Nursing quality risk assessment system based on medical big data

Abstract

The invention discloses a nursing quality risk assessment system based on medical big data, and relates to the technical field of nursing quality risk assessment, and the system comprises a data collection module, a dynamic assessment module, a dynamic prediction module and a closed-loop management module. A dynamic risk prediction value in the future 6 hours is calculated through combined calculation, 'gate forward movement 'of risk management is realized, meanwhile, by calculating a nursing manpower scheduling value, nurses are allocated in advance according to the dynamic risk prediction value, efficient utilization of nursing resources is improved, a differential treatment strategy is matched based on N-value three-level early warning (yellow / orange / red), and the risk management efficiency is improved. Through real-time correction of error prediction and dynamic adjustment of model parameters, the system continuously adapts to clinical scene changes, and the timeliness of risk management is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of nursing quality risk assessment, and in particular to a nursing quality risk assessment system based on medical big data. Background Art

[0002] With the rapid advancement of information technology, the healthcare sector is experiencing an unprecedented explosion of data. This growth is primarily driven by diverse data sources such as electronic medical records, medical imaging, and genetic sequences, which together form the foundation of healthcare big data. This massive amount of data not only deepens our understanding of human health and disease mechanisms but also provides unprecedented depth and breadth of information for assessing nursing quality risks. Advanced data analytics technologies enable more comprehensive and accurate identification of potential risk factors in the nursing process, providing scientific, data-driven support for nursing decision-making and significantly improving nursing quality and patient safety.

[0003] However, despite the promising application prospects of medical big data, current nursing quality risk assessment systems still have significant shortcomings in terms of real-time and dynamic performance. Existing methods mostly rely on retrospective analysis, which involves analyzing historical data to identify risk events and their causes. While this approach can provide some risk insights, it cannot capture dynamic risk signals during the nursing process, such as sudden changes in a patient's vital signs. This lag limits the effectiveness of risk warning systems and makes it difficult to achieve "front-end" risk management, that is, to effectively prevent and control risks before they actually occur.

[0004] In addition, the existing system is disconnected in the feedback control mechanism between risk analysis results and nursing process optimization. Risk analysis often remains at the reporting level and lacks an automated closed-loop management system that is closely integrated with nursing practice. This means that even if high-risk scenarios are identified (such as increased nursing needs in specific patient groups and tight nursing resources), they cannot be quickly converted into actual nursing process adjustments or resource reallocation strategies. This disconnect not only weakens the timeliness of risk management, but also limits the efficient use of nursing resources, affecting the improvement of overall nursing quality. Summary of the Invention

[0005] In response to the above situation, the present invention dynamically evaluates the comprehensive evaluation value of patient nursing quality in the ward, the equipment risk value, and the text risk value, and jointly calculates the dynamic risk prediction value within the next 6 hours. Nurses are deployed in advance according to the dynamic risk prediction value, and dynamic early warning and iterative optimization are carried out at the same time, thereby improving the timeliness of risk management and the efficient use of nursing resources.

[0006] The technical solution includes a data acquisition module, a dynamic evaluation module, a dynamic prediction module and a closed-loop management module. The data acquisition module collects patient vital sign data A, nursing operation records B, medication information C, equipment alarm logs D, and nursing text reports E in real time. The dynamic evaluation module extracts the heart rate A1, blood pressure A2, and blood oxygen saturation A3 of the patient's vital signs data A, the nursing frequency B1, nursing time B2, and nursing content B3 of the nursing operation record B, and the drug type C1, dosage C2, and medication time C3 of the medication information C, calculates the patient's vital signs evaluation value A4, the nursing operation record evaluation value B4, and the medication information evaluation value C4, and finally calculates the comprehensive evaluation value M, M= A4×0.2+B4×0.2+ C4×0.4; The dynamic prediction module extracts the alarm type D1, alarm frequency D2, and alarm duration D3 from the device alarm log D, calculates the device risk value D4, extracts the treatment measure compliance E1 and the number of abnormal event marks E2 from the nursing text report E, and calculates the text risk value E3; The closed-loop management module calculates the dynamic risk prediction value N(t+6 hours) in the next 6 hours = 0.4×M(t)+0.3×D4(t)+0.2×E3(t)+0.1×M1, where t represents the current time value and M1 is the real-time change rate of the comprehensive assessment value M; When N (t + 6 hours) is greater than 0.7, a list of high-risk patients is generated and pushed to the head nurse's terminal. The frequency of nursing rounds is adjusted to once every 15 minutes, and the relevant medical order modification permissions in the electronic medical record system are locked. At the same time, the nursing manpower scheduling value W is calculated, W = N (t + 6 hours) × L, where L represents the number of patients in the corresponding ward. W nurses are selected and deployed to prioritize monitoring patients on the high-risk patient list. Otherwise, nursing work is arranged normally.

[0007] Furthermore, it also includes a dynamic early warning module, which responds in a graded manner. When 0.5 < N (t + 6 hours) ≤ 0.7, a yellow early warning is issued and a pop-up window prompts the nurse station screen. 0.7 < N (t+6 hours) ≤ 0.85, orange alert, nurses’ personal terminals vibrate to sound an alarm; N (t+6 hours) > 0.85, red alert, the hospital-wide broadcasting system activates emergency broadcasting.

[0008] Furthermore, it also includes an iterative optimization module, which pushes the result of the dynamic risk prediction value N (t+6 hours) to the nursing expert terminal for manual review. If it is accurate, the result of the dynamic risk prediction value N (t+6 hours) is regarded as correct data. Otherwise, it is regarded as erroneous data. When it is erroneous data, the dynamic warning operation caused by the erroneous judgment is immediately canceled manually, and the original dynamic risk prediction value N (t+6 hours) is replaced by the manual judgment result. At the same time, the correct data or erroneous data is used as the optimization sample of this system model to iteratively update the risk weight value of the nursing quality risk assessment.

[0009] Furthermore, the data acquisition module is connected to the medical information system, electronic medical record system, Internet of Things medical equipment, and nursing operation terminal through a standardized interface to collect patient vital sign data A, nursing operation records B, medication information C, equipment alarm log D, and nursing text report E in real time; Clean the collected data to remove noise, erroneous data, and inconsistent data, and fuse the data to integrate data from different sources into a unified CSV data format.

[0010] Furthermore, the dynamic assessment module extracts the heart rate A1, blood pressure A2, and blood oxygen saturation A3 of the patient's vital sign data A, the nursing frequency B1, nursing time B2, and nursing content B3 of the nursing operation record B, and the drug type C1, dosage C2, and medication time C3 of the medication information C, and calculates the patient's vital sign assessment value A4, A4=A11×0.3+A21×0.3+A31×0.4, where the characteristic heart rate A1 is within the normal range, A11 is 0, otherwise, A11 is 1; when the blood pressure A2 is normal, A21 is 0, otherwise, A21 is 1; when the blood oxygen saturation A3 is normal, A31 is 0, otherwise, A31 is 1; Nursing operation record evaluation value B4, B4=B11×0.3+ B21×0.3+ B31×0.4, where B11, B21, and B31 are the corresponding completion degrees of nursing frequency B1, nursing time B2, and nursing content B3; Medication information evaluation value C4, C4=C11×0.4+C21×0.4+C31×0.2, where if the drug type C1 is correct, C11 is 0, otherwise, C11 is 1; if the dosage C2 meets the requirements, C21 is 0, otherwise, C21 is 1; if the medication time C3 is correct, C31 is 0, otherwise, C21 is 1; Finally, calculate the comprehensive evaluation value M, M= A4×0.2+B4×0.2+ C4×0.4.

[0011] Furthermore, the dynamic prediction module extracts the alarm type D1, alarm frequency D2, and alarm duration D3 from the device alarm log D, and calculates the device risk value D4, D4=D11×0.4+D2÷D21×0.4+D3÷24 hours×0.3, where the alarm type D1 is a device failure, D11 is 0, and if the vital signs are abnormal, D11 is 1, and D21 is the alarm frequency threshold; Extract the compliance degree of medical measures E1 and the number of abnormal event marks E2 from the nursing text report E, and calculate the text risk value E3: E3 = E1 × 0.6 + E2 ÷ E21 × 0.4, where E1 is the 0-1 standardized value and E21 is the threshold of the number of abnormal event marks; Furthermore, the nursing quality risk assessment system uses 5G communication for information transmission, and physical layer encryption and dynamic key management for information encryption. The physical layer encryption is based on the physical layer security coding of Polar code, and the dynamic key management uses the 5G network SUCI solution to update the device access key every 30 minutes.

[0012] Due to the adoption of the above technical solution, the present invention has the following advantages compared with the prior art: 1. By dynamically evaluating the comprehensive assessment value of patient nursing quality in the ward, as well as the equipment risk value and text risk value, and jointly calculating the dynamic risk prediction value within the next 6 hours, the risk management checkpoint is moved forward. At the same time, by calculating the nursing manpower scheduling value, nurses are deployed in advance according to the dynamic risk prediction value, improving the efficient use of nursing resources. Based on the three-level N value warning (yellow / orange / red), differentiated handling strategies are matched, which not only avoids the waste of resources caused by excessive warnings, but also ensures 100% coverage of high-risk events through strong notification methods such as hospital-wide broadcasts. The iterative optimization module introduces a manual review and feedback mechanism. Through real-time correction of erroneous predictions and dynamic adjustment of model parameters, the system continuously adapts to changes in clinical scenarios, improving the timeliness of risk management. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a flow chart of a nursing quality risk assessment system based on medical big data in the present invention. DETAILED DESCRIPTION

[0014] The above and other technical contents, features and effects of the present invention are described below with reference to the attached Figure 1 The detailed description of the embodiments will clearly show that the structural contents mentioned in the following embodiments are all based on the accompanying drawings.

[0015] Example 1, based on existing technologies, in order to improve nursing quality and achieve the "forward shift" of risk management, includes a data acquisition module, a dynamic assessment module, a dynamic prediction module, and a closed-loop management module. The data acquisition module collects patient vital sign data A, nursing operation records B, medication information C, equipment alarm logs D, and nursing text reports E in real time; The dynamic evaluation module extracts the heart rate A1, blood pressure A2, and blood oxygen saturation A3 of the patient's vital signs data A, the nursing frequency B1, nursing time B2, and nursing content B3 of the nursing operation record B, and the drug type C1, dosage C2, and medication time C3 of the medication information C, calculates the patient's vital signs evaluation value A4, the nursing operation record evaluation value B4, and the medication information evaluation value C4, and finally calculates the comprehensive evaluation value M, M= A4×0.2+B4×0.2+ C4×0.4; The dynamic prediction module extracts the alarm type D1, alarm frequency D2, and alarm duration D3 from the device alarm log D, calculates the device risk value D4, extracts the treatment measure compliance E1 and the number of abnormal event marks E2 from the nursing text report E, and calculates the text risk value E3; The closed-loop management module calculates the dynamic risk prediction value N(t+6 hours) in the next 6 hours = 0.4×M(t)+0.3×D4(t)+0.2×E3(t)+0.1×M1, where t represents the current time value and M1 is the real-time change rate of the comprehensive assessment value M; When N (t + 6 hours) is greater than 0.7, a list of high-risk patients is generated and pushed to the head nurse's terminal. The frequency of nursing rounds is adjusted to once every 15 minutes, and the relevant medical order modification permissions in the electronic medical record system are locked. At the same time, the nursing manpower scheduling value W is calculated, W = N (t + 6 hours) × L, where L represents the number of patients in the corresponding ward. W nurses are selected and deployed to prioritize monitoring patients on the high-risk patient list. Otherwise, nursing work is arranged normally.

[0016] Furthermore, it also includes a dynamic early warning module, which responds in a graded manner. When 0.5 < N (t + 6 hours) ≤ 0.7, a yellow early warning is issued and a pop-up window prompts the nurse station screen. 0.7 < N (t+6 hours) ≤ 0.85, orange alert, nurses’ personal terminals vibrate to sound an alarm; N (t+6 hours) > 0.85, red alert, the hospital-wide broadcasting system activates emergency broadcasting.

[0017] Furthermore, it also includes an iterative optimization module, which pushes the result of the dynamic risk prediction value N (t+6 hours) to the nursing expert terminal for manual review. If it is accurate, the result of the dynamic risk prediction value N (t+6 hours) is regarded as correct data. Otherwise, it is regarded as erroneous data. When it is erroneous data, the dynamic warning operation caused by the erroneous judgment is immediately canceled manually, and the original dynamic risk prediction value N (t+6 hours) is replaced by the manual judgment result. At the same time, the correct data or erroneous data is used as the optimization sample of this system model to iteratively update the risk weight value of the nursing quality risk assessment.

[0018] Furthermore, the data acquisition module is connected to the medical information system, electronic medical record system, Internet of Things medical equipment, and nursing operation terminal through a standardized interface to collect patient vital sign data A, nursing operation records B, medication information C, equipment alarm log D, and nursing text report E in real time; Clean the collected data to remove noise, erroneous data, and inconsistent data, and fuse the data to integrate data from different sources into a unified CSV data format.

[0019] Furthermore, the dynamic assessment module extracts the heart rate A1, blood pressure A2, and blood oxygen saturation A3 of the patient's vital sign data A, the nursing frequency B1, nursing time B2, and nursing content B3 of the nursing operation record B, and the drug type C1, dosage C2, and medication time C3 of the medication information C, and calculates the patient's vital sign assessment value A4, A4=A11×0.3+A21×0.3+A31×0.4, where the characteristic heart rate A1 is within the normal range, A11 is 0, otherwise, A11 is 1; when the blood pressure A2 is normal, A21 is 0, otherwise, A21 is 1; when the blood oxygen saturation A3 is normal, A31 is 0, otherwise, A31 is 1; Nursing operation record evaluation value B4, B4=B11×0.3+ B21×0.3+ B31×0.4, where B11, B21, and B31 are the corresponding completion degrees of nursing frequency B1, nursing time B2, and nursing content B3; Medication information evaluation value C4, C4=C11×0.4+C21×0.4+C31×0.2, where if the drug type C1 is correct, C11 is 0, otherwise, C11 is 1; if the dosage C2 meets the requirements, C21 is 0, otherwise, C21 is 1; if the medication time C3 is correct, C31 is 0, otherwise, C21 is 1; Finally, calculate the comprehensive evaluation value M, M= A4×0.2+B4×0.2+ C4×0.4.

[0020] Furthermore, the dynamic prediction module extracts the alarm type D1, alarm frequency D2, and alarm duration D3 from the device alarm log D, and calculates the device risk value D4, D4=D11×0.4+D2÷D21×0.4+D3÷24 hours×0.3, where the alarm type D1 is a device failure, D11 is 0, and if the vital signs are abnormal, D11 is 1, and D21 is the alarm frequency threshold; Extract the compliance degree of medical measures E1 and the number of abnormal event marks E2 from the nursing text report E, and calculate the text risk value E3: E3 = E1 × 0.6 + E2 ÷ E21 × 0.4, where E1 is the 0-1 standardized value and E21 is the threshold of the number of abnormal event marks; Furthermore, the nursing quality risk assessment system uses 5G communication for information transmission, and physical layer encryption and dynamic key management for information encryption. The physical layer encryption is based on the physical layer security coding of Polar code, and the dynamic key management uses the 5G network SUCI solution to update the device access key every 30 minutes.

[0021] When the present invention is used specifically, based on the existing technology, the data acquisition module collects patient vital sign data A, nursing operation records B, medication information C, equipment alarm log D, and nursing text report E in real time; The dynamic evaluation module extracts the heart rate A1, blood pressure A2, and blood oxygen saturation A3 of the patient's vital signs data A, the nursing frequency B1, nursing time B2, and nursing content B3 of the nursing operation record B, and the drug type C1, dosage C2, and medication time C3 of the medication information C, calculates the patient's vital signs evaluation value A4, the nursing operation record evaluation value B4, and the medication information evaluation value C4, and finally calculates the comprehensive evaluation value M, M= A4×0.2+B4×0.2+ C4×0.4; The dynamic prediction module extracts the alarm type D1, alarm frequency D2, and alarm duration D3 from the device alarm log D, calculates the device risk value D4, extracts the treatment measure compliance E1 and the number of abnormal event marks E2 from the nursing text report E, and calculates the text risk value E3; The closed-loop management module calculates the dynamic risk prediction value N(t+6 hours) in the next 6 hours = 0.4×M(t)+0.3×D4(t)+0.2×E3(t)+0.1×M1, where t represents the current time value and M1 is the real-time change rate of the comprehensive assessment value M; When N (t + 6 hours) is greater than 0.7, a list of high-risk patients is generated and pushed to the head nurse's terminal. The frequency of nursing rounds is adjusted to once every 15 minutes, and the relevant medical order modification permissions in the electronic medical record system are locked. At the same time, the nursing manpower scheduling value W is calculated, W = N (t + 6 hours) × L, where L represents the number of patients in the corresponding ward. W nurses are selected and deployed to prioritize monitoring patients on the high-risk patient list. Otherwise, nursing work is arranged normally. By dynamically evaluating the comprehensive assessment value of patient nursing quality in the ward, the equipment risk value, and the text risk value, the dynamic risk prediction value within the next 6 hours is jointly calculated. Nurses are deployed in advance based on the dynamic risk prediction value. At the same time, dynamic early warning and iterative optimization are carried out to improve the timeliness of risk management and the efficient use of nursing resources.

[0022] The above is a further detailed description of the present invention in combination with specific implementation methods, and it cannot be determined that the specific implementation of the present invention is limited to this; for technical personnel in the technical fields to which the present invention belongs and related technical fields, based on the technical solution ideas of the present invention, the expansion and replacement of operating methods and data should all fall within the scope of protection of the present invention.

Claims

1. A nursing quality risk assessment system based on medical big data, characterized by: It includes a data acquisition module, a dynamic evaluation module, a dynamic prediction module and a closed-loop management module. The data acquisition module collects patient vital sign data A, nursing operation records B, medication information C, equipment alarm log D, and nursing text report E in real time; The dynamic evaluation module extracts the heart rate A1, blood pressure A2, and blood oxygen saturation A3 of the patient's vital signs data A, the nursing frequency B1, nursing time B2, and nursing content B3 of the nursing operation record B, and the drug type C1, dosage C2, and medication time C3 of the medication information C, calculates the patient's vital signs evaluation value A4, the nursing operation record evaluation value B4, and the medication information evaluation value C4, and finally calculates the comprehensive evaluation value M, M= A4×0.2+B4×0.2+ C4×0.4; The dynamic prediction module extracts the alarm type D1, alarm frequency D2, and alarm duration D3 from the device alarm log D, calculates the device risk value D4, extracts the treatment measure compliance E1 and the number of abnormal event marks E2 from the nursing text report E, and calculates the text risk value E3; The closed-loop management module calculates the dynamic risk prediction value N(t+6 hours) in the next 6 hours = 0.4×M(t)+0.3×D4(t)+0.2×E3(t)+0.1×M1, where t represents the current time value and M1 is the real-time change rate of the comprehensive assessment value M; When N (t + 6 hours) is greater than 0.7, a list of high-risk patients is generated and pushed to the head nurse's terminal. The frequency of nursing rounds is adjusted to once every 15 minutes, and the relevant medical order modification permissions in the electronic medical record system are locked. At the same time, the nursing manpower scheduling value W is calculated, W = N (t + 6 hours) × L, where L represents the number of patients in the corresponding ward. W nurses are selected and deployed to prioritize monitoring patients on the high-risk patient list. Otherwise, nursing work is arranged normally.

2. A nursing quality risk assessment system based on medical big data according to claim 1, characterized in that: It also includes a dynamic early warning module, which responds in a graded manner. When 0.5 < N (t + 6 hours) ≤ 0.7, a yellow early warning is issued, and a pop-up window prompts the nurse station screen. 0.7 < N (t+6 hours) ≤ 0.85, orange warning, the nurse's personal terminal vibrates; N (t+6 hours) > 0.85, red alert, emergency broadcast activated on the entire hospital broadcasting system.

3. The nursing quality risk assessment system based on medical big data according to claim 2 is characterized in that: It also includes an iterative optimization module, which pushes the result of the dynamic risk prediction value N (t+6 hours) to the nursing expert terminal for manual review. If it is accurate, the result of the dynamic risk prediction value N (t+6 hours) is regarded as correct data. Otherwise, it is regarded as erroneous data. When it is erroneous data, the dynamic early warning operation caused by the erroneous judgment is immediately canceled manually, and the original dynamic risk prediction value N (t+6 hours) is replaced by the manual judgment result. At the same time, the correct data or erroneous data is used as the optimization sample of this system model to iteratively update the risk weight value of the nursing quality risk assessment.

4. The nursing quality risk assessment system based on medical big data according to claim 1 is characterized in that: The data acquisition module is connected to the medical information system, electronic medical record system, Internet of Things medical equipment, and nursing operation terminal through a standardized interface to collect patient vital sign data A, nursing operation records B, medication information C, equipment alarm log D, and nursing text report E in real time; Clean the collected data to remove noise, erroneous data, and inconsistent data, and fuse the data to integrate data from different sources into a unified CSV data format.

5. The nursing quality risk assessment system based on medical big data according to claim 1 is characterized in that: The dynamic evaluation module extracts the heart rate A1, blood pressure A2, and blood oxygen saturation A3 of the patient's vital sign data A, the nursing frequency B1, nursing time B2, and nursing content B3 of the nursing operation record B, and the drug type C1, dosage C2, and medication time C3 of the medication information C, and calculates the patient's vital sign evaluation value A4, A4=A11×0.3+A21×0.3+A31×0.4, where if the characteristic heart rate A1 is in the normal range, A11 is 0, otherwise, A11 is 1; if the blood pressure A2 is normal, A21 is 0, otherwise, A21 is 1; if the blood oxygen saturation A3 is normal, A31 is 0, otherwise, A31 is 1; Nursing operation record evaluation value B4, B4=B11×0.3+ B21×0.3+ B31×0.4, where B11, B21, and B31 are the corresponding completion degrees of nursing frequency B1, nursing time B2, and nursing content B3; Medication information evaluation value C4, C4=C11×0.4+C21×0.4+C31×0.2, where if the drug type C1 is correct, C11 is 0, otherwise, C11 is 1; if the dosage C2 meets the requirements, C21 is 0, otherwise, C21 is 1; if the medication time C3 is correct, C31 is 0, otherwise, C21 is 1; Finally, calculate the comprehensive evaluation value M, M= A4×0.2+B4×0.2+ C4×0.

4.

6. The nursing quality risk assessment system based on medical big data according to claim 1 is characterized in that: The dynamic prediction module extracts the alarm type D1, alarm frequency D2, and alarm duration D3 from the device alarm log D, and calculates the device risk value D4, D4=D11×0.4+D2÷D21×0.4+D3÷24 hours×0.3, where the alarm type D1 is a device failure, D11 is 0, and if the vital signs are abnormal, D11 is 1, and D21 is the alarm frequency threshold; Extract the compliance degree of medical measures E1 and the number of abnormal event marks E2 from the nursing text report E, and calculate the text risk value E3, E3 = E1×0.6+E2÷E21×0.4, where E1 is the standardized value of 0-1 and E21 is the threshold of the number of abnormal event marks.

7. A nursing quality risk assessment system based on medical big data according to any one of claims 1 to 6, characterized in that: The nursing quality risk assessment system uses 5G communication for information transmission, and physical layer encryption and dynamic key management for information encryption. The physical layer encryption is based on the physical layer security coding of Polar code, and the dynamic key management uses the 5G network SUCI solution to update the device access key every 30 minutes.

Images