Intelligent physical examination offline data collection method

Abstract

The application discloses a kind of wisdom physical examination offline data acquisition methods, it is related to electric digital data processing technical field, including the following steps, identity token is generated by identification and extraction of identification information through acquisition terminal, identity token is synchronized to physical examination equipment, and original physiological index data is collected through physical examination equipment, and structured data file is generated, structured data file is encrypted and handled, and storage is carried out, respectively calculate the hash value of structured data file and original physiological index data after encryption processing, and comparison verification is carried out, form metadata file, metadata file and structured data file are associated storage, break through network limit and stable data acquisition, ensure data security by identity token binding and encrypted storage, double hash check prevents tampering and can be traced back, built-in lightweight decision engine multidimensional analysis index judges chronic disease risk, sound and light signal helps medical staff to distinguish key population quickly, improve primary physical examination efficiency and chronic disease early screening ability.

Description

Technical Field

[0001] This invention relates to the field of electronic digital data processing technology, and in particular to a smart physical examination offline data acquisition method. Background Technology

[0002] In recent years, with the vigorous development of smart healthcare technology, medical data collection, as a core link in health management and disease prevention, has become increasingly important in terms of accuracy, security, and real-time performance. The efficient collection of physiological indicator data not only provides a basis for individualized health assessment, but also serves as a key support for early screening of chronic diseases and follow-up management in primary healthcare.

[0003] However, significant pain points still exist in existing offline physical examination data collection scenarios. Traditional solutions struggle to reliably bind the examinee's identity to the device cluster in environments with no or weak network coverage. Inaccurate extraction of identity information often leads to difficulties in data traceability. The structured processing of raw physiological indicator data lacks unified medical standards, encrypted storage mechanisms have security vulnerabilities, and data integrity verification relies on a single hash algorithm, which cannot resist the risk of data tampering in complex environments. At the same time, existing collection terminals lack real-time intelligent screening capabilities, making it difficult to simultaneously complete chronic disease risk assessment during the data collection phase, resulting in efficiency bottlenecks for primary healthcare resources in the later stages of data processing. Summary of the Invention

[0004] The technical problem addressed by this invention is as follows: However, significant pain points still exist in existing offline physical examination data collection scenarios. Traditional solutions struggle to reliably bind the examinee's identity to the device cluster in environments with no or weak network coverage. Inaccurate extraction of identity information often leads to difficulties in data traceability. The structured processing of raw physiological indicator data lacks unified medical standards, the encrypted storage mechanism has security vulnerabilities, and data integrity verification relies on a single hash algorithm, which cannot resist the risk of data tampering in complex environments. At the same time, existing collection terminals lack real-time intelligent screening capabilities, making it difficult to simultaneously complete chronic disease risk assessment during the data collection stage, resulting in efficiency bottlenecks for primary healthcare resources in the later stages of data processing.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a smart physical examination offline data collection method, comprising the following steps:

[0006] Step S1: The identification information is identified and extracted by the acquisition terminal to generate an identity token, and the identity token is synchronized to the physical examination device.

[0007] Step S2: Collect raw physiological indicator data through the physical examination equipment, generate a structured data file, encrypt the structured data file, and store it;

[0008] Step S3: Calculate the hash values ​​of the encrypted structured data file and the original physiological indicator data respectively, compare and verify them to form a metadata file, and store the metadata file and the structured data file together.

[0009] As a preferred embodiment of the intelligent physical examination offline data collection method described in this invention, the identity token is obtained by the collection terminal by recognizing the examinee's medical insurance card or ID card;

[0010] The data collection terminal establishes a short-range communication link with one or more medical examination devices via Bluetooth or NFC, and synchronizes the identity token to the medical examination device to form a binding relationship between the examinee's identity and the medical examination device cluster.

[0011] The physical examination equipment collects raw physiological index data in environments with no or weak network coverage.

[0012] The raw physiological index data are generated into a structured data file according to a preset medical data standard;

[0013] The structured data file includes physiological indicator data, collection timestamps, unique codes of the medical examination equipment, and identity token hash values.

[0014] As a preferred embodiment of the intelligent physical examination offline data collection method of the present invention, the collection terminal has a built-in lightweight decision engine.

[0015] The lightweight decision engine has a built-in preset chronic disease screening and judgment standard. When physiological indicator data is acquired, the screening logic corresponding to the physiological indicator data is called, and real-time verification is performed according to the chronic disease screening and judgment standard, and the verification result is output.

[0016] The screening logic includes:

[0017] Feature parameters are extracted from the physiological index data;

[0018] The characteristic parameters include physiological indicator measurements, abnormal fluctuation range of indicators, indicator trend parameters, ratio parameters of multiple related indicators, and indicator measurement time interval parameters.

[0019] The multiple correlation indicators include the combination of blood glucose and insulin correlation indicators, the combination of blood pressure and blood lipid correlation indicators, and the combination of body mass index and waist circumference correlation indicators.

[0020] Determine the verification dimensions and threshold ranges for the physiological indicator data;

[0021] The verification dimensions include basic physiological indicators, dynamic changes of indicators, multi-indicator correlation, and population feature adaptation.

[0022] The threshold ranges for the verification dimensions include the threshold ranges for basic physiological indicators, the threshold ranges for dynamic changes in indicators, the threshold ranges for multi-indicator correlation, and the threshold ranges for population feature adaptation.

[0023] As a preferred embodiment of the intelligent physical examination offline data acquisition method of the present invention, the extracted feature parameters are compared with the verification dimension threshold range, and logical verification operations are performed, including:

[0024] The feature parameters are matched with the verification dimensions respectively;

[0025] Once the matching is complete, the feature parameters are compared with the threshold range of the corresponding verification dimension to determine the threshold compliance of the feature parameters.

[0026] Based on the threshold compliance, single-dimensional verification results are generated for the basic physiological indicator dimension, indicator dynamic change dimension, multi-indicator correlation dimension, and population feature adaptation dimension, respectively, including the verification results for the basic physiological indicator dimension, the verification results for the indicator dynamic change dimension, the verification results for the multi-indicator correlation dimension, and the verification results for the population feature adaptation dimension.

[0027] The verification results of the basic physiological indicators, the dynamic changes of indicators, the multi-indicator correlation, and the population feature adaptation are respectively divided into normal state, critical state, and abnormal state according to the threshold compliance.

[0028] As a preferred embodiment of the intelligent physical examination offline data collection method of the present invention, the single-dimensional verification results are combined and calculated based on a preset logical operation strategy to generate a combined operation result;

[0029] The logical operation strategy includes AND operation, OR operation, and priority determination;

[0030] The combined operation results include normal operation, abnormal operation, and abnormal AND operation.

[0031] The normal operation includes:

[0032] When the verification results of the basic physiological indicators, the dynamic changes of indicators, the multi-indicator correlation, and the population feature adaptation are all in a normal state, and no critical or abnormal state occurs, the operation is judged to be normal.

[0033] The OR operation exceptions include:

[0034] When any single-dimensional verification result among the basic physiological indicator dimension verification result, indicator dynamic change dimension verification result, multi-indicator correlation dimension verification result, and population feature adaptation dimension verification result is in an abnormal or critical state, it is determined to be an OR operation abnormality.

[0035] The AND operation exceptions include:

[0036] When the verification results of the basic physiological indicator dimension, the indicator dynamic change dimension, the multi-indicator correlation dimension, and the population feature adaptation dimension are all in an abnormal state, it is determined that the AND operation is abnormal.

[0037] As a preferred embodiment of the intelligent physical examination offline data collection method described in this invention, the priority determination includes:

[0038] The verification results of the basic physiological indicators have a higher priority than the verification results of the dynamic changes in the indicators.

[0039] The verification results of the dynamic change dimension of the indicator have higher priority than the verification results of the multi-indicator correlation dimension.

[0040] The multi-indicator correlation dimension verification result has a higher priority than the population feature adaptation dimension verification result.

[0041] When single-dimensional verification results conflict, the higher-priority single-dimensional verification result is used for combined calculation.

[0042] The verification result is determined based on the combined calculation result, which represents the degree of matching between the feature parameters and the preset chronic disease screening judgment criteria.

[0043] As a preferred embodiment of the intelligent physical examination offline data collection method of the present invention, the chronic disease risk level is determined based on the verification result;

[0044] The chronic disease risk level is divided into low risk, medium risk and high risk according to the degree of matching;

[0045] The low-risk determination criteria include:

[0046] When the verification results of the basic physiological indicators, the dynamic changes of indicators, the multi-indicator correlation, and the population feature adaptation are all in a normal or critical state, and no abnormalities or computational errors are triggered, the risk is determined to be low.

[0047] The criteria for determining medium risk include:

[0048] If any single dimension verification result of the basic physiological indicator dimension verification result, indicator dynamic change dimension verification result, multi-indicator correlation dimension verification result, and population feature adaptation dimension verification result is abnormal, and the duration of the abnormal state does not exceed the set time, or if there is a triggering or calculation error, it is judged as medium risk.

[0049] The high-risk determination criteria include:

[0050] If any single dimension verification result of the basic physiological indicator dimension verification result, indicator dynamic change dimension verification result, multi-indicator correlation dimension verification result, and population feature adaptation dimension verification result is abnormal, and the abnormal state lasts for more than the set time, or triggers an abnormal operation, it is judged as high risk.

[0051] As a preferred embodiment of the intelligent physical examination offline data collection method of the present invention, when the verification result meets the preset chronic disease screening judgment criteria, the collection terminal prompts the primary medical staff through sound and light signals to include the physical examination examinee with the identity token corresponding to the preset chronic disease screening judgment criteria into the follow-up management of key populations.

[0052] When the chronic disease risk level reaches or exceeds the preset risk threshold, it is determined that the preset chronic disease screening criteria are met.

[0053] The criteria for determining the preset risk threshold include:

[0054] When the risk assessment conditions are met, if the basic physiological indicator dimension verification result is abnormal and the multi-indicator correlation dimension verification result is abnormal, or the basic physiological indicator dimension verification result, the indicator dynamic change dimension verification result, and the multi-indicator correlation dimension verification result all show a critical state, then it is determined that the preset risk threshold has been reached.

[0055] When the high-risk criteria are met, the risk threshold is directly determined to have been reached.

[0056] The audio-visual signals include a preset frequency buzzing tone and a combination of flashing lights of different colors;

[0057] The preset frequency of the buzzer alert tone is positively correlated with the chronic disease risk level, including:

[0058] Low risk corresponds to low frequency, medium risk corresponds to medium frequency, and high risk corresponds to high frequency.

[0059] The color of the flashing light is set to green, yellow, or red according to the risk level, and the flashing frequency increases as the risk level increases.

[0060] As a preferred embodiment of the intelligent physical examination offline data collection method described in this invention, the structured data file is encrypted in blocks by the collection terminal using a lightweight encryption algorithm and stored in the terminal's local secure storage area;

[0061] Metadata files are generated through double hash verification and stored in association with structured data files, including:

[0062] When the data acquisition terminal receives raw physiological index data, it simultaneously uses the MD5 algorithm to calculate the first hash value, which serves as the data fingerprint of the raw physiological index data.

[0063] By using the security encryption module built into the acquisition terminal, the SHA-256 algorithm is used to perform a hash operation on the encrypted structured data file to generate a second hash value of fixed length.

[0064] The first hash value and the second hash value are associated and stored as a basis for data consistency verification.

[0065] As a preferred embodiment of the intelligent physical examination offline data collection method described in this invention, the method includes: verifying the consistency of the original physiological indicator data and performing double verification to generate metadata files, including:

[0066] Decrypt the encrypted structured data file, extract the original physiological indicator data, and recalculate the MD5 value of the original physiological indicator data as the third hash value;

[0067] The third hash value is compared with the first hash value generated during data collection to verify the integrity of the original physiological indicator data.

[0068] The integrity of the encrypted file is verified by comparing the current SHA-256 hash value of the encrypted structured data file with the stored second hash value.

[0069] The metadata file includes a double verification result, which is generated through double verification and includes:

[0070] If the third hash value is consistent with the first hash value, and the current SHA-256 hash value is consistent with the second hash value, then the double verification result is that the original data is complete and the encrypted file is complete, and a metadata file that has passed the verification is generated.

[0071] If the third hash value is inconsistent with the first hash value, and the current SHA-256 hash value is inconsistent with the second hash value, then the double verification result is "the original data is incomplete and the encrypted file is incomplete", generating a metadata file that fails the verification.

[0072] If the third hash value is inconsistent with the first hash value, but the current SHA-256 hash value is consistent with the second hash value, then the result of the double verification is that the original data is incomplete, and a metadata file that fails the verification is generated.

[0073] If the third hash value is consistent with the first hash value, but the current SHA-256 hash value is inconsistent with the second hash value, then the result of the double verification is that the original data is complete, and a metadata file that fails the verification is generated.

[0074] The metadata file includes the double verification result, the first hash value, the second hash value, the third hash value, the current SHA-256 hash value, and the comparison timestamp. It establishes a one-to-one correspondence with the structured data file through the unique encoding of the physical examination device and the collection timestamp, and is stored independently in encrypted form.

[0075] The beneficial effects of this invention are: overcoming network limitations to achieve stable data collection in environments with no or weak network coverage; ensuring data security and integrity through identity token binding and encrypted storage; and making data tampering traceable through dual hash verification. The built-in lightweight decision engine analyzes physiological indicators from multiple dimensions to accurately determine chronic disease risks. The intuitive audio-visual signals simplify complex data collection, helping medical personnel quickly identify key populations, significantly improving the efficiency of physical examinations in grassroots and remote areas, strengthening early screening capabilities for chronic diseases, and enabling smart healthcare to benefit more scenarios, providing an efficient, safe, and intelligent solution for health management. Attached Figure Description

[0076] Figure 1 This is a flowchart illustrating the steps of an offline data collection method for intelligent physical examination provided in one embodiment of the present invention. Detailed Implementation

[0077] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0078] Example, refer to Figure 1 As an embodiment of the present invention, a smart physical examination offline data collection method is provided, comprising the following steps:

[0079] Step S1: The identification information is identified and extracted by the data collection terminal to generate an identity token, and the identity token is synchronized to the physical examination device.

[0080] Step S2: Collect raw physiological indicator data through physical examination equipment, generate structured data files, encrypt the structured data files, and store them;

[0081] Step S3: Calculate the hash values ​​of the encrypted structured data file and the original physiological indicator data respectively, compare and verify them to form a metadata file, and store the metadata file and the structured data file together.

[0082] In one embodiment, during offline data collection for smart health checkups, the collection terminal identifies and extracts identification information to generate an identity token, which is then synchronized to the health checkup device. Subsequently, the health checkup device collects raw physiological indicator data and generates a structured data file. The hash values ​​of the encrypted structured data file and the raw physiological indicator data are calculated and compared for verification, forming a metadata file. The metadata file and the structured data file are then stored together, achieving accurate binding of the examinee's identity, encrypted data storage, and integrity verification, effectively improving the security and reliability of offline health checkup data collection.

[0083] The identity token is obtained by identifying the examinee's medical insurance card or ID card through the data collection terminal;

[0084] The data collection terminal establishes a short-range communication link with one or more medical examination devices via Bluetooth or NFC, and synchronizes the identity token to the medical examination device to form a binding relationship between the examinee's identity and the medical examination device cluster.

[0085] The medical examination equipment collects raw physiological data in environments with no or weak network access.

[0086] The raw physiological data are used to generate structured data files according to preset medical data standards;

[0087] The structured data file includes physiological indicator data, collection timestamps, unique codes of medical examination equipment, and identity token hash values.

[0088] In one embodiment, the identity token is acquired by the data acquisition terminal identifying the examinee's medical insurance card or ID card. For example, the terminal reads the identity information in the card chip through a built-in card reader, generates a unique identity token through hash calculation, and enables Bluetooth (operating in the 2.4GHz band, communication distance set to 0-10 meters) or NFC (using ISO 14443 standard, communication distance controlled within 0-10 centimeters) to establish a short-range communication link with one or more medical examination devices, such as a blood pressure monitor and an electrocardiograph, and synchronizes the identity token to each medical examination device, forming a clear binding relationship, for example, examinee Zhang San - blood pressure monitor #001, electrocardiograph #005. In environments with no network or weak network speeds below 200kbps, the medical examination devices normally collect raw physiological index data of blood pressure and heart rate, and collect the data according to the preset FHIR medical data standard (based on HL7). The V2 / V3 extension generates a structured data file, which includes specific measurement values ​​(such as blood pressure 120 / 80 mmHg), a collection timestamp accurate to milliseconds, a unique device code (such as ECG-2023-001), and an identity token hash value. This method ensures the accurate association between identity and device in offline scenarios. The structured data conforms to medical industry standards, providing a consistent foundation for subsequent data sharing and analysis. At the same time, short-range communication reduces the risk of leakage during data transmission.

[0089] The data acquisition terminal has a built-in lightweight decision engine;

[0090] The lightweight decision engine has built-in preset chronic disease screening criteria. When physiological indicator data is acquired, it calls the screening logic corresponding to the physiological indicator data, performs real-time verification according to the chronic disease screening criteria, and outputs the verification results.

[0091] The screening logic includes:

[0092] Feature parameter extraction from physiological indicator data;

[0093] The characteristic parameters include the measured values ​​of physiological indicators, the amplitude of abnormal fluctuations in indicators, the trend parameters of indicator changes, the ratio parameters of multiple related indicators, and the parameters of indicator measurement time intervals;

[0094] Multiple correlation indicators include combinations of blood glucose and insulin correlation indicators, combinations of blood pressure and blood lipid correlation indicators, and combinations of body mass index and waist circumference correlation indicators;

[0095] Determine the validation dimensions and threshold ranges for the physiological indicator data;

[0096] The verification dimensions include basic physiological indicators, dynamic changes of indicators, multi-indicator correlation, and population feature adaptation.

[0097] The threshold ranges for the verification dimensions include the threshold ranges for basic physiological indicators, the threshold ranges for dynamic changes in indicators, the threshold ranges for multi-indicator correlation, and the threshold ranges for population feature adaptation.

[0098] The extracted feature parameters are compared with the threshold range of the verification dimension, and logical verification operations are performed, including:

[0099] The feature parameters are matched with the verification dimensions respectively;

[0100] Once the matching is complete, the feature parameters are compared with the threshold range of the corresponding verification dimension to determine the threshold compliance of the feature parameters.

[0101] Based on the threshold compliance, single-dimensional verification results are generated for the basic physiological indicators dimension, the dynamic change of indicators dimension, the multi-indicator correlation dimension, and the population feature adaptation dimension, including verification results for the basic physiological indicators dimension, the dynamic change of indicators dimension, the multi-indicator correlation dimension, and the population feature adaptation dimension.

[0102] The verification results of the basic physiological indicators, the dynamic changes of indicators, the multi-indicator correlation, and the population feature adaptation are respectively divided into normal state, borderline state, and abnormal state according to the threshold compliance.

[0103] In one embodiment, the lightweight decision engine of the data acquisition terminal incorporates screening criteria based on guidelines for chronic disease prevention and control. After acquiring physiological indicator data, it automatically initiates the screening logic for the corresponding disease: First, it extracts feature parameters, such as for diabetes screening, extracting fasting blood glucose measurement values ​​(e.g., 6.5 mmol / L), the amplitude of the last three blood glucose fluctuations (2.0 mmol / L), the weekly blood glucose trend (continuously rising), the blood glucose to insulin ratio (8.2), and the measurement interval (12 hours), while also including the ratio parameters of correlation indicators such as blood pressure and blood lipids, and body mass index and waist circumference; then, it determines four verification dimensions and threshold ranges. In the basic physiological indicator dimension, the normal threshold for fasting blood glucose is 3.9-6.1 mmol / L, the borderline is 6.1-6.9 mmol / L, and the abnormal threshold is greater than or equal to 7.0 mmol / L. In the indicator dynamic change dimension, the normal blood glucose fluctuation is less than or equal to 1.4 mmol / L, the borderline is 1.4-2.2 mmol / L, and the abnormal threshold is greater than or equal to 1.4 mmol / L. The fasting blood glucose threshold for individuals aged 65 and above was relaxed to 6.1-7.2 mmol / L. The data was analyzed using a multi-indicator correlation dimension, where the body mass index (BMI) to waist circumference ratio was normally less than 0.3, borderline 0.3-0.35, and abnormal greater than or equal to 0.35. Logical verification was then performed, matching the extracted feature parameters with the thresholds for each dimension. A fasting blood glucose level of 6.5 mmol / L was borderline in the baseline dimension, a fluctuation range of 2.0 mmol / L was borderline in the dynamic change dimension, and a blood glucose to insulin ratio of 8.2 was abnormal in the multi-indicator correlation dimension. Combined with age, this was considered normal in the population characteristic adaptation dimension. Finally, single-dimensional verification results were generated for each dimension. This process, through standardized parameter extraction and multi-dimensional threshold comparison, achieved real-time and accurate screening of chronic disease risks, providing clear indicators of abnormalities for primary healthcare workers and enhancing the immediate application value of physical examination data in offline scenarios.

[0104] Based on a preset logical operation strategy, the single-dimensional verification results are combined and operated to generate a combined operation result;

[0105] Logical operation strategies include AND operation, OR operation, and priority determination;

[0106] The results of combined operations include normal operation, abnormal operation, and abnormal AND operation.

[0107] Normal operation includes:

[0108] When the verification results of the basic physiological indicators dimension, the dynamic change of indicators dimension, the multi-indicator correlation dimension, and the population characteristic adaptation dimension are all in a normal state, and no critical or abnormal state appears, the operation is judged to be normal.

[0109] Or, operational anomalies include:

[0110] When any single dimension verification result among the basic physiological indicator dimension verification result, indicator dynamic change dimension verification result, multi-indicator correlation dimension verification result, and population feature adaptation dimension verification result is in an abnormal or critical state, it is determined to be an OR operation abnormality.

[0111] AND operation exceptions include:

[0112] When the verification results of the basic physiological indicators dimension, the dynamic change of indicators dimension, the multi-indicator correlation dimension, and the population feature adaptation dimension are all in an abnormal state, it is determined that the AND operation is abnormal.

[0113] In one embodiment, the lightweight decision engine performs combined calculations on the single-dimensional verification results based on a preset logical operation strategy. When the verification results of the basic physiological indicator dimension, the indicator dynamic change dimension, the multi-indicator correlation dimension, and the population characteristic adaptation dimension are all in a normal state (e.g., blood pressure 120 / 80 mmHg, blood glucose fluctuation 1.0 mmol / L, body mass index 22.5 to waist circumference 80cm ratio 0.28, and blood pressure threshold adaptation for 60-year-olds are normal), and there are no critical or abnormal states, the calculation is judged to be normal. If any dimension of the basic physiological indicator dimension, the indicator dynamic change dimension, the multi-indicator correlation dimension, or the population characteristic adaptation dimension is in a critical or abnormal state (e.g., the basic physiological indicator dimension of systolic blood pressure 135 mmHg is in a critical state), the calculation is judged to be normal. If the body mass index (BMI) to waist circumference ratio of 0.36 is considered abnormal (e.g., abnormal state), then an OR operation is triggered. If all dimensions of the basic physiological indicators, dynamic changes of indicators, multi-indicator correlation, and population characteristic adaptation are abnormal (e.g., systolic blood pressure 160 mmHg, blood glucose fluctuation 3.0 mmol / L, blood pressure to blood lipid ratio 4.5, and population adaptation dimension fails), then an AND operation is determined to be abnormal. Through the above strategies, the engine can integrate scattered single-dimensional results into combined operation results with clinical guidance significance, providing primary healthcare personnel with a tiered risk assessment basis, significantly improving the processing efficiency of complex indicator combinations in offline screening, avoiding the problems of misjudgment of single indicators or lag in multi-indicator correlation analysis, and achieving rapid and accurate classification of chronic disease risks.

[0114] Priority determination includes:

[0115] The results of the verification of basic physiological indicators have higher priority than the results of the verification of dynamic changes in indicators.

[0116] The results of dynamic change dimension verification have higher priority than the results of multi-indicator correlation dimension verification.

[0117] The results of multi-indicator correlation dimension verification have higher priority than the results of population feature adaptation dimension verification.

[0118] When single-dimensional verification results conflict, the higher-priority single-dimensional verification result is used for combined calculation.

[0119] The verification result is determined based on the combined calculation result, which represents the degree of matching between the feature parameters and the preset chronic disease screening criteria.

[0120] The chronic disease risk level is determined based on the verification results;

[0121] Chronic disease risk levels are classified into low risk, medium risk, and high risk based on the degree of matching.

[0122] Low-risk criteria include:

[0123] When the verification results of the basic physiological indicators, the dynamic changes of indicators, the multi-indicator correlation, and the population characteristic adaptation are all in a normal or critical state, and there are no triggers or abnormal operations, it is judged as low risk.

[0124] The criteria for determining medium risk include:

[0125] If any single dimension verification result of the basic physiological indicator dimension verification result, the indicator dynamic change dimension verification result, the multi-indicator correlation dimension verification result, and the population characteristic adaptation dimension verification result is abnormal, and the duration of the abnormal state does not exceed the set time, or if the trigger or calculation is abnormal, it is judged as medium risk.

[0126] High-risk determination criteria include:

[0127] If any single dimension verification result of the basic physiological indicator dimension verification result, the indicator dynamic change dimension verification result, the multi-indicator correlation dimension verification result, or the population characteristic adaptation dimension verification result is abnormal, and the abnormal state lasts for more than the set time, or if an abnormality is triggered or calculated, it is judged as high risk.

[0128] In one embodiment, the lightweight decision engine prioritizes the results of basic physiological indicator verification as the first priority, followed by the results of dynamic indicator change verification as the second priority, multi-indicator correlation verification as the third priority, and population feature adaptation verification as the fourth priority. When single-dimensional verification results conflict, the higher priority takes precedence. For example, if the basic physiological indicator dimension is judged as abnormal while the population feature adaptation dimension is judged as normal, the basic physiological indicator dimension result is used for combined calculation. Based on this, the engine classifies chronic disease risk levels according to the combined calculation results: low risk means that the results of basic physiological indicator verification, dynamic indicator change verification, multi-indicator correlation verification, and population feature adaptation verification are all normal or borderline (e.g., basic physiological indicator borderline, dynamic change normal, multi-indicator correlation borderline, and population adaptation normal). The following criteria define risk levels: Medium risk refers to an abnormal result in any single dimension of the following dimensions (basic physiological indicators, dynamic changes in indicators, multi-indicator correlation, and population characteristic adaptation) that does not trigger an error or is abnormal (e.g., abnormal basic physiological indicators lasting 3 days); High risk refers to an abnormal result in any dimension of the following dimensions (basic physiological indicators, dynamic changes in indicators, multi-indicator correlation, and population characteristic adaptation) that lasts more than 7 days, or an abnormal AND operation (e.g., all dimensions are abnormal). This mechanism achieves precise stratification by setting priorities and combining them with risk classification, ensuring that primary healthcare workers can quickly identify high-priority abnormalities and risk levels in offline scenarios, thereby improving the timeliness and targeting of chronic disease intervention.

[0129] When the verification result meets the preset criteria for chronic disease screening, the data collection terminal prompts primary healthcare workers through audio and visual signals to include the examinee with the identity token corresponding to the preset criteria for chronic disease screening into the follow-up management of key populations.

[0130] When the risk level of chronic disease reaches or exceeds the preset risk threshold, it is determined that the preset chronic disease screening criteria are met.

[0131] The criteria for determining the preset risk threshold include:

[0132] When the risk assessment conditions are met, if the basic physiological indicator dimension verification result is abnormal and the multi-indicator correlation dimension verification result is abnormal, or the basic physiological indicator dimension verification result, the indicator dynamic change dimension verification result, and the multi-indicator correlation dimension verification result all show a critical state, then it is determined that the preset risk threshold has been reached.

[0133] When the high-risk criteria are met, the risk threshold is directly determined to have been reached.

[0134] The audio-visual signals include a preset frequency beeping tone and a combination of flashing lights of different colors;

[0135] The preset frequency of the beeping alert tone is positively correlated with the chronic disease risk level, including:

[0136] Low risk corresponds to low frequency, medium risk corresponds to medium frequency, and high risk corresponds to high frequency.

[0137] The color of the flashing light is set to green, yellow, or red according to the risk level, and the flashing frequency increases as the risk level increases.

[0138] In one embodiment, when the chronic disease risk level reaches or exceeds a preset risk threshold, the data collection terminal prompts primary healthcare personnel via audio-visual signals to include examinees with identity tokens corresponding to the preset chronic disease screening criteria into key follow-up management. The specific judgment logic is as follows: if the risk level is medium risk and simultaneously meets the following criteria: abnormalities in basic physiological indicators (e.g., systolic blood pressure 145 mmHg) and abnormalities in multiple indicator correlation dimensions (body mass index to waist circumference ratio 0.36), or if the three dimensions of basic physiological indicators, dynamic changes in indicators, and multiple indicator correlation are all at a critical state (e.g., blood pressure critically 135 mmHg, blood glucose fluctuation critically 1.8 mmol / L, blood pressure to blood lipid ratio critically 3.8), then the preset risk threshold is reached; if the risk level is high risk (e.g., any...), then the preset risk threshold is reached. If an indicator remains abnormal for more than 7 days or is abnormal across all dimensions, the threshold is directly triggered. At this time, the terminal provides real-time prompts via audio and visual signals: low risk corresponds to a green light (flashing frequency 5 times / minute) and a low-frequency beep (100Hz), medium risk corresponds to a yellow light (10 times / minute) and a medium-frequency beep (300Hz), and high risk corresponds to a red light (20 times / minute) and a high-frequency beep (500Hz). This mechanism uses differentiated audio and visual signals to intuitively present the risk level, helping grassroots staff quickly locate examinees who need key management (such as linking the abnormal results of examinee Li Si - blood pressure monitor #003 through an identity token), significantly improving the timeliness and accuracy of chronic disease follow-up management in offline scenarios, and avoiding intervention delays caused by information lag.

[0139] The structured data files are encrypted in blocks by the acquisition terminal using a lightweight encryption algorithm and stored in the terminal's local secure storage area;

[0140] Metadata files are generated through double hash verification and stored in association with structured data files, including:

[0141] When the data acquisition terminal receives raw physiological index data, it simultaneously uses the MD5 algorithm to calculate the first hash value, which serves as the data fingerprint of the raw physiological index data.

[0142] By using the security encryption module built into the acquisition terminal, the SHA-256 algorithm is used to perform a hash operation on the encrypted structured data file to generate a second hash value of fixed length.

[0143] The first hash value and the second hash value are associated and stored as a basis for data consistency verification.

[0144] The consistency of the original physiological indicator data was verified, and a dual-check was performed to generate metadata files, including:

[0145] Decrypt the encrypted structured data file, extract the original physiological index data, and recalculate the MD5 value of the original physiological index data as the third hash value;

[0146] The third hash value is compared with the first hash value generated during data collection to verify the integrity of the original physiological indicator data.

[0147] The integrity of the encrypted file is verified by comparing the current SHA-256 hash value of the encrypted structured data file with the stored second hash value.

[0148] The metadata file includes the results of the double-checked verification, which are generated through double-checking and include:

[0149] If the third hash value matches the first hash value, and the current SHA-256 hash value matches the second hash value, then the double verification result is that the original data is complete and the encrypted file is complete, and a metadata file that has passed the verification is generated.

[0150] If the third hash value is inconsistent with the first hash value, and the current SHA-256 hash value is inconsistent with the second hash value, then the double verification result is "the original data is incomplete and the encrypted file is incomplete", generating a metadata file that fails the verification.

[0151] If the third hash value is inconsistent with the first hash value, but the current SHA-256 hash value is consistent with the second hash value, then the result of the double verification is that the original data is incomplete, and a metadata file that fails the verification is generated.

[0152] If the third hash value is the same as the first hash value, but the current SHA-256 hash value is different from the second hash value, then the result of the double verification is that the original data is complete, and a metadata file that fails the verification is generated.

[0153] The metadata file includes the double verification result, the first hash value, the second hash value, the third hash value, the current SHA-256 hash value, and the comparison timestamp. It establishes a one-to-one correspondence with the structured data file through the unique encoding of the physical examination device and the collection timestamp, and is stored independently in encrypted form.

[0154] In one embodiment, the structured data file achieves end-to-end security protection through lightweight encryption and double hash verification. The acquisition terminal uses the AES-128 algorithm to encrypt the structured data file containing physiological indicators such as blood pressure and blood sugar in blocks, and stores it in a local secure storage area based on a Trusted Execution Environment (TEE). Simultaneously, during the data acquisition phase, the MD5 algorithm is used to calculate the first hash value of the original physiological indicator data as a data fingerprint, and the SHA-256 algorithm is called by the security encryption module to generate a second hash value for the encrypted file. The two are then stored together to form a metadata file. During data verification, the terminal first decrypts the structured data file, recalculates the MD5 value of the original data as a third hash value, and compares it with the first hash value to verify the integrity of the original data. The system simultaneously calculates the current SHA-256 value of the encrypted file and compares it with the second hash value to verify the integrity of the encrypted file. For example, if the third hash value matches the first hash value and the current SHA-256 value matches the second hash value, the dual verification result indicates that both the original data and the encrypted file are complete, and a metadata file including a verification timestamp and a unique device code (such as BP-2023-001) is generated. If either hash value is inconsistent, a verification result indicating that either the original data or the encrypted file is incomplete is generated based on the conflict type. This mechanism prevents data leakage through block encryption and achieves bidirectional consistency verification between the original data and the encrypted file through dual hash verification, effectively resisting the risk of data tampering in offline scenarios. It meets the requirements for medical data storage security in the "Guideline for Health and Medical Data Security" (WS / T 743-2021) and provides underlying protection for the trustworthiness of subsequent data uploads to the cloud.

[0155] This invention overcomes network limitations to achieve stable data collection in offline and weak network environments. Through identity token binding and encrypted storage, it ensures data security and integrity, while dual hash verification makes data tampering traceable. A built-in lightweight decision engine analyzes physiological indicators from multiple dimensions to accurately determine chronic disease risk. Intuitive audio-visual signals simplify complex data collection, helping medical personnel quickly identify key populations, significantly improving the efficiency of physical examinations in grassroots and remote areas, strengthening early screening capabilities for chronic diseases, and enabling smart healthcare to benefit more scenarios, providing an efficient, safe, and intelligent solution for health management.

[0156] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0157] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for offline data collection in intelligent health checkups, characterized in that, Includes the following steps: Step S1: The identification information is identified and extracted by the acquisition terminal to generate an identity token, and the identity token is synchronized to the physical examination device. Step S2: Collect raw physiological indicator data through the physical examination equipment, generate a structured data file, encrypt the structured data file, and store it; Step S3: Calculate the hash values ​​of the encrypted structured data file and the original physiological indicator data respectively, compare and verify them to form a metadata file, and store the metadata file and the structured data file together. The identity token is obtained by the data collection terminal by identifying the examinee's medical insurance card or ID card; The data collection terminal establishes a short-range communication link with one or more medical examination devices via Bluetooth or NFC, and synchronizes the identity token to the medical examination device to form a binding relationship between the examinee's identity and the medical examination device cluster. The physical examination equipment collects raw physiological index data in environments with no or weak network coverage. The raw physiological index data are generated into a structured data file according to a preset medical data standard; The structured data file includes physiological indicator data, collection timestamps, unique codes of the physical examination equipment, and identity token hash values; The structured data file is encrypted in blocks by the acquisition terminal using a lightweight encryption algorithm and stored in the terminal's local secure storage area; Metadata files are generated through double hash verification and stored in association with structured data files, including: When the data acquisition terminal receives raw physiological index data, it simultaneously uses the MD5 algorithm to calculate the first hash value, which serves as the data fingerprint of the raw physiological index data. By using the security encryption module built into the acquisition terminal, the SHA-256 algorithm is used to perform a hash operation on the encrypted structured data file to generate a second hash value of fixed length. The first hash value and the second hash value are associated and stored as a basis for data consistency verification.

2. The intelligent physical examination offline data collection method as described in claim 1, characterized in that: The data acquisition terminal has a built-in lightweight decision engine; The lightweight decision engine has a built-in preset chronic disease screening and judgment standard. When physiological indicator data is acquired, the screening logic corresponding to the physiological indicator data is called, and real-time verification is performed according to the chronic disease screening and judgment standard, and the verification result is output. The screening logic includes: Feature parameters are extracted from the physiological index data; The characteristic parameters include physiological indicator measurements, abnormal fluctuation range of indicators, indicator trend parameters, ratio parameters of multiple related indicators, and indicator measurement time interval parameters. The multiple correlation indicators include the combination of blood glucose and insulin correlation indicators, the combination of blood pressure and blood lipid correlation indicators, and the combination of body mass index and waist circumference correlation indicators. Determine the verification dimensions and threshold ranges for the physiological indicator data; The verification dimensions include basic physiological indicators, dynamic changes of indicators, multi-indicator correlation, and population feature adaptation. The threshold ranges for the verification dimensions include the threshold ranges for basic physiological indicators, the threshold ranges for dynamic changes in indicators, the threshold ranges for multi-indicator correlation, and the threshold ranges for population feature adaptation.

3. The intelligent physical examination offline data collection method as described in claim 2, characterized in that: The extracted feature parameters are compared with the verification dimension threshold range, and logical verification operations are performed, including: The feature parameters are matched with the verification dimensions respectively; Once the matching is complete, the feature parameters are compared with the threshold range of the corresponding verification dimension to determine the threshold compliance of the feature parameters. Based on the threshold compliance, single-dimensional verification results are generated for the basic physiological indicator dimension, indicator dynamic change dimension, multi-indicator correlation dimension, and population feature adaptation dimension, respectively, including the verification results for the basic physiological indicator dimension, the verification results for the indicator dynamic change dimension, the verification results for the multi-indicator correlation dimension, and the verification results for the population feature adaptation dimension. The verification results of the basic physiological indicators, the dynamic changes of indicators, the multi-indicator correlation, and the population feature adaptation are respectively divided into normal state, critical state, and abnormal state according to the threshold compliance.

4. The intelligent physical examination offline data acquisition method as described in claim 3, characterized in that: Based on a preset logical operation strategy, the single-dimensional verification results are combined to generate a combined operation result. The logical operation strategy includes AND operation, OR operation, and priority determination; The combined operation results include normal operation, abnormal operation, and abnormal AND operation. The normal operation includes: When the verification results of the basic physiological indicators, the dynamic changes of indicators, the multi-indicator correlation, and the population feature adaptation are all in a normal state, and no critical or abnormal state occurs, the operation is judged to be normal. The OR operation exception includes: When any single-dimensional verification result among the basic physiological indicator dimension verification result, indicator dynamic change dimension verification result, multi-indicator correlation dimension verification result, and population feature adaptation dimension verification result is in an abnormal or critical state, it is determined to be an OR operation abnormality. The AND operation exceptions include: When the verification results of the basic physiological indicator dimension, the indicator dynamic change dimension, the multi-indicator correlation dimension, and the population feature adaptation dimension are all in an abnormal state, it is determined that the AND operation is abnormal.

5. The intelligent physical examination offline data acquisition method as described in claim 4, characterized in that: The priority determination includes: The verification results of the basic physiological indicators have a higher priority than the verification results of the dynamic changes in the indicators. The verification results of the dynamic change dimension of the indicator have higher priority than the verification results of the multi-indicator correlation dimension. The multi-indicator correlation dimension verification result has a higher priority than the population feature adaptation dimension verification result. When single-dimensional verification results conflict, the higher-priority single-dimensional verification result is used for combined calculation. The verification result is determined based on the combined calculation result, which represents the degree of matching between the feature parameters and the preset chronic disease screening judgment criteria.

6. The intelligent physical examination offline data acquisition method as described in claim 5, characterized in that: The chronic disease risk level is determined based on the verification results; The chronic disease risk level is divided into low risk, medium risk and high risk according to the degree of matching; The low-risk determination criteria include: When the verification results of the basic physiological indicators, the dynamic changes of indicators, the multi-indicator correlation, and the population feature adaptation are all in a normal or critical state, and no abnormalities or computational errors are triggered, the risk is determined to be low. The criteria for determining medium risk include: If any single dimension verification result of the basic physiological indicator dimension verification result, indicator dynamic change dimension verification result, multi-indicator correlation dimension verification result, and population feature adaptation dimension verification result is abnormal, and the duration of the abnormal state does not exceed the set time, or if there is a triggering or calculation error, it is judged as medium risk. The high-risk determination criteria include: If any single dimension verification result of the basic physiological indicator dimension verification result, indicator dynamic change dimension verification result, multi-indicator correlation dimension verification result, and population feature adaptation dimension verification result is abnormal, and the abnormal state lasts for more than the set time, or triggers an abnormal operation, it is judged as high risk.

7. The intelligent physical examination offline data acquisition method as described in claim 6, characterized in that: When the verification result meets the preset chronic disease screening criteria, the data acquisition terminal prompts primary healthcare workers through sound and light signals to include the examinee with the identity token corresponding to the preset chronic disease screening criteria into the key population follow-up management. When the chronic disease risk level reaches or exceeds the preset risk threshold, it is determined that the preset chronic disease screening criteria are met. The criteria for determining the preset risk threshold include: When the risk assessment conditions are met, if the basic physiological indicator dimension verification result is abnormal and the multi-indicator correlation dimension verification result is abnormal, or the basic physiological indicator dimension verification result, the indicator dynamic change dimension verification result, and the multi-indicator correlation dimension verification result all show a critical state, then it is determined that the preset risk threshold has been reached. When the high-risk criteria are met, the risk threshold is directly determined to have been reached. The audio-visual signals include a preset frequency buzzing tone and a combination of flashing lights of different colors; The preset frequency of the buzzer alert tone is positively correlated with the chronic disease risk level, including: Low risk corresponds to low frequency, medium risk corresponds to medium frequency, and high risk corresponds to high frequency. The color of the flashing light is set to green, yellow, or red according to the risk level, and the flashing frequency increases as the risk level increases.

8. The intelligent physical examination offline data acquisition method as described in claim 7, characterized in that: The consistency of the original physiological indicator data was verified, and a dual-check was performed to generate metadata files, including: Decrypt the encrypted structured data file, extract the original physiological indicator data, and recalculate the MD5 value of the original physiological indicator data as the third hash value; The third hash value is compared with the first hash value generated during data collection to verify the integrity of the original physiological indicator data. The integrity of the encrypted file is verified by comparing the current SHA-256 hash value of the encrypted structured data file with the stored second hash value. The metadata file includes a double verification result, which is generated through double verification and includes: If the third hash value is consistent with the first hash value, and the current SHA-256 hash value is consistent with the second hash value, then the double verification result is that the original data is complete and the encrypted file is complete, and a metadata file that has passed the verification is generated. If the third hash value is inconsistent with the first hash value, and the current SHA-256 hash value is inconsistent with the second hash value, then the double verification result is "the original data is incomplete and the encrypted file is incomplete", generating a metadata file that fails the verification. If the third hash value is inconsistent with the first hash value, but the current SHA-256 hash value is consistent with the second hash value, then the result of the double verification is that the original data is incomplete, and a metadata file that fails the verification is generated. If the third hash value is consistent with the first hash value, but the current SHA-256 hash value is inconsistent with the second hash value, then the result of the double verification is that the original data is complete, and a metadata file that fails the verification is generated. The metadata file includes the double verification result, the first hash value, the second hash value, the third hash value, the current SHA-256 hash value, and the comparison timestamp. It establishes a one-to-one correspondence with the structured data file through the unique encoding of the physical examination device and the collection timestamp, and is stored independently in encrypted form.

Images