared, and the candidate pain type with the highest index value will be selected as the target pain type for the target patient. For example, if a cancer patient has both: an index value of 76 for cancer bone pain and an index value of 68 for chemotherapy drug neuralgia, cancer bone pain will be selected as the target pain type and given priority. The above scheme comprehensively considers the multimodal medical data of the target patient, including the patient's physiological indicators, medical history, symptoms and other information, so it can more comprehensively reflect the patient's pain condition and convert the target patient's medical data into objective numerical values ​​for the candidate pain type. The calculation of the index value can explain the mechanism of pain occurrence under the corresponding cause of the candidate pain type, thereby improving the accuracy of the target pain type of the target patient determined based on the index value of each candidate pain type.

[0037] In one implementation scenario, a first confidence threshold and a second confidence threshold are obtained for each candidate pain type. The first confidence threshold is greater than the second confidence threshold. The first confidence threshold serves as a high-confidence diagnostic standard, representing the strict indicator threshold required for a confirmed diagnosis. The second confidence threshold serves as a preliminary warning standard for identifying situations with potential disease risks. It should be noted that the numerical values ​​of the first confidence threshold and the second confidence threshold for different types of candidate pain types may be consistent or inconsistent, and the specific numerical values ​​are not limited in this application.

[0038] In one specific implementation scenario, in response to the indicator value of a candidate pain type being no less than a first confidence threshold, the candidate pain type is determined as a target pain type, and a first prompt is generated to indicate treatment. Specifically, the first prompt includes a standardized treatment plan recommendation, such as medication dosage, physical therapy method, and follow-up visit frequency. It can also trigger an electronic medical record system to record the diagnosis and complete the patient's medical record.

[0039] In another specific implementation scenario, in response to the indicator value of the candidate pain type being less than the first confidence threshold and not less than the second confidence threshold, a second prompt is generated to indicate further examination. Specifically, the second prompt is based on the clinical evidence-based medicine knowledge base, and accurately recommends suitable supplementary examination items for the current suspected pain type. For example, if the candidate pain type points to "rheumatoid arthritis" and the existing indicator value does not meet the diagnosis standard, the second prompt may include specific test items such as joint ultrasound examination, as well as recommended examination priorities and time nodes. For another example, the second prompt is accompanied by a description of the potential development risks of related diseases, and provides patients with a popular science explanation of the necessity of the examination. In addition, the second prompt can also be linked to the hospital's examination appointment system to provide a convenient examination item appointment portal and precautions, thereby improving the timeliness and consistency of medical services.

[0040] In another specific implementation scenario, in response to the index value of the candidate pain type being less than the second confidence threshold, the candidate pain type is excluded. That is, when the index value of the candidate pain type is lower than the second confidence threshold, it means that the current medical data has an extremely low degree of match with the pain type and is insufficient to support any diagnostic hypothesis. The pain type is removed from the candidate list, and the reason for exclusion can also be recorded in the diagnostic log, including specific index value and threshold comparison data, for subsequent optimization.

[0041] In one implementation scenario, after selecting a candidate pain type as the target pain type of the target patient based on the indicator values ​​of each candidate pain type, the current indicator value and historical indicator values ​​of the target pain type of the target patient of the same type can be obtained to generate pain trends and monitor the effectiveness of treatment. For example, the historical indicator value of a fibromyalgia patient before treatment was 15.2, and the current indicator value after three months of treatment was 9.3, indicating that the treatment plan was effective. The first indicator value of the current target pain type of the target patient within a preset time period and the second indicator values ​​of several historical target pain types can also be used to judge the influence relationship between different pain types and provide reference information for the subsequent diagnosis and treatment of the target patient.

[0042] The above scheme obtains configuration data for a number of preset pain types and medical data related to the pain location of a target patient. The configuration data for the preset pain types includes pain configuration conditions for the preset pain types and a score calculation method. Based on the matching results between the medical data and the pain configuration conditions of each preset pain type, the preset pain type is selected as a candidate pain type. Based on the medical data and the configuration data for the candidate pain types, an index value for the candidate pain type is calculated. Based on the index value of each candidate pain type, the candidate pain type is selected as the target pain type for the target patient. On the one hand, screening the candidate pain types based on the matching results between the pain configuration conditions of each preset pain type and the medical data of the target patient helps to screen out preset pain types that match the cause of the target patient while taking into account "same disease, different symptoms, different diseases, same symptoms" and select them as candidate pain types for further analysis. On the other hand, calculating the index value for each candidate pain type based on the score calculation formula can convert the medical data of the target patient into an objective numerical value for the candidate pain type. The calculation of the index value can explain the mechanism of pain occurrence under the cause corresponding to the candidate pain type, thereby improving the accuracy of the target pain type of the target patient determined based on the index value of each candidate pain type. Therefore, the objectivity and accuracy of pain type identification can be improved.

[0043] See also Figure 2 , Figure 2 This is a flow chart of another embodiment of the pain recognition method of the present application. Specifically, the following steps may be included: Step S11: Acquire configuration data of several preset pain types, and acquire medical data related to the pain site of the target patient.

[0044] For details, please refer to the specific steps of the aforementioned embodiment, which will not be repeated here for the sake of brevity.

[0045] Step S121: Determine whether the medical data of the target patient meets the necessary pain conditions.

[0046] For details, please refer to the specific steps of the aforementioned embodiment, which will not be repeated here for the sake of brevity.

[0047] Step S122: In response to the target patient's medical data not satisfying the necessary pain conditions, the current preset pain type is excluded.

[0048] For details, please refer to the specific steps of the aforementioned embodiment, which will not be repeated here for the sake of brevity.

[0049] Step S123: In response to the medical data of the target patient satisfying the necessary pain condition, determining whether the medical data of the target patient triggers a pain determination condition.

[0050] For details, please refer to the specific steps of the aforementioned embodiment, which will not be repeated here for the sake of brevity.

[0051] Step S124: In response to the target patient's medical data triggering the pain judgment condition, the current preset pain type is excluded.

[0052] For details, please refer to the specific steps of the aforementioned embodiment, which will not be repeated here for the sake of brevity.

[0053] Step S131: In response to the target patient's medical data not triggering the pain judgment condition, a preset pain type is used as a candidate pain type, and an index value is calculated based on a score calculation method for the candidate pain type.

[0054] For details, please refer to the specific steps of the aforementioned embodiment, which will not be repeated here for the sake of brevity.

[0055] Step S141: performing threshold comparison and sorting based on the index values ​​of each candidate pain type to obtain the target pain type of the target patient.

[0056] For details, please refer to the specific steps of the aforementioned embodiment, which will not be repeated here for the sake of brevity.

[0057] The above scheme obtains configuration data for a number of preset pain types and medical data related to the pain location of a target patient. The configuration data for the preset pain types includes pain configuration conditions for the preset pain types and a score calculation method. Based on the matching results between the medical data and the pain configuration conditions of each preset pain type, the preset pain type is selected as a candidate pain type. Based on the medical data and the configuration data for the candidate pain types, an index value for the candidate pain type is calculated. Based on the index value of each candidate pain type, the candidate pain type is selected as the target pain type for the target patient. On the one hand, screening the candidate pain types based on the matching results between the pain configuration conditions of each preset pain type and the medical data of the target patient helps to screen out preset pain types that match the cause of the target patient while taking into account "same disease, different symptoms, different diseases, same symptoms" and select them as candidate pain types for further analysis. On the other hand, calculating the index value for each candidate pain type based on the score calculation formula can convert the medical data of the target patient into an objective numerical value for the candidate pain type. The calculation of the index value can explain the mechanism of pain occurrence under the cause corresponding to the candidate pain type, thereby improving the accuracy of the target pain type of the target patient determined based on the index value of each candidate pain type. Therefore, the objectivity and accuracy of pain type identification can be improved.

[0058] See also Figure 3 , Figure 3 This is a flow chart of an embodiment of the index value prediction method of the present application. Specifically, it may include the following steps: Step S31: obtaining an indicator value prediction model based on the sample indicator values ​​of a plurality of sample patients for the same preset pain type and the sample medical data of the preset types of the sample patients.

[0059] In the disclosed embodiment, the sample index value is calculated based on any of the aforementioned pain recognition method embodiments, and will not be described again here for the sake of brevity.

[0060] In one implementation scenario, sample medical data is predicted based on an initial model to obtain an expected index value. Based on the difference between the expected index value and the corresponding sample index value, the initial model is adjusted until convergence to obtain an index value prediction model.

[0061] In a specific implementation scenario, the indicator value prediction model can be built and trained using machine learning or deep learning algorithms. By learning and analyzing the medical data of a large number of sample patients, the model can automatically extract key features related to pain types and establish a mapping relationship between features and pain types.

[0062] In another specific implementation scenario, based on the mapping relationship between the sample index values ​​of each sample patient for the same preset pain type and the sample medical data of the preset types of sample patients, a functional relationship between the data is constructed, and the constructed functional relationship is used as an index value prediction model.

[0063] It should be noted that the above embodiments are only possible implementation methods, and the specific structure of the indicator value prediction model is not limited in this application.

[0064] Step S32: predicting the target medical data of the preset type of the patient to be tested based on the index value prediction model to obtain the predicted index value of the preset pain type of the patient to be tested.

[0065] In one implementation scenario, during the prediction phase, the indicator value prediction model can predict the corresponding predicted indicator value based on a relatively single, preset type of medical data from the patient to be tested. This approach improves the effectiveness of medical testing by predicting relatively accurate indicator values ​​based on the indicator value prediction model derived from sample indicator values ​​of several sample patients for the same preset pain type and sample medical data from the sample patients of the preset type, even when the type of medical data from the patient to be tested is limited.

[0066] In a specific implementation scenario, a large amount of sample patient data is selected from a medical database. This data includes complete sample medical data for predefined categories and sample index values ​​corresponding to clinically confirmed pain types. An initial model uses deep learning or machine learning algorithms, such as convolutional neural networks and random forests, to extract features and perform pattern recognition on the sample medical data for the predefined categories, outputting predicted index values ​​for each predefined pain type. By calculating the difference between the predicted index values ​​and the actual sample index values ​​of the sample patients, using metrics such as mean squared error and cross-entropy loss, a backpropagation algorithm is used to iteratively adjust the parameters of the index prediction sub-model to optimize the model's prediction accuracy and generalization ability, enabling the model to more accurately capture the mapping between medical data and pain types. In a practical application scenario, when diagnosing a patient, the system obtains target medical data for the predefined categories and inputs them into the trained index prediction sub-model. Based on the learned features and patterns, the model analyzes and processes the target medical data and outputs predicted index values ​​corresponding to the predicted pain type. This predicted index value is presented as the final pathology prediction result, and a visual prediction analysis report is generated, demonstrating the contribution of each dimension of medical data to the prediction result.

[0067] The above scheme obtains an index value prediction model based on sample index values ​​of several sample patients for the same preset pain type and sample medical data of the sample patients of the preset category, and the sample index values ​​are calculated based on any of the aforementioned pain recognition method embodiments. The target medical data of the preset category of the patient to be tested is predicted based on the index value prediction model to obtain a predicted index value for the preset pain type of the patient to be tested. Therefore, in the case where the type of medical data of the patient to be tested is limited, a relatively accurate index value is predicted by the index value prediction model based on the sample index values ​​of several sample patients for the same preset pain type and the sample medical data of the sample patients of the preset category, thereby improving the detection effect of the medical test.

[0068] See also Figure 4 , Figure 4 This is a flow chart of an embodiment of the drug efficacy response method of the present application. Specifically, it may include the following steps: Step S41: constructing sample pairs based on the first historical indicator value of each sample patient of the same preset pain type before using the preset medicine, the second historical indicator value after using the preset medicine for a preset period, and the preset period.

[0069] In the embodiment of the present disclosure, the first historical indicator value and the second historical indicator value are calculated based on any of the aforementioned pain recognition method embodiments, and are not described again here for the sake of brevity.

[0070] Step S42: obtaining a drug efficacy response model based on a number of sample pairs.

[0071] In one implementation scenario, a drug response model can be built and trained using machine learning or deep learning algorithms. By learning and analyzing the medical data of a large number of sample patients, the model can automatically extract key features related to pain types and establish a mapping relationship between these features and pain types.

[0072] In another implementation scenario, based on the mapping relationship between the first historical indicator value of each sample patient of the same preset pain type who did not use the preset drug, the second historical indicator value after using the preset drug for a preset period, and the preset period, a functional relationship between the data is constructed, and the constructed functional relationship is used as a drug efficacy response model.

[0073] It should be noted that the above embodiments are only possible implementation methods, and the specific structure of the indicator value prediction model is not limited in this application.

[0074] Step S43: predicting the current index value of the preset pain type of the patient to be tested and the usage data of the preset drug based on the drug efficacy response model to obtain the drug efficacy response trend of the patient to be tested using the preset drug.

[0075] In the disclosed embodiment, the current index value is calculated based on any of the aforementioned pain recognition method embodiments, and will not be described in detail here for the sake of brevity.

[0076] In one implementation scenario, the pharmacodynamic response trend for a patient using a pre-determined medication represents the pattern and direction of changes in the patient's condition indicators and symptoms over time after using the pre-determined medication. Specifically, the pharmacodynamic response trend is presented through quantitative data and visualization. For example, the pain score is expected to decrease by 30% on the 7th day after medication, by 50% on the 14th day, and maintain a 60% relief level on the 30th day.

[0077] In a specific implementation scenario, a large number of sample patients' historical medical data are screened from a medical database. For each pre-prescribed medication, the system extracts the patient's medical data before medication use and quantifies it into a first historical indicator value. This indicator encompasses multiple dimensions, such as pain intensity scores, inflammatory indicators, and physiological function parameters, reflecting the patient's baseline condition before medication use. Simultaneously, the system extracts medical data after the patient has used the pre-prescribed medication for a predetermined period, such as 7 or 14 days. This second historical indicator value can be flexibly set based on the drug's characteristics and clinical experience, to visually reflect the changes in the patient's condition after medication intervention. Based on this data, the system constructs a first sample pair—a three-dimensional dataset consisting of "first historical indicator value, second historical indicator value, and predetermined period." This first sample pair includes a comparison of the patient's condition before and after medication use and introduces a time dimension variable, enabling the model to learn the dynamic changes in drug efficacy over time. Using this large number of first sample pairs as input, the efficacy prediction sub-model uses time series data analysis algorithms, such as recurrent neural networks (RNNs) and long short-term memory networks (LSTMs), to perform in-depth training on the sample data. During the training process, the model continuously adjusts its parameters to optimize its ability to fit the complex relationship between drug efficacy and time and disease indicators, thereby establishing a stable and reliable prediction mechanism. When applied to patients to be tested, the system first collects their current medical data, including basic medical history, symptoms, latest test results, etc. After inputting these data into the trained drug efficacy prediction sub-model, the model simulates and predicts the drug efficacy trend of patients after using preset drugs based on the learned drug efficacy rules and the individual characteristics of the patients to be tested. This trend is presented in the form of visual charts and quantitative data, such as the pain score decline curve, the inflammatory index change line chart, the expected pain relief rate, the degree of improvement in indicators, etc., which intuitively show the possible therapeutic effects of the drug at different time points.

[0078] The above scheme constructs sample pairs based on the first historical index value of each sample patient of the same preset pain type before using the preset drug, the second historical index value after using the preset drug for a preset period, and the preset period. The first historical index value and the second historical index value are calculated based on any of the aforementioned pain identification method embodiments. Based on several sample pairs, a drug efficacy response model is obtained. Based on the drug efficacy response model, the current index value of the preset pain type of the patient to be tested and the usage data of the preset drug are predicted to obtain the drug efficacy response trend of the patient to be tested using the preset drug, and the current index value is calculated based on any of the aforementioned pain identification method embodiments. Therefore, predicting the trend of indicator changes after the patient takes the drug through the drug efficacy response model is beneficial to predicting the efficacy before medication to avoid ineffective drug trials, as well as dynamically monitoring and optimizing the scheme during medication, thereby improving treatment efficiency and safety.

[0079] See also Figure 5 , Figure 5 This is a flow chart of an embodiment of the disease prediction method of the present application. Specifically, it may include the following steps: Step S51: obtaining a symptom prediction model based on the sample index values ​​and sample symptoms of each sample patient of the same preset pain type.

[0080] In the disclosed embodiment, the sample index value is calculated based on any of the aforementioned pain recognition method embodiments, and will not be described again here for the sake of brevity.

[0081] In one implementation scenario, a symptom prediction model can be built and trained using machine learning or deep learning algorithms. By learning and analyzing the medical data of a large number of sample patients, the model can automatically extract key features related to pain types and establish a mapping relationship between these features and pain types.

[0082] In another implementation scenario, based on the mapping relationship between the sample index values ​​of each sample patient of the same preset pain type and the sample symptoms, a functional relationship between the data is constructed, and the constructed functional relationship is used as a symptom prediction model.

[0083] It should be noted that the above embodiments are only possible implementation methods, and the specific structure of the indicator value prediction model is not limited in this application.

[0084] Step S52: Predict the current indicator value of the patient to be tested based on the disease prediction model to obtain the disease prediction result of the patient to be tested.

[0085] In the disclosed embodiment, the current index value is calculated based on any of the aforementioned pain recognition method embodiments, and will not be described in detail here for the sake of brevity.

[0086] In one implementation scenario, the disease prediction results of the patient to be tested also include the disease risk prediction within a preset period. For example, for every 10-point increase in the sample index value of cancer bone pain, the risk of pathological fracture within 6 months increases by 28%. The disease risk can be predicted based on the current index value of the patient to be tested.

[0087] In a specific implementation scenario, during the training process, the disease prediction model uses deep learning algorithms such as Transformers, graph neural networks, or traditional machine learning algorithms such as support vector machines and random forests to extract features and recognize patterns in sample diseases. By analyzing the potential feature combinations in the sample medical data, it learns the contribution weights of different data dimensions to disease diagnosis. Combining the logical relationship between sample indicator values ​​and sample symptoms, it establishes a predictive mapping from data to symptoms. By continuously adjusting model parameters to minimize prediction error, it ultimately forms a disease prediction model that converges with training. When applied to a patient, current medical data, including the latest symptom description, laboratory test reports, and imaging results, is first collected to obtain current indicator values. These values ​​are then input into the trained disease prediction model. The model then analyzes and predicts the patient's condition based on the learned disease diagnostic patterns, outputting a disease prediction result. Specifically, the disease prediction result includes not only the name of the possible pain condition, but also information such as the prediction confidence and key diagnostic evidence. It is presented in the form of visual charts and text reports, and provides differentiation suggestions from other similar conditions to assist in a comprehensive assessment of the pain type.

[0088] The above scheme obtains a symptom prediction model based on the sample index values ​​and sample symptoms of each sample patient with the same preset pain type, where the sample index values ​​are calculated based on any of the aforementioned pain identification method embodiments. The symptom prediction model then predicts the current index values ​​of the patient to be tested, where the current index values ​​are calculated based on any of the aforementioned pain identification method embodiments, to obtain a symptom prediction result for the patient to be tested. Therefore, the symptom prediction model, derived from the sample data of several sample patients, can accurately infer the pain symptoms of the patient to be tested, improving diagnostic accuracy and efficiency while accommodating individual differences as much as possible.

[0089] See also Figure 6 , Figure 6 This is a schematic diagram of the framework of an embodiment of the pain recognition device of the present application. Figure 6As shown, the pain recognition device 60 includes an acquisition module 61, a selection module 62, a calculation module 63 and an identification module 64. The acquisition module 61 is used to obtain configuration data of several preset pain types and obtain medical data related to the pain site of the target patient; wherein the configuration data of the preset pain type includes the pain configuration conditions and score calculation method of the preset pain type; the selection module 62 is used to select the preset pain type as the candidate pain type based on the matching results between the medical data and the pain configuration conditions of each preset pain type; the calculation module 63 is used to calculate the index value of the candidate pain type based on the medical data and the configuration data of the candidate pain type; the identification module 64 is used to select the candidate pain type as the target pain type of the target patient based on the index value of each candidate pain type.

[0090] In the above scheme, the pain identification device 60 obtains configuration data for a number of preset pain types and medical data related to the pain location of a target patient. The configuration data for the preset pain types includes pain configuration conditions for the preset pain types and a score calculation method. Based on the matching results between the medical data and the pain configuration conditions of each preset pain type, the preset pain type is selected as a candidate pain type. Based on the medical data and the configuration data for the candidate pain types, an index value for the candidate pain type is calculated. Based on the index value for each candidate pain type, the candidate pain type is selected as the target pain type for the target patient. On the one hand, screening the candidate pain types based on the matching results between the pain configuration conditions of each preset pain type and the medical data of the target patient facilitates screening the preset pain types that match the etiology of the target patient, taking into account both "same disease, different symptoms" and "different diseases, same symptoms," as candidate pain types for further analysis. On the other hand, calculating the index value for each candidate pain type based on the score calculation formula converts the medical data of the target patient into an objective numerical value specific to the candidate pain type. The calculated index value can explain the mechanism of pain occurrence under the corresponding etiology of the candidate pain type, thereby improving the accuracy of the target pain type determined based on the index value of each candidate pain type. Therefore, the objectivity and accuracy of pain type identification can be improved.

[0091] In the embodiment of the present disclosure, the calculation module 63 also includes a support condition determination module (not shown), which is used to determine the pain support condition that matches the score calculation method of the candidate pain type; the calculation module 63 also includes a first data determination module (not shown), which is used to obtain the first data corresponding to the pain support condition based on the medical data consistent with the pain support condition data type; the calculation module 63 also includes an indicator value calculation submodule (not shown), which is used to calculate the indicator value of the candidate pain type based on the score calculation method and the first data corresponding to the pain support condition.

[0092] In the embodiment of the present disclosure, the configuration data of the preset pain type also includes a pain exclusion condition. After the index value of the candidate pain type is calculated based on the score calculation method and the first data corresponding to the pain support condition, the pain identification device 60 also includes a contradiction item calculation module (not shown), which is used to respond to the medical data including the second data matching the pain exclusion condition, and to weight the target weight and the second data corresponding to the pain exclusion condition to obtain a contradiction item score; the pain identification device 60 also includes a contradiction difference module (not shown), which is used to obtain an updated index value based on the difference between the index value of the candidate pain type and the contradiction item score.

[0093] In the embodiment of the present disclosure, the pain configuration conditions include pain-required conditions and pain-exclusion conditions. The selection module 62 also includes a first matching module (not shown) for obtaining a first matching result between the medical data and the pain-required conditions; the selection module 62 also includes a second matching module (not shown) for obtaining a second matching result between the medical data and the pain-exclusion conditions in response to the first matching result indicating that the medical data meets the pain-required conditions; the selection module 62 also includes a third matching module (not shown) for selecting a preset pain type as a candidate pain type in response to the second matching result indicating that the medical data does not trigger the pain-exclusion conditions.

[0094] In the embodiment of the present disclosure, the identification module 64 also includes a confidence acquisition module (not shown) for acquiring a first confidence threshold and a second confidence threshold for each candidate pain type; wherein the first confidence threshold is greater than the second confidence threshold; the identification module 64 also includes a first identification submodule (not shown) for, in response to the index value of the candidate pain type being not less than the first confidence threshold, taking the candidate pain type as the target pain type and generating a first prompt for indicating treatment; the identification module 64 also includes a second identification submodule (not shown) for, in response to the index value of the candidate pain type being less than the first confidence threshold and not less than the second confidence threshold, generating a second prompt for indicating further examination; the identification module 64 also includes a third identification submodule (not shown) for, in response to the index value of the candidate pain type being less than the second confidence threshold, excluding the candidate pain type.

[0095] See also Figure 7 , Figure 7 This is a schematic diagram of the framework of an embodiment of the index value prediction device of this application. Figure 7As shown, the index value prediction device 70 includes: a first model module 71 and a first prediction module 72. The first model module 71 is used to obtain an index value prediction model based on sample index values ​​of several sample patients for the same preset pain type and sample medical data of preset types of sample patients; wherein the sample index values ​​are calculated based on any of the aforementioned pain recognition method embodiments; the first prediction module 72 is used to predict the target medical data of the preset type of the patient to be detected based on the index value prediction model, and obtain the predicted index value of the preset pain type of the patient to be detected.

[0096] In the above scheme, the index value prediction device 70 obtains an index value prediction model based on the sample index values ​​of several sample patients for the same preset pain type and the sample medical data of the sample patients of the preset type, and the sample index values ​​are calculated based on any of the aforementioned pain recognition method embodiments. The target medical data of the preset type of the patient to be tested is predicted based on the index value prediction model to obtain a predicted index value for the preset pain type of the patient to be tested. Therefore, in the case where the type of medical data of the patient to be tested is limited, a relatively accurate index value is predicted by the index value prediction model obtained based on the sample index values ​​of several sample patients for the same preset pain type and the sample medical data of the sample patients of the preset type, thereby improving the detection effect of the medical test.

[0097] In some disclosed embodiments, the first model module 71 also includes a first expectation module (not shown), which is used to predict the sample medical data based on the initial model to obtain the expected index value; the first model module 71 also includes a first adjustment module (not shown), which is used to adjust the initial model until convergence based on the difference between the expected index value and the corresponding sample index value to obtain an index value prediction model.

[0098] See also Figure 8 , Figure 8 This is a schematic diagram of the framework of an embodiment of the drug efficacy response device of the present application. Figure 8 As shown, the drug efficacy response device 80 includes: a second construction module 81, a second model module 82 and a second prediction module 83, the second construction module 81 is used to construct sample pairs based on the first historical indicator value of each sample patient of the same preset pain type who did not use the preset drug, the second historical indicator value after using the preset drug for a preset period, and the preset period; wherein the first historical indicator value and the second historical indicator value are calculated based on any of the aforementioned pain recognition method embodiments; the second model module 82 is used to obtain a drug efficacy response model based on a number of sample pairs; the second prediction module 83 predicts the current indicator value of the preset pain type of the patient to be detected and the usage data of the preset drug based on the drug efficacy response model, and obtains the drug efficacy response trend of the patient to be detected using the preset drug; wherein the current indicator value is calculated based on any of the aforementioned pain recognition method embodiments.

[0099] In the above scheme, the drug efficacy response device 80 constructs sample pairs based on the first historical indicator value of each sample patient of the same preset pain type before using the preset drug, the second historical indicator value after using the preset drug for a preset period, and the preset period. The first historical indicator value and the second historical indicator value are calculated based on any of the aforementioned pain identification method embodiments. Based on several sample pairs, a drug efficacy response model is obtained. Based on the drug efficacy response model, the current indicator value of the preset pain type of the patient to be tested and the usage data of the preset drug are predicted to obtain the drug efficacy response trend of the patient to be tested using the preset drug, and the current indicator value is calculated based on any of the aforementioned pain identification method embodiments. Therefore, predicting the indicator change trend after the patient takes the drug through the drug efficacy response model is beneficial for predicting the efficacy of the drug before taking the drug to avoid ineffective drug trials, as well as for dynamic monitoring and optimization of the drug regimen during use, thereby improving treatment efficiency and safety.

[0100] See also Figure 9 , Figure 9 This is a schematic diagram of the framework of an embodiment of the disease prediction device of the present application. Figure 9 As shown, the symptom prediction device 90 includes: a third model module 91 and a third prediction module 92. The third model module 91 is used to obtain a symptom prediction model based on the sample index values ​​and sample symptoms of each sample patient of the same preset pain type; wherein the sample index value is calculated based on any of the aforementioned pain identification method embodiments; the third prediction module 92 is used to predict the current index value of the patient to be tested based on the symptom prediction model to obtain a symptom prediction result of the patient to be tested; wherein the current index value is calculated based on any of the aforementioned pain identification method embodiments.

[0101] In the above scheme, the symptom prediction device 90 obtains a symptom prediction model based on the sample index values ​​and sample symptoms of each sample patient with the same preset pain type, where the sample index values ​​are calculated based on any of the aforementioned pain identification method embodiments. The symptom prediction model then predicts the current index values ​​of the patient to be tested, where the current index values ​​are calculated based on any of the aforementioned pain identification method embodiments, to obtain a symptom prediction result for the patient to be tested. Therefore, the symptom prediction model, derived from the sample data of several sample patients, can accurately infer the pain symptoms of the patient to be tested, thereby improving diagnostic accuracy and efficiency while accommodating individual differences as much as possible.

[0102] See also Figure 10 , Figure 10It is a schematic diagram of the framework of an embodiment of an electronic device of the present application. The electronic device 100 includes at least a memory 101 and a processor 102 coupled to each other, the memory 101 stores at least program instructions, and the processor 102 is used to execute the program instructions to implement any of the above-mentioned pain recognition methods, or, indicator value prediction methods, or, drug efficacy response methods, or, steps in the embodiment of the disease prediction method. For details, please refer to the aforementioned disclosed embodiments, which will not be repeated here. It should be noted that the electronic device 100 may include but is not limited to learning machines, office books, smart large screens and other devices, and the specific type of the electronic device 100 is not limited here.

[0103] Specifically, processor 102 is used to control itself and memory 101 to implement the steps of any of the aforementioned pain identification methods, or index value prediction methods, or drug efficacy response methods, or symptom prediction method embodiments. Processor 102 may also be referred to as a CPU (Central Processing Unit). Processor 102 may be an integrated circuit chip with signal processing capabilities. Processor 102 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component. A general-purpose processor may be a microprocessor, or the processor may be any conventional processor. Furthermore, processor 102 may be implemented by an integrated circuit chip.

[0104] In the above scheme, the electronic device 100 obtains configuration data for a number of preset pain types and medical data related to the pain location of a target patient. The configuration data for the preset pain types includes pain configuration conditions for the preset pain types and a score calculation method. Based on the matching results between the medical data and the pain configuration conditions of each preset pain type, the preset pain type is selected as a candidate pain type. Based on the medical data and the configuration data for the candidate pain types, an index value for the candidate pain type is calculated. Based on the index value for each candidate pain type, the candidate pain type is selected as the target pain type for the target patient. On the one hand, screening the candidate pain types based on the matching results between the pain configuration conditions of each preset pain type and the medical data of the target patient helps to screen out preset pain types that match the etiology of the target patient, taking into account both "same disease, different symptoms" and "different diseases, same symptoms," as candidate pain types for further analysis. On the other hand, calculating the index value for each candidate pain type based on the score calculation formula can convert the medical data of the target patient into an objective numerical value for the candidate pain type. The calculated index value can explain the mechanism of pain occurrence under the corresponding etiology of the candidate pain type, thereby improving the accuracy of the target pain type determined based on the index value of each candidate pain type. Therefore, the objectivity and accuracy of pain type identification can be improved.

[0105] See also Figure 11 , Figure 11 Schematic diagram of a computer-readable storage medium according to an embodiment of the present invention. Computer-readable storage medium 110 stores program instructions 111 executable by a processor. Program instructions 111 are used to implement the steps of any of the aforementioned pain identification methods, indicator value prediction methods, drug efficacy response methods, or symptom prediction methods.

[0106] In some embodiments, the functions or modules included in the device provided by the embodiments of the present disclosure can be used to execute the method described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments. For the sake of brevity, it will not be repeated here.

[0107] The above description of the various embodiments tends to emphasize the differences between the various embodiments. The same or similar aspects can be referenced with each other and will not be repeated herein for the sake of brevity.

[0108] In the several embodiments provided in this application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device implementation methods described above are only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0109] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0110] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0111] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the various implementation methods of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0112] If the technical solution of this application involves personal information, the product that applies the technical solution of this application has clearly informed the personal information processing rules and obtained the individual's voluntary consent before processing personal information. If the technical solution of this application involves sensitive personal information, the product that applies the technical solution of this application has obtained the individual's separate consent before processing sensitive personal information, and at the same time meets the "explicit consent" requirement. For example, on personal information collection devices such as cameras, a clear and prominent sign is set to inform that the personal information collection scope has been entered and personal information will be collected. If the individual voluntarily enters the collection scope, it is deemed that they agree to the collection of their personal information; or on the personal information processing device, when the personal information processing rules are notified by obvious signs / information, the individual's authorization is obtained through pop-up information or by asking the individual to upload their personal information; among which, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the type of personal information processed.

Claims

1. A pain recognition method, characterized in that: include: Acquiring configuration data for a plurality of preset pain types and obtaining medical data related to the pain site of a target patient; wherein the configuration data for the preset pain types includes pain configuration conditions and a score calculation method for the preset pain types; selecting the preset pain type as a candidate pain type based on a matching result between the medical data and the pain configuration conditions of each preset pain type; Calculating an index value of the candidate pain type based on the medical data and the configuration data of the candidate pain type; Based on the index value of each candidate pain type, the candidate pain type is selected as the target pain type of the target patient.

2. The method according to claim 1, characterized in that The step of calculating the index value of the candidate pain type based on the medical data and the configuration data of the candidate pain type includes: determining a pain support condition in the configuration data that matches the score calculation method of the candidate pain type; obtaining first data corresponding to the pain support condition based on the medical data consistent with the data type of the pain support condition; Based on the score calculation method and the first data corresponding to the pain support condition, an index value of the candidate pain type is calculated.

3. The method according to claim 2, characterized in that The configuration data of the preset pain type further includes a pain exclusion condition. After calculating the index value of the candidate pain type based on the score calculation method and the first data corresponding to the pain support condition, the method further includes: In response to the medical data including second data matching the pain elimination condition, weighting the second data based on a target weight corresponding to the pain elimination condition to obtain a contradiction item score; Based on the difference between the index value of the candidate pain type and the conflict item score, an updated index value is obtained.

4. The method according to claim 1, wherein The pain configuration condition includes a pain requirement condition and a pain exclusion condition. The selecting of the preset pain type as a candidate pain type based on the matching result between the medical data and the pain configuration condition of each preset pain type includes: Obtaining a first matching result between the medical data and the necessary pain condition; In response to the first matching result indicating that the medical data meets the pain-requiring condition, obtaining a second matching result between the medical data and the pain-eliminating condition; In response to the second matching result indicating that the medical data does not trigger the pain exclusion condition, the preset pain type is selected as a candidate pain type.

5. The method according to claim 1, characterized in that The step of selecting the candidate pain type as the target pain type of the target patient based on the index value of each candidate pain type includes: Obtaining a first confidence threshold and a second confidence threshold for each candidate pain type; wherein the first confidence threshold is greater than the second confidence threshold; In response to the indicator value of the candidate pain type being not less than the first confidence threshold, using the candidate pain type as the target pain type and generating a first prompt for instructing treatment; In response to the indicator value of the candidate pain type being less than the first confidence threshold and not less than the second confidence threshold, generating a second prompt for instructing further examination; In response to the indicator value of the candidate pain type being less than the second confidence threshold, the candidate pain type is excluded.

6. A method for predicting an index value, characterized in that: include: Based on sample index values ​​of a plurality of sample patients for the same preset pain type and sample medical data of preset types of the sample patients, an index value prediction model is obtained; wherein the sample index values ​​are calculated based on the pain recognition method according to any one of claims 1 to 5; The target medical data of the preset type of the patient to be detected is predicted based on the indicator value prediction model to obtain the prediction indicator value of the preset pain type of the patient to be detected.

7. The method according to claim 6, characterized in that The indicator value prediction model is obtained based on the sample indicator values ​​of a plurality of sample patients for the same preset pain type and the sample medical data of the preset types of the sample patients, including: Predicting the sample medical data based on the initial model to obtain expected indicator values; Based on the difference between the expected index value and the corresponding sample index value, the initial model is adjusted until convergence to obtain the index value prediction model.

8. A drug efficacy response method, characterized in that: include: Constructing a sample pair based on a first historical indicator value of each sample patient of the same preset pain type before using a preset drug, a second historical indicator value after using the preset drug for a preset period, and the preset period; wherein the first historical indicator value and the second historical indicator value are calculated based on the pain recognition method according to any one of claims 1 to 5; Based on the plurality of sample pairs, a drug efficacy response model is obtained; Based on the drug efficacy response model, the current index value of the preset pain type of the patient to be tested and the usage data of the preset drug are predicted to obtain the drug efficacy response trend of the patient to be tested using the preset drug; wherein, the current index value is calculated based on the pain recognition method according to any one of claims 1 to 5.

9. A disease prediction method, characterized in that: include: Based on the sample index values ​​and sample symptoms of each sample patient of the same preset pain type, a symptom prediction model is obtained; wherein the sample index values ​​are calculated based on the pain recognition method according to any one of claims 1 to 5; The current index value of the patient to be tested is predicted based on the disease prediction model to obtain a disease prediction result of the patient to be tested; wherein, the current index value is calculated based on the pain recognition method according to any one of claims 1 to 5.

10. An electronic device, characterized in that: It includes a memory and a processor coupled to each other, wherein the memory stores program instructions, and the processor is used to execute the program instructions to implement the pain recognition method described in any one of claims 1 to 5, or to implement the indicator value prediction method described in any one of claims 6 to 7, or to implement the drug efficacy response method described in claim 8, or to implement the disease prediction method described in claim 9.

11. A computer-readable storage medium having program instructions stored thereon, characterized in that: When the program instructions are executed by the processor, the pain recognition method described in any one of claims 1 to 5 is implemented, or the index value prediction method described in any one of claims 6 to 7 is implemented, or the drug efficacy response method described in claim 8 is implemented, or the disease prediction method described in claim 9 is implemented.

Images