Hospitalized patient classification management method and system

Abstract

The application discloses a kind of inpatient hierarchical management method and system, belong to medical risk management technical field.The method includes: obtaining the personal information of inpatient and determining its risk index;According to risk index, dynamic correction electronic fence;Through real-time acquisition patient position by wearable device, combine fence relationship to calculate risk index;While calculating vital sign index according to vital sign information;Comprehensive risk index and vital sign index determine early warning grade, and execute corresponding early warning prompt.The application realizes the change from passive patrol to active early warning through multi-dimensional risk fusion and dynamic fence adjustment, improves the timeliness and accuracy of high-risk patient monitoring, while reducing false positive rate and nursing interference.

Description

Technical Field

[0001] This application relates to the field of medical risk management technology, specifically to a method and system for the hierarchical management of hospitalized patients. Background Technology

[0002] With the advancement of smart hospital construction, the Internet of Things (IoT) and wearable technologies are being gradually introduced into the safety management of hospitalized patients. Bluetooth positioning, LoRa wireless transmission, and multi-parameter vital sign monitoring can now enable real-time tracking and physiological data collection of patients; various early warning scores (NEWS) are also widely used to assess the risk of disease deterioration.

[0003] For example, Chinese invention patent CN113784289B, entitled "A Method for Assisted Care of Hospital Patients," discloses a method for assisted care of hospital patients, including the following steps: S1: Establish a hospital map in the terminal system, using rooms as the smallest unit, and obtain the coordinates and names of each corresponding room; S2: Divide the area into sections using rooms as the smallest fence unit, and select the required electronic fence range; S3: Manage the selected electronic fence range by area level, with different levels of fence areas corresponding to different associated personnel; S4: When the terminal system detects that a patient has entered a fence area, it immediately sends the patient's fence boundary violation information to the associated personnel corresponding to that area, so that the associated personnel can promptly locate the patient who has crossed the fence. This method can effectively care for patients, accurately reporting any boundary violations to the associated personnel. Moreover, the fence area provided by this method can be modified at any time to adapt to complex settings, accurately detects the patient's specific location, improves work efficiency, and ensures patient safety.

[0004] However, this method also has some problems. It treats "location" and "risk" separately: although the electronic fence can be modified at any time, such modification is a temporary operation by the management personnel. Its alarm threshold remains unchanged throughout the process, and the risk level depends only on vital signs or a single assessment, lacking real-time correction for individual differences and historical trajectories. When patients leave the ward or stay in the bathroom for a long time, the fixed threshold is very prone to false alarms or missed alarms, resulting in frequent invalid pop-ups at the nurse station, clinical response fatigue, and the defects of "nursing gaps" and "alarm storms" coexisting.

[0005] Therefore, there is an urgent need for a graded alarm scheme for hospitalized patients that couples multidimensional risks with dynamic location, can adaptively adjust alarm boundaries, and can self-evolve with the usage cycle. Summary of the Invention

[0006] The purpose of this application is to address the shortcomings of the aforementioned background technology and provide a method and system for the hierarchical management of hospitalized patients.

[0007] The technical solution of this application is as follows: The purpose of this application is to provide a method, system, equipment and medium for hierarchical management of hospitalized patients, in response to the deficiencies of the existing technology.

[0008] The technical solution adopted in this application is: a method for hierarchical management of hospitalized patients, including... Obtain personal information collected from hospitalized patients via wearable devices, and determine the patient's risk index within the current sampling period based on the personal information; Based on the risk index, the preset initial electronic fence is modified to obtain a modified electronic fence; The wearable device collects the patient's location information in real time, and calculates the patient's risk index based on the relationship between the location information and the location of the modified electronic fence. Based on the vital signs information in the aforementioned personal information, obtain the patient's vital signs index; The warning level for the patient is determined based on the risk index and the vital signs index. Based on the aforementioned warning level, the corresponding warning prompt will be executed.

[0009] According to the method for hierarchical management of hospitalized patients provided in this application, the personal information includes the patient's identity information, medical condition information, self-care ability information, tube information, fall risk information, suicide risk information, and vital sign information; the vital sign information includes one or more of, for example, heart rate, blood oxygen saturation, and body temperature.

[0010] According to the method for hierarchical management of hospitalized patients provided in this application, the step of determining the patient's risk index based on personal information includes: scoring the patient's medical condition information, self-care ability information, tube information, fall risk information, and suicide risk information respectively, and determining the highest risk score among them as the patient's risk index.

[0011] According to the inpatient triage management method provided in this application, the method for risk scoring of various information includes: Condition information score: Stable condition, stable vital signs and no acute symptoms, score 0; Mild condition, requires routine monitoring and no immediate risk, score 0.3; Severe condition and requires close monitoring or intervention, score 0.6; Critical condition and requires emergency intervention or ICU monitoring, score 0.8; Life-threatening condition with multiple organ failure or end-stage condition, score 1.0. Self-care ability assessment: Completely self-reliant, 0 points; Mildly dependent, 0.2 points; Moderately dependent, 0.5 points; Severely dependent, 0.8 points; Completely dependent, 1.0 point; Piping information scoring: No piping, 0 points; Low-risk piping, 0.2 points; Medium-risk piping, 0.5 points; High-risk piping, 0.8 points; Extremely high-risk piping, 1.0 point; Fall risk information scoring: No history of falls, able to move freely and conscious, score 0; Low risk, slightly unsteady gait but no assistance required, score 0.3; Medium risk, using a walking aid, elderly or visually impaired, score 0.6; High risk, history of frequent falls, confusion or orthostatic hypotension, score 0.8; Very high risk, delirium, severe dizziness, severe limb weakness or paralysis, score 1.0; Suicide Risk Information Score: No suicidal ideation or history, stable mood, 0 points; Mild risk, occasional negative thoughts but no plans or behaviors, 0.4 points; Moderate risk, suicidal thoughts or history but no recent plans, 0.7 points; High risk, clear plans or preparatory behaviors, 0.9 points; Very high risk, recent suicide attempts, strong suicidal intentions or command hallucinations, 1.0 point.

[0012] According to the inpatient triage management method provided in this application, the step of correcting the initial electronic fence based on the risk index includes: when the risk index is not greater than a first threshold a, the patient's risk level is determined to be level one, and the initial electronic fence is extended outward by a first distance n to form the corrected electronic fence; when the risk index is greater than a and not greater than a second threshold b, the risk level is determined to be level two, and the initial electronic fence is directly used as the corrected electronic fence; when the risk index is greater than b, the risk level is determined to be level three, and the initial electronic fence is contracted inward by a second distance m to form the corrected electronic fence; wherein, 0 < a < b ≤ 1, and n and m are preset distance values ​​greater than 0.

[0013] According to the inpatient triage management method provided in this application, the initial electronic fence is an area defined according to the ward to which the patient belongs, excluding toilets and balconies.

[0014] According to the inpatient triage management method provided in this application, the calculation of the risk index based on location information and modified electronic fence includes: if the patient's location is within the range of the modified electronic fence, the risk index is 0; if the patient's location is outside the range of the modified electronic fence, but the stay time does not exceed a preset time threshold T, the risk index is 0.5; if the stay time exceeds the preset time threshold T, the risk index is 1.0.

[0015] According to the method for hierarchical management of hospitalized patients provided in this application, the method for obtaining the vital sign index of a patient based on the vital sign information in the patient's personal information includes: obtaining the respiratory rate, oxygen saturation, oxygen inhalation status, level of consciousness, systolic blood pressure, heart rate and body temperature in the patient's vital sign information, scoring the above-mentioned physiological indicators of the patient according to the NEWS2 model, and obtaining the vital sign index of the patient.

[0016] According to the inpatient triage management method provided in this application, the determination of the warning level based on the risk index and vital sign index includes: pre-defining five progressively increasing risk levels (first to fifth) according to the numerical range of the vital sign index; determining the patient's basic risk level based on the patient's vital sign index; if the risk index is 0, then the basic risk level is used as the warning level; if the risk index is 0.5, then the warning level is increased by one level above the basic risk level; if the risk index is 1.0, then the warning level is increased by two levels above the basic risk level; if the increased level exceeds the fifth risk level, then the warning level is determined to be the highest level.

[0017] According to the triage management method for hospitalized patients provided in this application, the early warning prompts are executed as follows based on the early warning level: When the warning level is the highest risk level, no proactive warning will be issued. When the warning level is the second risk level, the wearable device will be displayed in solid green and the status will be recorded on the electronic whiteboard at the nurse station. When the warning level is the third risk level, the wearable device will flash yellow and record the status on the electronic whiteboard at the nurse station. When the warning level is the fourth risk level, the wearable device will flash red and pop up and vibrate to alert the nurse station electronic whiteboard, PDA and medical staff wristbands. When the warning level is the fifth risk level, the wearable device will flash red, and pop-up windows and vibration prompts will appear on the electronic whiteboard at the nurse station, PDA, and medical staff wristbands, and an audible alarm will be activated simultaneously.

[0018] This application also relates to a hospitalized patient triage management system for performing the above-described method, the system comprising: Wearable devices are used to collect personal information and real-time location information of hospitalized patients; The processing unit is configured as follows: The patient's risk index is determined based on the aforementioned personal information; The initial electronic fence is modified based on the risk index to obtain a modified electronic fence; Based on the relationship between the real-time location information and the modified electronic fence, the patient's risk index is calculated; Based on the vital signs information in the aforementioned personal information, obtain the patient's vital signs index; The patient's warning level is determined based on the risk index and the vital signs index; The early warning unit is used to trigger a corresponding early warning notification based on the early warning level.

[0019] The advantages of this application are as follows: 1. This application provides a hierarchical management method for hospitalized patients. This application transforms the passive mode that relies on manual patrols into a data-driven proactive early warning mode, compressing the window time from the occurrence of an event to the nurse's response; This application integrates three types of risk information with different natures and update frequencies—static / semi-static risk assessment (risk index), dynamic behavioral risk (hazard index), and real-time physiological risk (vital sign index)—step by step in a logical manner (first, the geographic baseline electronic fence of behavioral monitoring is corrected with static risk, and then behavioral risk and physiological risk are superimposed and graded); This integration method is a concrete manifestation of coupling in a unified model, so that the final early warning level comprehensively reflects the patient's overall safety status, rather than a single-dimensional abnormality; 2. This application defines the dimensions of personal information and clarifies the diverse data sources upon which risk assessment relies; it covers everything from objective medical conditions (illness, tubes), to functional abilities (self-care ability, fall risk), to psychosocial risks (suicide risk), and core physiological indicators (vital signs); this comprehensiveness ensures that the subsequent risk index can accurately and comprehensively depict the patient's complex and individualized risk profile, which is a prerequisite for achieving precise hierarchical management; 3. This application specifies that multiple risk information are scored separately and the highest score is taken as the risk index. The strategy of taking the highest risk score is in line with the principle of focusing on the most dangerous safety issues in clinical nursing. This method simplifies the decision-making logic and ensures that the system can make the most timely and conservative response to the patient's primary risks, avoiding the downplaying of high-risk factors due to risk averaging. 4. This application provides specific, tiered scoring standards for various risks (illness, self-care ability, etc.), transforming the clinical judgment that originally relied on subjective descriptions into objective scores that can be processed by computers. This is a key step in digitizing the multidimensional risk scale, enabling risks from different sources and of different natures to be compared and calculated on a unified numerical scale, providing accurate numerical input for subsequent electronic fence correction and comprehensive early warning. The descriptions of each tier (such as command hallucinations and orthostatic hypotension) have clear clinical indications, indicating that the scoring system originates from medical practice and improves the professionalism and credibility of the system's judgment. 5. This application dynamically adjusts the electronic fence based on the risk index by expanding outward, maintaining its original state, or contracting inward; achieving individualized and dynamic matching of alarm boundaries: this is the core means to achieve real-time matching of alarm boundaries with individual patient status; for low-risk patients, it grants them a larger safe activity space, reduces invalid alarms triggered by normal activities (such as walking slowly in the corridor), significantly improves specificity, reduces the pop-up load and noise interference at the nurse station, and improves the nursing experience; for high-risk patients, their safe activity range is restricted to a more core area, and an alarm is triggered once they leave, greatly improving monitoring sensitivity and ensuring close monitoring of high-risk patients; for indoor positioning of wearable devices, there are unavoidable errors; expanding the fence outward can accommodate the positioning drift of low-risk patients and avoid false alarms; while contracting the fence inward ensures that for high-risk patients, even if there are positioning errors, their true location is highly likely to be restricted to a smaller safe range, and once an alarm is triggered, the actual risk is extremely high; 6. This application stipulates that the initial electronic fence is the ward area, excluding toilets and balconies; using wards as the basic safety area is in line with common sense in medical management; actively excluding toilets and balconies, which are high-risk locations for patient falls, suicides and other accidents, is a highly practical design; it avoids the contradiction that the system remains silent when patients enter these high-risk areas because the patients are in the permitted area, forcing the system to rely on other rules or vital signs to judge the risk, which in effect defaults to these special areas as areas that require extra attention, reflecting the rigor of the design logic; 7. This application assigns different risk indices based on whether the patient is outside the modified electronic fence and the length of time they stay; it introduces a time tolerance window to balance response timeliness and operational error tolerance, and sets a time threshold T to create a stay threshold effect that decreases linearly with the risk level; for brief, accidental boundary crossings (such as approaching the door), the system provides an observation period (risk index 0.5) and does not immediately trigger a high-level alarm, effectively reducing false alarms caused by brief behavioral fluctuations; only continuous boundary crossings are judged as the highest risk (1.0), which is more in line with real risk scenarios and improves the accuracy of alarms; 8. This application uses the NEWS2 model to score multiple vital signs to obtain a vital sign index; it introduces an internationally recognized and objective early warning tool for physiological deterioration, NEWS2 (UK National Early Warning Score 2nd Edition), which is a widely validated standardized tool that can scientifically and quantitatively reflect the severity and trend of a patient's condition; integrating it into the system makes the assessment of real-time vital signs move away from the simple judgment of a single indicator threshold, and become a comprehensive, scientifically weighted index, which greatly enhances the professionalism and early warning value of the vital sign assessment part; 9. This application determines the basic risk level based on vital sign indices and then upgrades it according to the danger index to determine the final warning level. This is the final decision point coupled in the unified model. The rule creatively stipulates that even if vital signs are stable (low basic risk level), abnormal bed-leaving behavior (high danger index) will upgrade the warning level; conversely, even if the patient is in bed, significant physiological deterioration will trigger an alarm. This non-linear superposition rule (fixed upgrade of 1 or 2 levels) concisely and effectively amplifies the composite risk, ensuring that the system can make the highest level of response to extremely high-risk scenarios such as abnormal vital signs and bed-leaving. It is the final algorithmic embodiment of the risk-driven concept. 10. This application specifies the detailed warning prompts for risk levels one through five, with each level progressively stronger. The warning prompts strictly correspond to the warning levels, progressing from a non-intrusive display (solid green) to visual alerts (flashing yellow / red), then to proactive push notifications from multiple terminals (PDA, wristband pop-up vibration), and finally to emergency alarms combining sound and light. This design enables precise allocation of nursing resources: low-level warnings are only recorded without interrupting the workflow; medium-level warnings attract attention; and high-level warnings require mandatory intervention. It systematically reduces noise interference under normal conditions while ensuring that information is conveyed without omission in truly critical situations. It is the final and crucial link in improving clinical safety. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the hierarchical management method for inpatients in this application. Detailed Implementation

[0021] The embodiments of this application are described in detail below, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0022] In the description of this application, it should be understood that the terms "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0023] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0024] The present application will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0025] This application relates to a triage management method for inpatients. The core of this method lies in collecting multidimensional personal information and real-time location data of patients through wearable devices. First, a risk index representing static medical risk is determined based on quantitative scoring rules. Then, a pre-set electronic fence is adjusted in a personalized manner to form a modified electronic fence. The system calculates a behavioral risk index by comparing the relationship between the patient's real-time location and the modified fence, as well as their dwell time. Simultaneously, a physiological risk index is calculated using a standardized model based on vital sign data. Finally, the behavioral risk index and the physiological risk index are coupled to determine a comprehensive early warning level from low to high, triggering a progressive response from status display to multi-terminal mandatory alarms. This solution realizes a shift from homogenized, timed rounds to individualized, risk-driven proactive care. By dynamically matching alarm boundaries with patient status, it significantly improves the timeliness of high-risk event response while effectively reducing nursing interference.

[0026] Specifically, such as Figure 1 As shown, this method begins with the acquisition and processing of multidimensional personal information about the patient and ultimately culminates in providing early warning prompts that match the risk level, forming a dynamic, closed-loop intelligent monitoring process. The method is executed sequentially according to the following steps: S101: Risk Index Determination The system collects the patient's personal information through a smart bracelet (a type of wearable device, but not limited to smart bracelets; other wearable devices are also acceptable), including but not limited to information on the patient's condition retrieved from electronic medical records and information on self-care ability entered into nursing assessments. The system's built-in risk assessment module calculates the above information based on a set of quantitative rules and outputs a risk index between 0 and 1. This index represents the patient's baseline risk level based on their static or semi-static medical condition. S102: Dynamic Correction of Electronic Fence The system has a preset initial electronic fence based on the patient's ward area; based on the risk index calculated by S101, the system calls the fence correction algorithm to intelligently adjust the boundary of the initial electronic fence and generate a corrected electronic fence that matches the individual risk level of the patient. S103: Risk Index Calculation The smart bracelet's built-in UWB or Bluetooth positioning module reports the patient's location coordinates in real time; the system compares this location with the corrected electronic fence generated by S102, and calculates a risk index that characterizes the risk of the patient's boundary-crossing behavior in combination with preset time logic. S104: Calculation of vital signs index The system extracts vital signs information such as heart rate, blood oxygen, and body temperature from the physiological data continuously collected by the smart bracelet, and uses a standardized early warning scoring model to calculate an objective vital signs index to quantify the patient's real-time physiological state risk. S105: Early Warning Level Decision The danger index obtained from S103 and the vital signs index obtained from S104 are input into the early warning decision engine. The decision engine first determines a basic risk level based on the vital signs index, and then increases the basic level by superimposing the value of the danger index, and finally determines a comprehensive early warning level from low to high. S106: Implementation of Tiered Early Warning Based on the warning level determined by S105, the system triggers a preset graded response protocol and executes precise warning prompts by adjusting the display of wearable devices, updating the status on the nurse station's large screen, sending pop-up and vibration reminders to medical staff's mobile terminals (PDAs, wristbands), and even activating sound and light alarms.

[0027] This application constructs a full-link technical solution from data perception to intelligent decision-making to precise execution; its core advantage lies in realizing a fundamental transformation of the monitoring mode: from a manual inspection mode that relies on fixed time intervals to a proactive risk warning mode continuously driven by real-time patient status data; by coupling the three key dimensions of static risk assessment, dynamic behavior monitoring and real-time physiological monitoring in a unified process, the system can identify complex risks earlier and more comprehensively, laying a methodological foundation for significantly reducing the window time from the occurrence of an event to the nurse's response.

[0028] In some embodiments of this application, this embodiment optimizes the above-mentioned risk assessment and dynamic adjustment of the fence. Specifically, this embodiment elaborates on how to quantify the risk index based on multidimensional personal information and how to use the index to realize the personalized and dynamic correction of the electronic fence.

[0029] The personal information in this application includes the patient's identity information (such as gender, age, name, home address, emergency contact, occupation, etc.), medical condition information (cause of hospitalization, diagnosis information, etc.), self-care ability information (whether the patient has the ability to take care of themselves), tube information (whether tubes are inserted), fall risk information (whether there is a risk of falling), suicide risk information (whether there is a risk of suicide), and vital signs information, including one or more of the following: heart rate, blood oxygen saturation, and body temperature.

[0030] To quantify the risk index, risk scores were assigned to the aforementioned medical condition information, self-care ability information, tube / tube information, fall risk information, and suicide risk information. The specific scoring methods are as follows: Condition information score: Stable condition, stable vital signs and no acute symptoms, score 0; Mild condition, requires routine monitoring and no immediate risk, score 0.3; Severe condition and requires close monitoring or intervention, score 0.6; Critical condition and requires emergency intervention or ICU monitoring, score 0.8; Life-threatening condition with multiple organ failure or end-stage condition, score 1.0. Self-care ability assessment: Completely self-reliant, 0 points; Mildly dependent, 0.2 points; Moderately dependent, 0.5 points; Severely dependent, 0.8 points; Completely dependent, 1.0 point; Piping information scoring: No piping, 0 points; Low-risk piping, 0.2 points; Medium-risk piping, 0.5 points; High-risk piping, 0.8 points; Extremely high-risk piping, 1.0 point; Fall risk information scoring: No history of falls, able to move freely and conscious, score 0; Low risk, slightly unsteady gait but no assistance required, score 0.3; Medium risk, using a walking aid, elderly or visually impaired, score 0.6; High risk, history of frequent falls, confusion or orthostatic hypotension, score 0.8; Very high risk, delirium, severe dizziness, severe limb weakness or paralysis, score 1.0; Suicide Risk Information Score: No suicidal ideation or history, stable mood, 0 points; Mild risk, occasional negative thoughts but no plans or behaviors, 0.4 points; Moderate risk, suicidal thoughts or history but no recent plans, 0.7 points; High risk, clear plans or preparatory behaviors, 0.9 points; Very high risk, recent suicide attempts, strong suicidal intentions or command hallucinations, 1.0 point.

[0031] For example, a patient diagnosed with heart failure and dyspnea may be rated as 0.6 (severe, requiring close monitoring); The patient is bedridden and requires assistance to turn over. His self-care ability score is 0.8 (severely dependent). The patient had a urinary catheter, which is considered a low-risk catheter, and was rated 0.2 points. Its fall risk information was rated 0 points due to being bedridden; Its suicide risk information is unavailable, and it is rated as 0 points; According to the highest score principle of claim 3, the patient's risk index was determined to be 0.8 points (based on self-care ability score).

[0032] After obtaining the score for each item, the highest risk score is determined as the patient's risk index.

[0033] This allows us to obtain the patient's risk index. By transforming complex clinical descriptions into structured, graded scores, we digitize and objectify risk factors, providing accurate and consistent input for all subsequent automated decision-making.

[0034] In other embodiments of this application, the initial electronic fence and the method for modifying the initial electronic fence described above are optimized. Specifically, the initial electronic fence is set as a polygon of the ward space where the patient's assigned bed is located. When drawing this polygon, the independent toilet and balcony areas in the ward are deliberately excluded. This means that even if the patient enters the toilet or balcony, it is logically considered to have left the initial electronic fence area. This design forces the system to rely on subsequent rules (such as time thresholds) to determine the risk of the behavior.

[0035] Specifically, after obtaining the initial electronic fence and the corresponding patient's risk index, the initial electronic fence is modified based on the risk index. The modification method includes: when the risk index is not greater than a first threshold 'a', the patient's risk level is determined to be Level 1, and the initial electronic fence is extended outward by a first distance 'n' to form the modified electronic fence; when the risk index is greater than 'a' and not greater than a second threshold 'b', the risk level is determined to be Level 2, and the initial electronic fence is directly used as the modified electronic fence; when the risk index is greater than 'b', the risk level is determined to be Level 3, and the initial electronic fence is contracted inward by a second distance 'm' to form the modified electronic fence; wherein, 0 < a < b ≤ 1, and n and m are preset distance values ​​greater than 0.

[0036] The system preset parameters are: first threshold a=0.3, second threshold b=0.7, expansion distance n=2 meters, and contraction distance m=1.5 meters. In actual applications, these values ​​are not limited to and can be set according to requirements.

[0037] For example, in case 1 (low risk): if the patient's risk index is 0.2 (≤a), the system determines that it is a level 1 risk; then, based on the ward boundary, a modified electronic fence is generated by extending 2 meters outward, allowing the patient to move appropriately in the corridor without triggering the location alarm; Scenario 2 (Medium Risk): If the patient's risk index is 0.5 (>a and ≤b), it is classified as Level 2 risk; the system will directly use the ward boundary as a corrected electronic fence, requiring the patient to remain within the ward. Scenario 3 (High Risk): If the patient's risk index is 0.8 (>b) as in the previous case, it is determined to be a level 3 risk. Then, based on the ward boundary, a modified electronic fence is generated by shrinking inward by 1.5 meters to form a more core safety area (such as limited to the area around the bed). Once the patient attempts to leave the bed area, a high alert state is entered.

[0038] This embodiment enables the electronic fence to intelligently expand and contract according to the patient's risk level; relaxing restrictions for low-risk patients reduces invalid alarms caused by normal activities, improves system specificity, and reduces nurses' interference; tightening the fence for high-risk patients greatly improves monitoring sensitivity and ensures strict protection for critically ill patients; this dynamic adjustment mechanism is the key technical means to achieve real-time matching between alarm boundaries and individual patient status, thereby balancing sensitivity and specificity.

[0039] In some other preferred embodiments of this application, the above-mentioned method for determining dangerous behavior is described. Specifically, if the patient's location is within the range of the modified electronic fence, the danger index is 0; if the patient's location is outside the range of the modified electronic fence, but the stay time does not exceed a preset time threshold T, the danger index is 0.5; if the stay time exceeds the preset time threshold T, the danger index is 1.0.

[0040] When the wristband (wearable device) is first activated (i.e., when it is worn and bound to the inpatient), a baseline sampling is performed on the Bluetooth positioning beacons in the ward using Bluetooth 4.0 electromagnetic fingerprint comparison technology. Simultaneously, sampling packets are transmitted back via LoRa uplink to obtain the initial positioning error φ0, which is then written to the wristband's memory. The Bluetooth positioning beacon set B = {b1, b2, ..., b...} n}, where b1, b2, ..., b n Representing the 1st, 2nd, ..., nth Bluetooth beacons respectively; the received signal strength indication matrix RSSO∈R is obtained through reference sampling. n *1, where the number of sampling points n 4. Simultaneously transmit RSSO; the initial positioning error is calculated using the following formula: in, For two-dimensional coordinates calculated based on RSSO ( x, y ), The actual coordinates of the Bluetooth positioning beacon ( x gt ,y gt ), .

[0041] The wristband broadcasts Bluetooth signals periodically (t), and the IoT base station array performs joint TOA / AOA ranging, outputting the original coordinate sequence. P 0 (t) .

[0042] Assume the system has a preset time threshold. T=120 The danger index is calculated in seconds as follows: If Patient A (low risk index, fence has been extended) walks in the corridor and remains within the modified electronic fence, then their risk index is always 0. Patient B (medium risk index) entered the bathroom (outside the corrected fence) to use the toilet; the system started timing: If it returns to the bed (inside the fence) after 115 seconds, its danger level is recorded as 0.5; If it does not return after 125 seconds, the system determines it as a timeout and its danger index is immediately updated to 1.0.

[0043] The time threshold T can be dynamically adjusted according to the medical staff configuration (such as day-night differences). For example, during the daytime period from 07:01 to 18:00, the time threshold T can be 120 seconds; during the nighttime period from 18:01 to 07:00, the time threshold T can be 120 seconds.

[0044] This embodiment sets a time threshold T, creating a buffer observation period; brief, purposeful boundary crossings (such as normal toilet use) will not immediately trigger advanced alarms, which effectively filters out a large number of non-emergency situations and significantly reduces the false alarm rate; while long-term abnormal stays will be accurately captured; combined with the dynamic fence of Embodiment 2, for high-risk patients, their fence has been contracted, and they may have crossed the boundary as soon as they move from the bed to the toilet door. At this time, their actual safe stay time is much lower than that of low-risk patients, realizing the intelligent effect of the stay threshold decreasing linearly with the risk level.

[0045] In a further embodiment of this application, the above-mentioned vital sign assessment method is optimized. Specifically, the respiratory rate, oxygen saturation, oxygen intake, level of consciousness, systolic blood pressure, heart rate and body temperature of the patient are obtained, and the above-mentioned physiological indicators of the patient are scored according to the NEWS2 model to obtain the patient's vital sign index.

[0046] The UK National Early Warning Score, version 2 (NEWS2) model was used. For example, the system acquired the patient's real-time data: heart rate 105 beats / min (score +1), blood oxygen saturation 92% (score +3), respiratory rate 22 breaths / min (score +3), body temperature 38.5℃ (score +1), conscious (score 0), systolic blood pressure 110 mmHg (score 0). The scores were added together to obtain the patient's vital signs index (NEWS2 total score) of 8.

[0047] Based on the range of vital signs index values, five progressively increasing risk levels are defined in advance. For example, the system presets: NEWS2 total score of 0-4 points is the first risk level (low), 5-6 points is the second level (medium), 7-8 points is the third level (high), ≥9 points is the fourth level (critical), and there is a fifth level which is the highest emergency level.

[0048] The method for determining the warning level based on the risk index and vital sign index is as follows: determine the patient's basic risk level according to the patient's vital sign index; if the risk index is 0, then the basic risk level is used as the warning level; if the risk index is 0.5, then the basic risk level is increased by one level as the warning level; if the risk index is 1.0, then the basic risk level is increased by two levels as the warning level; if the increased level exceeds the fifth risk level, then the warning level is determined to be the highest level.

[0049] For example, in scenario 1: the patient's vital signs index is 3 (Level 1) and the danger index is 0 (within the safe zone); then the final warning level is Level 1. Scenario 2: The patient's vital signs index is 3 (Level 1), and the risk index is 1.0 (prolonged absence from bed); then the warning level is raised two levels to Level 3. Scenario 3: The patient's vital signs index is 8 (Level 3), and the risk index is 0.5 (short-term bed alighting); then the warning level is raised by one level to Level 4. Scenario 4: The patient's vital signs index is 10 (level 4), and the risk index is 1.0; if it is raised two levels and exceeds the highest level, the warning level will be directly set to level 5.

[0050] The advantages of this embodiment lie in the scientific integration of multi-source risks and the precise allocation of resources; the objective and authoritative physiological assessment, using the internationally recognized NEWS2 model (UK National Early Warning Score 2nd Edition), standardizes and scientizes vital sign assessment, improving the accuracy and clinical acceptance of physiological risk judgment; the early warning decision rule (risk index increases the basic level) is a key innovation, realizing the dynamic weighted superposition of behavioral risks and physiological risks, rather than a simple addition; this ensures that extreme high-risk situations, such as critical physiological condition and unauthorized departure from bed, can be instantly identified and escalated to the highest alert level, demonstrating the system's high sensitivity to complex risks.

[0051] In some embodiments of this application, the execution method of the above-mentioned graded early warning prompts has been optimized, specifically: Warning Level 1: No proactive alerts, patient's wristband displays a solid green light, and the bed is marked in green on the nurse station's large screen; Warning Level 2: The wristband is constantly lit in green, the large screen is marked in green and records low-risk observations; Warning Level 3: The wristband flashes yellow, the large screen displays a yellow marker and pops up the patient's information card; Warning Level 4: The wristband flashes red, and red pop-up windows appear simultaneously on the large screen, the responsible nurse's PDA, and the smart wristband, vibrating continuously until confirmed; Warning Level 5: In addition to all the prompts in Level 4, the nurse station will simultaneously activate a buzzer alarm.

[0052] This embodiment uses a tiered warning system to precisely match the clinical workflow. Low-level warnings only provide status alerts without interrupting the work, while high-level warnings force intervention through multimodal (visual, tactile, and auditory) and multi-terminal (large screen, mobile device) methods to ensure that emergency information is delivered. This systematically reduces alarm noise in normal situations and ensures efficient and accurate response during high-risk events, which is the ultimate guarantee for improving overall clinical safety.

[0053] In addition, it is necessary to statistically analyze the accuracy and false alarm rates of alarms in order to avoid false alarms that could increase the workload of medical staff.

[0054] False alarm rate = Number of false alarms / Total number of alarms. The number of false alarms and total number of alarms are counted within 24 hours. The total number of alarms includes both false alarms and true alarms. If the false alarm rate exceeds the set false alarm threshold, such as 5%, the system is considered to have a significant error and requires system calibration.

[0055] When the inpatient triage management method of this application is actually implemented, it can be carried out according to the following steps; 1. System initialization and patient record creation After a new patient is admitted, nursing staff enters their personal information (identity, condition, self-care ability, tubes / tubes, fall risk, suicide risk) through the system terminal or automatically obtains it from the hospital information system (HIS). At the same time, the patient is fitted with a medical smart bracelet (wearable device) and linked to the patient's file. The system automatically generates an initial electronic fence (excluding toilets and balconies) based on the patient's ward. 2. Personalized baseline risk assessment and fence setting The system activates the risk assessment module based on the entered personal information. This module scores the patient's condition and self-care ability in five dimensions according to a preset quantitative scoring table, and takes the highest score as the patient's risk index. Then, the electronic fence engine applies preset threshold rules (such as a=0.3, b=0.7) based on this risk index: expand the initial fence by n meters for low-risk patients, keep it unchanged for medium-risk patients, and shrink it by m meters for high-risk patients, thereby generating a unique modified electronic fence that matches the patient's current condition. 3. Real-time dynamic monitoring and risk calculation During the patient's hospitalization, the system performs dual-thread real-time monitoring: Behavior monitoring thread: The smart bracelet periodically reports real-time location information; the system compares it with the patient's modified electronic fence; if the patient is inside the fence, the danger index is 0; if the patient is outside the fence, a timer is started; if the preset time threshold T (e.g., 120 seconds) is not exceeded, the danger index is 0.5; if the threshold T is exceeded, the danger index is updated to 1.0. Physiological monitoring thread: The smart bracelet continuously collects vital signs information such as heart rate, blood oxygen, and body temperature, and uploads it; the system uses the NEWS2 model to standardize and score these indicators, and adds up the scores to obtain the real-time updated vital signs index; 4. Multi-source risk fusion and early warning decision-making The early warning decision engine continuously receives hazard indices and vital sign indices. Its decision-making logic is as follows: First, a basic risk level is determined based on the vital sign index against the NEWS2 standard (e.g., 0-4 points is one level). Then, this basic level is upgraded based on the hazard index: a hazard index of 0.5 upgrades it by one level, and 1.0 upgrades it by two levels. Finally, a comprehensive early warning level is determined from level one (lowest) to level five (highest). This rule ensures that the coupled risk of physiological and behavioral abnormalities can be amplified and identified. 5. Tiered Response and Early Warning Execution Based on the final determined warning level, the system automatically executes the preset response protocol: Level 1 / 2: Information prompts; patient status color (green) is updated only on the electronic whiteboard at the nurse station, without causing any active interference; Level 3: Active visual cue; the patient's wristband flashes yellow, and a yellow pop-up window appears on the nurses' station screen to attract the nurses' attention; Level 4: Mandatory intervention prompt; The patient's wristband flashes red, and red alarms and vibrations simultaneously pop up on the nurse station's large screen, the responsible nurse's PDA, and the wristband, requiring immediate action; Level 5: Emergency Alarm; Based on Level 4, activate the audible alarm at the nurses' station, broadcast throughout the hospital, and initiate the highest level of emergency response; 6. Solution closed-loop and continuous optimization After receiving an alert, the nurse goes to handle the situation and confirms the handling on the system. Once the alert is lifted, the system resumes normal monitoring. The patient's personal information (such as condition and self-care ability) will be updated with nursing assessments, thereby triggering the recalculation of risk indices and the correction of electronic fences, achieving a dynamic and adaptive closed loop in the entire management method.

[0056] In addition, this application also relates to an inpatient triage management system, which includes: Wearable device: A medical smart bracelet integrating a UWB / Bluetooth positioning module, heart rate and blood oxygen sensor, accelerometer and display screen; responsible for collecting the patient's real-time location information and vital signs information, and receiving instructions to provide light prompts; Processing unit: Deployed on hospital servers or in the cloud, including: Risk assessment module: Performs risk scoring and index calculation as described in Example 2; Electronic fence engine: Executes the fence generation and dynamic correction logic in Examples 2 and 3; Location Analysis Module: Performs the hazard index calculation as described in Example 3; Vital signs analysis module: Performs NEWS2 score calculation as described in Example 4; Early warning decision engine: Executes the early warning level fusion decision in Example 4; Early warning unit: includes a central electronic whiteboard (display screen) at the nurse station, PDAs (mobile nursing terminals) provided to medical staff, smart bracelets, and sound and light alarms in the ward area; this unit receives early warning level instructions issued by the processing unit and triggers corresponding graded prompt operations; When the system is running, the wearable device uploads data to the processing unit. After the processing unit completes all analysis and calculations, it sends the warning command to the corresponding device in the warning unit, thus completing the closed loop from perception to warning.

[0057] This embodiment integrates the innovative method into a specific physical device. Its advantage lies in providing a complete and feasible solution, clearly defining the three core hardware components required and their collaborative relationships, and providing a clear blueprint for the manufacturing, deployment, and application of the product. From a patent protection perspective, the device claims and method claims form a three-dimensional protection, enhancing the stability and strength of the patent rights.

[0058] The foregoing has shown and described the basic principles, main features, and advantages of this application. Those skilled in the art should understand that this application is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of this application. Various changes and modifications can be made to this application without departing from the spirit and scope thereof, and all such changes and modifications fall within the scope of this application as claimed. The scope of protection of this application is defined by the appended claims and their equivalents.

Claims

1. A method for hierarchical management of hospitalized patients, characterized in that: include, Obtain personal information collected from hospitalized patients via wearable devices, and determine the patient's risk index within the current sampling period based on the personal information; Based on the risk index, the preset initial electronic fence is modified to obtain a modified electronic fence; The wearable device collects the patient's location information in real time, and calculates the patient's risk index based on the relationship between the location information and the location of the modified electronic fence. Based on the vital signs information in the aforementioned personal information, obtain the patient's vital signs index; The warning level for the patient is determined based on the risk index and the vital signs index. Based on the aforementioned warning level, the corresponding warning prompt will be executed.

2. The method for hierarchical management of inpatients according to claim 1, characterized in that: The personal information includes the patient's identity information, medical condition information, self-care ability information, tubing information, fall risk information, suicide risk information, and vital signs information; the vital signs information includes one or more of, for example, heart rate, blood oxygen saturation, and body temperature. The process of determining the patient's risk index based on personal information includes: scoring the patient's medical condition information, self-care ability information, tube information, fall risk information, and suicide risk information, and determining the highest risk score as the patient's risk index.

3. The method for hierarchical management of inpatients according to claim 2, characterized in that: The method for risk scoring of various information items includes: Condition information score: Stable condition, stable vital signs and no acute symptoms, score 0; Mild condition, requires routine monitoring and no immediate risk, score 0.3; Severe condition and requires close monitoring or intervention, score 0.6; Critical condition and requires emergency intervention or ICU monitoring, score 0.8; Life-threatening condition with multiple organ failure or end-stage condition, score 1.

0. Self-care ability assessment: Completely self-reliant, 0 points; Mildly dependent, 0.2 points; Moderately dependent, 0.5 points; Severely dependent, 0.8 points; Completely dependent, 1.0 point; Piping information scoring: No piping, 0 points; Low-risk piping, 0.2 points; Medium-risk piping, 0.5 points; High-risk piping, 0.8 points; Extremely high-risk piping, 1.0 point; Fall risk information scoring: No history of falls, able to move freely and conscious, score 0; Low risk, slightly unsteady gait but no assistance required, score 0.3; Medium risk, using a walking aid, elderly or visually impaired, score 0.6; High risk, history of frequent falls, confusion or orthostatic hypotension, score 0.8; Very high risk, delirium, severe dizziness, severe limb weakness or paralysis, score 1.0; Suicide Risk Information Score: No suicidal ideation or history, stable mood, 0 points; Mild risk, occasional negative thoughts but no plans or behaviors, 0.4 points; Moderate risk, suicidal thoughts or history but no recent plans, 0.7 points; High risk, clear plans or preparatory behaviors, 0.9 points; Very high risk, recent suicide attempts, strong suicidal intentions or command hallucinations, 1.0 point.

4. The method for hierarchical management of inpatients according to claim 3, characterized in that: The modification of the initial electronic fence based on the risk index includes: when the risk index is not greater than a first threshold a, the patient's risk level is determined to be level one, and the initial electronic fence is extended outward by a first distance n to form the modified electronic fence; when the risk index is greater than a and not greater than a second threshold b, the risk level is determined to be level two, and the initial electronic fence is directly used as the modified electronic fence; when the risk index is greater than b, the risk level is determined to be level three, and the initial electronic fence is contracted inward by a second distance m to form the modified electronic fence; wherein, 0 < a < b ≤ 1, and n and m are preset distance values ​​greater than 0.

5. The method for hierarchical management of inpatients according to claim 1, characterized in that: The initial electronic fence is an area defined based on the patient's ward, excluding toilets and balconies.

6. The method for hierarchical management of inpatients according to claim 1, characterized in that: The calculation of the risk index based on location information and modified electronic fence includes: if the patient's location is within the range of the modified electronic fence, the risk index is 0; if the patient's location is outside the range of the modified electronic fence, but the stay time does not exceed a preset time threshold T, the risk index is 0.5; if the stay time exceeds the preset time threshold T, the risk index is 1.

0.

7. The method for hierarchical management of inpatients according to claim 1, characterized in that: The method for obtaining the patient's vital sign index based on the vital sign information in the patient's personal information includes: obtaining the patient's respiratory rate, oxygen saturation, oxygen inhalation status, level of consciousness, systolic blood pressure, heart rate and body temperature from the patient's vital sign information, scoring the above physiological indicators of the patient according to the NEWS2 model, and obtaining the patient's vital sign index.

8. The method for hierarchical management of inpatients according to claim 7, characterized in that: The method of determining the warning level based on the risk index and vital sign index includes: pre-defining five progressively increasing risk levels (first to fifth) according to the numerical range of the vital sign index; determining the patient's baseline risk level based on the vital sign index; if the risk index is 0, then the baseline risk level is used as the warning level; if the risk index is 0.5, then the baseline risk level is increased by one level as the warning level; if the risk index is 1.0, then the baseline risk level is increased by two levels as the warning level; if the increased level exceeds the fifth risk level, then the warning level is determined to be the highest level.

9. A method for hierarchical management of inpatients according to claim 8, characterized in that: The warning prompts will be executed as follows based on the warning level: When the warning level is the highest risk level, no proactive warning will be issued. When the warning level is the second risk level, the wearable device will be displayed in solid green and the status will be recorded on the electronic whiteboard at the nurse station. When the warning level is the third risk level, the wearable device will flash yellow and record the status on the electronic whiteboard at the nurse station. When the warning level is the fourth risk level, the wearable device will flash red and pop up and vibrate to alert the nurse station electronic whiteboard, PDA and medical staff wristbands. When the warning level is the fifth risk level, the wearable device will flash red, and pop-up windows and vibration prompts will appear on the electronic whiteboard at the nurse station, PDA, and medical staff wristbands, and an audible alarm will be activated simultaneously.

10. A triage management system for inpatients, characterized in that: The system for performing the method as described in any one of claims 1 to 9 includes: Wearable devices are used to collect personal information and real-time location information of hospitalized patients; The processing unit is configured as follows: The patient's risk index is determined based on the aforementioned personal information; The initial electronic fence is modified based on the risk index to obtain a modified electronic fence; Based on the relationship between the real-time location information and the modified electronic fence, the patient's risk index is calculated; Based on the vital signs information in the aforementioned personal information, obtain the patient's vital signs index; The patient's warning level is determined based on the risk index and the vital signs index; The early warning unit is used to trigger a corresponding early warning notification based on the early warning level.

Images