A multi-modal sign-based hemodialysis complication early warning system and method

Abstract

The application discloses a hemodialysis complication early warning system and method based on multi-modal signs, relates to the technical field of hemodialysis monitoring, and comprises the following steps: determining the risk of cerebral tissue perfusion deficiency, the risk of peripheral tissue perfusion deficiency and the risk of muscle spasm according to the occurrence time of sign sets and tissue oxygen change; determining the ultrafiltration risk according to the lower limb tissue oxygen change, cumulative ultrafiltration volume and ultrafiltration rate in the sign set; identifying chest tightness according to the sign set to form a risk set; dividing early warning levels according to the risk set; matching safety measures from preset nursing rules according to the early warning levels, nursing permissions and medical order permissions to form nursing matters and medical order matters; and recording the actual execution results and execution time of the nursing matters to form a treatment record; and the individual oxygen baseline of the forehead and both lower limbs is established by using the tissue oxygen saturation in the initial stage of hemodialysis, so that early tissue perfusion change and subsequent clinical manifestations can be corresponded in the same time sequence information chain.

Description

Technical Field

[0001] This invention relates to the field of hemodialysis monitoring technology, and in particular to a hemodialysis complication early warning system and method based on multimodal signs. Background Technology

[0002] During hemodialysis, the patient's blood volume, peripheral perfusion, and electrolyte status are constantly changing. Clinically, complications are usually monitored by combining blood pressure, heart rate, ultrafiltration volume, ultrafiltration rate, equipment operating parameters, and symptoms and signs. Near-infrared tissue oxygenation monitoring and electrocardiogram monitoring are being gradually introduced to evaluate the perfusion status of the brain and peripheral tissues. Related early warning methods mostly use fixed thresholds, trend comparisons, or multi-parameter joint interpretation to provide risk warnings for abnormalities such as hypotension, inadequate perfusion, muscle spasms, and chest tightness.

[0003] Current technologies still have shortcomings. Monitoring methods that rely primarily on single thresholds or discrete sign interpretations struggle to address the sequential and persistent relationship between tissue oxygenation changes and symptoms, easily leading to a disconnect between early perfusion changes and subsequent clinical manifestations. Procedures that rely mainly on manual intervention after alarm triggering typically lack a continuous correlation between intervention time, tissue oxygenation recovery, and sign withdrawal, making it difficult to establish a verifiable closed loop between risk assessment and intervention response. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a method for early warning of hemodialysis complications based on multimodal signs to solve the problem of insufficient temporal correlation of multimodal abnormal information.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a method for early warning of hemodialysis complications based on multimodal signs, comprising, Continuously collect patient vital signs, tissue oxygen saturation, cumulative ultrafiltration volume, ultrafiltration rate, nursing authority and medical order authority, and establish individual oxygen saturation lines based on tissue oxygen saturation at the beginning of hemodialysis; Individual oxygen line and tissue oxygen saturation are compared to obtain tissue oxygen changes. Patient signs are correlated with tissue oxygen changes according to the time of occurrence to form a sign set. The risk of insufficient brain perfusion, insufficient peripheral tissue perfusion, and muscle spasm is determined based on the timing of the occurrence of physical signs and tissue oxygenation changes. The risk of ultrafiltration is determined based on the changes in lower limb tissue oxygenation, cumulative ultrafiltration volume, and ultrafiltration rate in the physical signs. Chest tightness is also identified based on the physical signs, thus forming a risk set. The warning level is divided according to the risk set. Based on the warning level, nursing authority and medical order authority, safety measures are matched from the preset nursing rules to form nursing matters and medical order matters. The actual execution results and execution time of the nursing matters are recorded to form a treatment record. The treatment records are used to assess the tissue oxygenation recovery in the forehead, left lower limb, and right lower limb, as well as the exit of corresponding vital signs. At the same time, the corresponding risks of the risk concentration are verified, and verification records are generated.

[0007] As a preferred embodiment of the hemodialysis complication early warning method based on multimodal signs described in this invention, wherein: establishing individual oxygen lines includes, The patient's physical signs include staring, slow reaction, yawning, changes in facial expression, cold sweat, body surface temperature, chest tightness, heart rate, forced sitting up, and muscle condition. Tissue oxygen saturation of the forehead, left lower limb, and right lower limb at the beginning of hemodialysis was collected to form three initial oxygen sequences; After removing the tissue oxygen saturation corresponding to sensor detachment, contact abnormalities, and invalid signals from the three initial oxygen sequences, three effective oxygen sequences were obtained. The median of the three effective oxygen sequences was then used as the individual oxygen line for the frontal limb, left lower limb, and right lower limb.

[0008] As a preferred embodiment of the hemodialysis complication early warning method based on multimodal signs described in this invention, wherein: obtaining tissue oxygenation changes includes, The difference between the prefrontal tissue oxygen saturation and the individual prefrontal oxygen line is compared to obtain the prefrontal oxygen change. When the prefrontal oxygen change reaches the preset decrease condition and continues for the preset duration, the prefrontal oxygen change is marked as a prefrontal decrease. The tissue oxygen saturation of the left and right lower limbs was divided into continuous observation segments, and the median tissue oxygen saturation of each continuous observation segment was calculated to obtain the lower limb oxygen sequence. When the oxygen sequence of the ipsilateral lower limb continues to decrease and reaches the preset decrease condition, the corresponding lower limb oxygen change is marked as lower limb decrease, and the decrease in forehead and lower limb oxygen is regarded as tissue oxygen change.

[0009] As a preferred embodiment of the hemodialysis complication early warning method based on multimodal signs described in this invention, wherein: the correlation between patient signs and tissue oxygenation changes according to the time of occurrence includes, Record changes in tissue oxygenation and the occurrence time of patient signs to form a sign time series. Divide the sign time series into multiple sign observation segments according to the preset observation duration. Changes in tissue oxygenation occurring within the same observation segment, along with staring, slowed reflexes, yawning, changes in facial expression, cold sweats, body surface temperature, chest tightness, heart rate, forced sitting up, and muscle condition, are grouped into observation segment sign groups and arranged in chronological order to form sign sets.

[0010] As a preferred embodiment of the hemodialysis complication early warning method based on multimodal signs described in this invention, wherein: the determination of the risk of insufficient brain perfusion, the risk of insufficient peripheral tissue perfusion, and the risk of muscle spasm includes, From the set of vital signs, forehead descent and gaze and yawning within the same observation segment were selected, and the first time interval between the occurrence of forehead descent and gaze and the second time interval between the occurrence of forehead descent and yawning were calculated respectively. When at least one of the first and second time intervals is within a preset time correlation range, frontal descent persists, and at least one of the following—gazing, slowed reaction, and yawning—occurs within the observation period of signs where frontal descent occurs, the patient is deemed to be at risk of insufficient brain perfusion. The third time interval between the time of lower limb descent and the time of facial changes, the fourth time interval between the time of lower limb descent and the time of cold sweating, and the fifth time interval between the time of lower limb descent and the time of sudden drop in body surface temperature were calculated separately. When at least one of the third, fourth, and fifth time intervals is within a preset time correlation range, the lower limb descent persists, and at least two of the following—facial changes, cold sweats, and a sudden drop in body surface temperature—occur within the observation period of the signs of lower limb descent, the patient is deemed to be at risk of peripheral tissue insufficiency. If the descent of the lower limb persists and abnormal contraction of the muscles on the same side occurs, it is determined that there is a risk of muscle spasm in the corresponding lower limb.

[0011] As a preferred embodiment of the hemodialysis complication early warning method based on multimodal signs described in this invention, wherein: the determination of ultrafiltration risk includes, The cumulative ultrafiltration volume and ultrafiltration rate were extracted according to the physical signs observed in the lower limbs to obtain the ultrafiltration observation group; The cumulative ultrafiltration change is calculated based on the cumulative ultrafiltration amount at adjacent time points in the ultrafiltration observation group, and the cumulative ultrafiltration change and ultrafiltration rate are arranged according to the corresponding time points to obtain the ultrafiltration change sequence; By comparing the recording time of the ultrafiltration change sequence with the occurrence time of lower limb descent, the cumulative ultrafiltration change, ultrafiltration rate, and lower limb descent status falling within the same sign observation segment are merged into an ultrafiltration oxygen change group. When the cumulative ultrafiltration change and ultrafiltration rate in the ultrafiltration oxygen change group both reach the preset ultrafiltration conditions and the lower limb descent persists, the patient is determined to have ultrafiltration risk.

[0012] As a preferred embodiment of the hemodialysis complication early warning method based on multimodal signs described in this invention, wherein: the identification of chest tightness includes, The heart rate state during the occurrence of chest tightness is divided into a slow heart rate state and a fast heart rate state, which is the chest tightness heart rate state. When the chest tightness and heart rate state is a bradycardia state, obtain the 12-lead ECG information and blood potassium information, compare the 12-lead ECG information and blood potassium information with the preset hyperkalemia conditions to obtain the hyperkalemia judgment result, confirm the hyperkalemia exclusion result based on the doctor's diagnosis information, and correlate the hyperkalemia exclusion result with tissue oxygenation changes, facial appearance changes and cold sweats over time to obtain the first chest tightness result. When the chest tightness and heart rate state is an tachycardia state, correlate the forced sitting up with the chest tightness over time to obtain the heart failure judgment result. Based on the doctor's diagnosis information, the heart failure exclusion result is confirmed. The heart failure exclusion result is correlated with the ST segment changes and tissue oxygenation changes in the twelve-lead electrocardiogram information over time to obtain the second chest tightness result. The first chest tightness result and the second chest tightness result are used as the chest tightness differential result. Together with the risks of insufficient brain tissue perfusion, insufficient peripheral tissue perfusion, muscle spasm, and ultrafiltration, a risk set is formed.

[0013] As a preferred embodiment of the hemodialysis complication early warning method based on multimodal signs described in this invention, the matching safety measures include: According to the warning level, the corresponding safety measures are selected from the preset nursing rules to obtain candidate measures. The nursing authority is used to verify the authority of the candidate measures. The candidate measures that can be executed by the nursing authority are marked as nursing matters. The doctor's order authority is used to verify the authority of the candidate measures that are not marked as nursing matters. The candidate measures that require doctor's confirmation are marked as doctor's order matters. Record the actual results and execution time of nursing matters, and combine the actual results, execution time, nursing matters and medical orders into a treatment record.

[0014] As a preferred embodiment of the hemodialysis complication early warning method based on multimodal signs described in this invention, wherein: the corresponding risks in the verification risk set include, The execution time in the treatment record is used as the starting point for response observation. The tissue oxygen saturation, corresponding signs, cumulative ultrafiltration volume and ultrafiltration rate of the forehead, left lower limb and right lower limb after the starting point of response observation are used to form a response sequence. The tissue oxygen saturation of the forehead, left lower limb and right lower limb in the response sequence is compared with the individual oxygen line of the forehead, the individual oxygen line of the left lower limb and the individual oxygen line of the right lower limb, respectively, to obtain the tissue oxygen recovery status. According to the response sequence, check the persistence status of the corresponding signs, mark the corresponding signs that stop appearing as exit signs, and obtain the sign exit status. Then, check the tissue oxygenation recovery status and the sign exit status with the corresponding risks in the risk set to obtain the risk response status. When tissue oxygen saturation recovers to the corresponding individual oxygen line and the corresponding signs disappear, the risk response status is marked as consistent response; when tissue oxygen saturation continues to decrease and the corresponding signs persist, the risk response status is marked as inconsistent response. Corresponding the risk response status with the handling records over time creates a verification record.

[0015] Secondly, the present invention provides a hemodialysis complication early warning system based on multimodal signs, comprising, The baseline generation module continuously collects patient vital signs, tissue oxygen saturation, cumulative ultrafiltration volume, ultrafiltration rate, nursing permissions, and medical order permissions, and establishes an individual oxygen baseline based on the tissue oxygen saturation at the beginning of hemodialysis. The vital signs association module compares individual oxygen lines and tissue oxygen saturation to obtain tissue oxygen changes, and associates patient vital signs with tissue oxygen changes according to the time of occurrence to form a vital signs set. The risk assessment module determines the risk of insufficient brain perfusion, insufficient peripheral tissue perfusion, and muscle spasm based on the occurrence time of physical signs and tissue oxygenation changes. It also determines the risk of ultrafiltration based on the changes in lower limb tissue oxygenation, cumulative ultrafiltration volume, and ultrafiltration rate in the physical signs, and identifies chest tightness based on the physical signs to form a risk set. The tiered treatment module classifies warning levels based on risk sets, matches safety measures from preset nursing rules based on warning level, nursing authority, and medical order authority, generates nursing and medical order items, and records the actual execution results and execution time of the nursing items to form a treatment record; The response verification module uses treatment records to determine the tissue oxygenation recovery status of the forehead, left lower limb, and right lower limb, as well as the exit status of corresponding vital signs. At the same time, it verifies the corresponding risks in the risk cluster and generates verification records.

[0016] The beneficial effects of this invention are as follows: By establishing individual oxygen saturation lines for the forehead and both lower limbs based on the tissue oxygen saturation at the beginning of hemodialysis, changes in tissue oxygenation, patient signs, and ultrafiltration status are made into a continuous basis for judgment according to their sequence and duration, so that early changes in tissue perfusion and subsequent clinical manifestations can be correlated in the same temporal information chain; in addition, by matching safety measures according to warning level, nursing authority, and medical order authority, risk assessment, treatment execution, and post-treatment response are made into a continuous correspondence, so that the corresponding risks can be verified based on the actual response of the patient after treatment. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a method for early warning of hemodialysis complications based on multimodal signs.

[0019] Figure 2 This is a schematic diagram of the structure of a hemodialysis complication early warning system based on multimodal signs.

[0020] Figure 3 A flowchart for the monitoring and management of risks associated with hemodialysis complications.

[0021] Figure 4 A schematic diagram illustrating the relationship between the entity set, risk set, and response verification. Detailed Implementation

[0022] 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.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0025] Reference Figures 1-4 This is one embodiment of the present invention, which provides a method for early warning of hemodialysis complications based on multimodal signs, including the following steps: S1: Continuously collect patient vital signs, tissue oxygen saturation, cumulative ultrafiltration volume, ultrafiltration rate, nursing authority and medical order authority, and establish individual oxygen saturation lines based on tissue oxygen saturation at the beginning of hemodialysis.

[0026] S1.1: After hemodialysis begins, the patient's vital signs are continuously collected, and the tissue oxygen saturation of the forehead, left lower limb and right lower limb is continuously collected. The cumulative ultrafiltration volume and ultrafiltration rate are continuously collected using the hemodialysis machine.

[0027] Patient signs include staring, slowed reaction, yawning, changes in facial expression, cold sweats, body temperature, chest tightness, heart rate, forced sitting up and muscle condition, violent struggling or grabbing at the throat (risk of suffocation).

[0028] Changes in facial appearance refer to changes in a patient's facial appearance relative to the state at the start of hemodialysis, which can be continuously observed or identified in terms of complexion, expression, or demeanor. Examples include paleness, cyanosis, dullness, apathy, frowning, and expressions of pain.

[0029] Based on the nursing execution requirements of each safety measure in the preset nursing rules, the execution authorization status corresponding to the safety measures that nursing staff are allowed to directly execute is recorded as nursing authority.

[0030] Based on the authorization records of each safety measure in the currently effective medical orders, the execution authorization status of the safety measures that have been authorized by the doctor's medical orders is recorded as the medical order permissions.

[0031] In one embodiment, the preset nursing rules are risk management rules pre-established based on the clinical treatment guidelines, nursing operation guidelines, and medical order management requirements for hemodialysis. Corresponding relationships are established according to risk type, warning level, safety measures, nursing execution requirements, and medical order execution requirements. Risk type is used to match the risk of insufficient brain perfusion, the risk of insufficient peripheral tissue perfusion, the risk of muscle spasm, the risk of ultrafiltration, and the identification results of chest tightness. Warning level is used to screen corresponding safety measures. Nursing execution requirements are used to determine whether the safety measures can be directly executed by nursing staff. Medical order execution requirements are used to determine whether the safety measures need to be confirmed by a doctor before execution.

[0032] S1.2: The start time of hemodialysis is taken as the starting time of tissue oxygen saturation. The tissue oxygen saturation of the forehead, left lower limb and right lower limb at the beginning of hemodialysis is extracted from the tissue oxygen saturation. The tissue oxygen saturation of the forehead, left lower limb and right lower limb are arranged in order of collection time to form three initial oxygen sequences.

[0033] Record the sensor connection and contact status when collecting tissue oxygen saturation data. When the sensor connection status changes from connected to disconnected, mark the collection time during the period when the connection status is disconnected as the collection time corresponding to the sensor detachment.

[0034] When the sensor connection status is "connected" and the contact status is "abnormal contact", the acquisition time during the period when the contact status is "abnormal contact" is marked as the acquisition time corresponding to the contact abnormality.

[0035] When tissue oxygen saturation values ​​are missing, tissue oxygen saturation acquisition fails, or tissue oxygen saturation acquisition status is invalid, the corresponding sampling result will be treated as an invalid signal, and the acquisition time corresponding to the invalid signal will be marked as the acquisition time corresponding to the invalid signal.

[0036] The tissue oxygen saturation at the acquisition time corresponding to sensor detachment, abnormal contact, and invalid signals was removed from the three initial oxygen sequences to obtain three effective oxygen sequences.

[0037] S1.3: Arrange the tissue oxygen saturation in the three effective oxygen sequences from largest to smallest, and calculate the median tissue oxygen saturation of the forehead, left lower limb, and right lower limb respectively.

[0038] The median oxygen saturation of the frontal tissue was used as the individual oxygen saturation line for the frontal tissue, the median oxygen saturation of the left lower limb tissue was used as the individual oxygen saturation line for the left lower limb, and the median oxygen saturation of the right lower limb tissue was used as the individual oxygen saturation line for the right lower limb.

[0039] S2: Compare individual oxygen lines and tissue oxygen saturation to obtain tissue oxygen changes, and associate patient signs with tissue oxygen changes according to the time of occurrence to form a sign set.

[0040] S2.1: Compare the prefrontal tissue oxygen saturation at each collection time with the prefrontal individual oxygen line. The difference between the prefrontal individual oxygen line and the prefrontal tissue oxygen saturation is taken as the prefrontal oxygen decrease. The ratio of the prefrontal oxygen decrease to the prefrontal individual oxygen line is taken as the prefrontal decrease amplitude. Arrange the prefrontal decrease amplitudes at each collection time according to the collection time to obtain the prefrontal oxygen change.

[0041] When the prefrontal oxygen level drops to the preset threshold, the start time of the drop is recorded. When the prefrontal oxygen level drops to the preset threshold and remains at the preset threshold for a preset duration, the change in prefrontal oxygen level from the start time of the drop to the end of the continuous holding period is marked as a prefrontal drop.

[0042] The preset decline condition refers to the decline in prefrontal tissue oxygen saturation relative to the prefrontal individual oxygen line reaching a preset decline range.

[0043] The preset duration refers to the time during which the oxygen saturation of the prefrontal tissue remains at the preset decline condition after it reaches the preset decline condition. The preset decline condition and preset duration are determined in advance based on hemodialysis nursing experience and clinical validation results.

[0044] In one embodiment, the process for determining the preset decrease conditions and preset duration is as follows: Data on changes in prefrontal tissue oxygen saturation under normal hemodialysis conditions are collected. After establishing the prefrontal individual oxygen line, the change in prefrontal tissue oxygen saturation relative to the prefrontal individual oxygen line is statistically analyzed. According to the statistics, the decrease in prefrontal tissue oxygen saturation relative to the prefrontal individual oxygen line under normal hemodialysis conditions is mainly distributed in the range of 0% to 8%, and the continuous maintenance time does not exceed 20 seconds. Therefore, tissue oxygen changes with a decrease of no more than 8% and a duration of no more than 20 seconds are considered as normal fluctuations.

[0045] Prefrontal tissue oxygen saturation was collected from patients exhibiting signs of staring and yawning during hemodialysis. Thresholds for the magnitude of the decrease were set at 10%, 12%, 15%, 18%, and 20%, and thresholds for the duration were set at 20 seconds, 30 seconds, 45 seconds, and 60 seconds. For each candidate condition, the trigger time for the prefrontal tissue oxygen change was recorded, and the time difference between the trigger time and the occurrence of staring or yawning was calculated. Simultaneously, the proportion of false triggers during normal dialysis was statistically analyzed. A candidate condition was deemed valid if the time difference corresponding to the candidate condition was greater than 0 seconds and the proportion of false triggers during normal dialysis did not exceed 5%. After screening the candidate conditions, the preset decrease condition was a 15% decrease in prefrontal tissue oxygen saturation relative to the prefrontal body oxygen line, with a preset duration of 30 seconds.

[0046] S2.2: Divide the tissue oxygen saturation of the left and right lower limbs into continuous observation segments with continuous time and no overlap according to the same preset observation time. Calculate the median of tissue oxygen saturation in each continuous observation segment and arrange the median of tissue oxygen saturation in the time order of the continuous observation segments to obtain the lower limb oxygen sequence corresponding to the left lower limb and the lower limb oxygen sequence corresponding to the right lower limb.

[0047] The median tissue oxygen saturation of adjacent time intervals in the ipsilateral lower limb oxygen sequence is compared sequentially. When the median tissue oxygen saturation of multiple consecutive observation segments decreases sequentially, it is determined that the ipsilateral lower limb oxygen sequence is continuously decreasing. The median tissue oxygen saturation during the continuous decrease is compared with the corresponding individual oxygen line of the left or right lower limb. When the decrease of the median tissue oxygen saturation relative to the corresponding individual oxygen line reaches the preset decrease condition, the oxygen change of the corresponding lower limb is marked as a decrease in the lower limb.

[0048] In one embodiment, the preset observation duration is 1 minute, and multiple consecutive observation segments are defined as 3 consecutive observation segments. When the median tissue oxygen saturation of adjacent consecutive observation segments decreases by no less than 0.5%, the corresponding decrease is considered as a reduction in the median tissue oxygen saturation.

[0049] The descent of the forehead, the descent of the lower limb corresponding to the left lower limb, and the descent of the lower limb corresponding to the right lower limb were recorded according to the time of occurrence to form changes in tissue oxygenation.

[0050] S2.3: During continuous collection of tissue oxygen saturation, the patient's vital signs, including staring, slow reaction, yawning, changes in facial expression, cold sweats, body surface temperature, chest tightness, heart rate, forced sitting up, and muscle status, should be recorded simultaneously, and the occurrence time of staring, slow reaction, yawning, changes in facial expression, cold sweats, body surface temperature, chest tightness, heart rate, forced sitting up, and muscle status should be recorded.

[0051] In one embodiment, a camera is used to continuously capture the patient's facial and behavioral states. Facial states include staring, yawning, and changes in facial color. Behavioral states include forced sitting up and muscle states. When the patient's gaze remains fixed for 5 seconds, the start time of the staring is recorded as the staring occurrence time. When the patient yawns 3 times within 5 minutes, the occurrence time of the 3rd yawn is recorded as the yawning occurrence time.

[0052] Issue preset interactive commands to the patient and record the patient's response time to complete the preset interactive commands. When the patient's response time is longer than 5 seconds compared to the individual's normal response state, or when the patient is unable to complete the preset interactive commands within 10 seconds but maintains spontaneous breathing and eye-opening, the corresponding state is recorded as sluggish response, and the time of occurrence of sluggish response is recorded.

[0053] Observe the patient's facial color changes and skin sweating. When the patient's facial color changes from normal to pale, record the time when the paleness begins as the time of occurrence. When the patient experiences cold sweats, record the time when the cold sweats begin as the time of occurrence. Continuously collect the patient's body surface temperature. When the body surface temperature drops by more than 0.5℃ within 5 minutes, record the time when the body surface temperature begins to drop as the time of occurrence.

[0054] Record the patient's chief complaint of chest tightness. When the patient reports chest tightness, record the time of the report as the time of occurrence of chest tightness. Continuously collect heart rate data. When the heart rate is below 60 beats / minute, record the heart rate status as a bradycardia state. When the heart rate is above 100 beats / minute, record the heart rate status as a tachycardia state. Record the time of occurrence of the bradycardia and tachycardia states.

[0055] Observe the patient's changes in body position. When the patient changes from a supine or semi-recumbent position to a sitting position and remains there for 30 seconds, record the time when the patient begins to sit up as the time when the patient is forced to sit up. When the patient experiences abnormal muscle contraction in the left or right lower limb for 3 seconds, record the time when the abnormal contraction begins as the time when the patient experiences the muscle state.

[0056] S2.4: Arrange the changes in tissue oxygenation, staring, sluggishness, yawning, changes in facial expression, cold sweats, body surface temperature, chest tightness, heart rate, forced sitting up, and muscle state in chronological order of occurrence to form a time sequence of vital signs.

[0057] The observation time sequence is divided into segments based on the preset observation duration, starting from the beginning time of the vital signs time sequence. This ensures that the end time of the previous observation segment is continuously connected to the start time of the next observation segment, resulting in multiple observation segments arranged in chronological order.

[0058] S2.5: Changes in tissue oxygenation, staring, sluggishness, yawning, changes in facial expression, cold sweats, body surface temperature, chest tightness, heart rate, forced sitting up, and muscle status occurring between the start and end times of the same sign observation segment are classified into the same sign observation segment. Changes in tissue oxygenation, staring, yawning, changes in facial expression, cold sweats, body surface temperature, chest tightness, heart rate, forced sitting up, and muscle status within the same sign observation segment are combined into a sign group for the observation segment.

[0059] The observation segments are arranged in order of their start time to form a set of vital signs.

[0060] S3: Based on the timing of the occurrence of physical signs and tissue oxygenation changes, determine the risk of insufficient brain perfusion, insufficient peripheral tissue perfusion, and muscle spasm. Based on the changes in lower limb tissue oxygenation, cumulative ultrafiltration volume, and ultrafiltration rate in the physical signs, determine the risk of ultrafiltration. Based on the physical signs, differentiate chest tightness and form a risk set.

[0061] S3.1: Select forehead descent and gaze and yawning within the same observation segment from the set of vital signs. Calculate the absolute value of the time difference between the occurrence time of forehead descent and gaze, and the absolute value of the time difference between the occurrence time of forehead descent and yawning. Take the absolute value of the time difference between the occurrence time of forehead descent and gaze as the first time interval, and take the absolute value of the time difference between the occurrence time of forehead descent and yawning as the second time interval.

[0062] When only one of the following two physical signs appears within the same observation segment: gazing or yawning, the time interval between forehead descent and the actual appearance of the patient's physical sign is calculated, and the time interval corresponding to the absence of the patient's physical sign is marked as an invalid time interval. Invalid time intervals are not included in the comparison of the preset time correlation range.

[0063] When at least one of the first and second time intervals is within a preset time correlation range, forehead descent persists, and at least one of the following—gazing, slowed reaction, and yawning—occurs within the observation period of the signs corresponding to forehead descent, the patient is deemed to be at risk of insufficient brain perfusion.

[0064] It should also be noted that staring and slowed reaction are non-specific signs of neurological changes. Staring and slowed reaction on their own cannot be used as a basis for disease diagnosis, but rather as signals of changes in patient status, which are used together with changes in tissue oxygenation, patient signs, and the time of occurrence to indicate risk.

[0065] The absolute time difference between the occurrence of lower limb descent and the occurrence of facial changes, the absolute time difference between the occurrence of lower limb descent and the occurrence of cold sweat, and the absolute time difference between the occurrence of lower limb descent and the occurrence of a sudden drop in body surface temperature were calculated separately. The absolute time difference between the occurrence of lower limb descent and the occurrence of facial changes was taken as the third time interval, the absolute time difference between the occurrence of lower limb descent and the occurrence of cold sweat was taken as the fourth time interval, and the absolute time difference between the occurrence of lower limb descent and the occurrence of a sudden drop in body surface temperature was taken as the fifth time interval.

[0066] When not all changes in facial features, cold sweats, and sudden drops in body surface temperature occur within the same observation period, the time interval between the descent of the lower limbs and the actual occurrence of the patient's signs is calculated. The time intervals corresponding to the absence of patient signs are marked as invalid time intervals, and invalid time intervals are not included in the comparison of the preset time correlation range.

[0067] When at least one of the third, fourth, and fifth time intervals falls within a preset time correlation range, the lower limb descent persists, and at least two of the following—facial changes, cold sweats, and a sudden drop in body surface temperature—occur within the observation period of the lower limb descent, the patient is deemed to be at risk of peripheral tissue insufficiency.

[0068] Peripheral tissue inadequacy risk refers to the risk of decreased blood supply to peripheral tissues such as the lower extremities during hemodialysis due to ultrafiltration, changes in circulating volume, or insufficient perfusion.

[0069] In one embodiment, the preset time correlation range is 0 to 5 minutes. The preset time correlation range can be adjusted according to the experience of nursing experts and the results of clinical verification. When multiple patients' vital signs appear within the same time range, the occurrence time of each patient's vital signs is recorded separately and used as a basis for risk assessment.

[0070] The lower limb descent and muscle status were examined in chronological order according to the time of the observation period of the physical signs. When the descent of the left lower limb was persistent and the muscle status of the left lower limb showed abnormal contraction, it was determined that there was a risk of muscle spasm in the left lower limb. When the descent of the right lower limb was persistent and the muscle status of the right lower limb showed abnormal contraction, it was determined that there was a risk of muscle spasm in the right lower limb.

[0071] The cumulative ultrafiltration volume and ultrafiltration rate were extracted according to the physical signs observed in the lower limbs. The cumulative ultrafiltration volume and ultrafiltration rate were then arranged in order of collection time to obtain the ultrafiltration observation group.

[0072] The difference between the cumulative ultrafiltration rate at the next acquisition time and the cumulative ultrafiltration rate at the previous acquisition time is taken as the cumulative ultrafiltration change at the next acquisition time. The cumulative ultrafiltration change and the ultrafiltration rate at the same acquisition time are arranged from first to last according to the acquisition time to obtain the ultrafiltration change sequence.

[0073] The recording times of each cumulative ultrafiltration change and ultrafiltration rate in the ultrafiltration change sequence, and the cumulative ultrafiltration change, ultrafiltration rate and lower limb decrease within the same vital sign observation segment, are combined into an ultrafiltration oxygen change group.

[0074] When the cumulative ultrafiltration change in the ultrafiltration oxygen change group reaches the preset ultrafiltration condition corresponding to the cumulative ultrafiltration change, the ultrafiltration rate reaches the preset ultrafiltration condition corresponding to the ultrafiltration rate, and the lower limb descent persists, the patient is determined to have an ultrafiltration risk.

[0075] The preset ultrafiltration conditions include the preset ultrafiltration conditions corresponding to the cumulative ultrafiltration change and the preset ultrafiltration conditions corresponding to the ultrafiltration rate.

[0076] In one embodiment, tissue oxygen saturation, cumulative ultrafiltration volume, and ultrafiltration rate are all accompanied by a collection time identifier. The observation segment formed by tissue oxygen saturation is used as a time alignment window to group the cumulative ultrafiltration changes and ultrafiltration rates that are collected within the same observation segment.

[0077] When there are multiple cumulative ultrafiltration change records within the same vital sign observation segment, the median value of the multiple cumulative ultrafiltration change records is taken as the cumulative ultrafiltration change of the corresponding vital sign observation segment. When there are multiple ultrafiltration rate records within the same vital sign observation segment, the median value of the multiple ultrafiltration rate records is taken as the ultrafiltration rate of the corresponding vital sign observation segment.

[0078] The preset ultrafiltration condition corresponding to the cumulative ultrafiltration change is 110% of the ultrafiltration volume of the current effective medical order within the observation period of vital signs. The preset ultrafiltration condition corresponding to the ultrafiltration rate is 110% of the ultrafiltration rate of the current effective medical order.

[0079] In another embodiment, when tissue oxygen saturation monitoring is not configured during hemodialysis monitoring, observations of staring, slow reaction, yawning, changes in facial expression, cold sweats, body surface temperature, chest tightness, heart rate, forced sitting up, and muscle status are continuously collected. The occurrence time and duration of each patient's signs are recorded. According to the preset observation time, each patient's signs are divided into consecutive sign observation segments based on their occurrence time. Patient signs within the same sign observation segment are combined according to their occurrence time to form a multimodal sign group that does not include changes in tissue oxygenation.

[0080] From the multimodal sign group, gaze, slow reaction, and yawning were selected. The occurrence time of gaze, slow reaction, and yawning was recorded respectively. When the time interval between the occurrence times of at least two of the signs of gaze, slow reaction, and yawning was within the preset time correlation range, and at least one patient sign was present continuously within the corresponding sign observation period, gaze, slow reaction, and yawning were combined into a neurological state change combination, and a brain tissue insufficiency risk warning was generated based on the neurological state change combination.

[0081] When only one of the following—gazing, slowed response, and yawning—occurs within the same sign observation segment, the actual patient sign is considered as a single neurological state change, without directly indicating a risk of insufficient brain perfusion. Further examination of new and persistent states of gaze, slowed response, and yawning is conducted in subsequent sign observation segments.

[0082] Changes in facial features, cold sweats, and a sudden drop in body surface temperature were selected from the multimodal sign group. The occurrence time of these three changes was recorded. When at least two of these changes occurred within the same sign observation period and at least one corresponding time interval was within a preset time correlation range, the changes in facial features, cold sweats, and a sudden drop in body surface temperature were combined to form a peripheral state change combination. Based on this peripheral state change combination, a risk warning for insufficient peripheral tissue perfusion was generated.

[0083] When only one of the following occurs within the same observation period: facial changes, cold sweats, or a sudden drop in body surface temperature, the actual patient sign is considered as a single peripheral status change and does not directly indicate a risk of insufficient peripheral tissue perfusion. In subsequent observation periods, new and persistent changes in facial expressions, cold sweats, and a sudden drop in body surface temperature are further examined.

[0084] The muscle status of the left and right lower limbs was examined in the multimodal signs group. When the left lower limb muscle status showed abnormal contraction and reached the preset duration condition, a risk indication of left lower limb muscle spasm was generated. When the right lower limb muscle status showed abnormal contraction and reached the preset duration condition, a risk indication of right lower limb muscle spasm was generated. When the abnormal contraction did not reach the preset duration condition, the abnormal contraction was recorded as a change in muscle status and subsequent observation continued.

[0085] The cumulative ultrafiltration volume and ultrafiltration rate are extracted from the vital signs observation segment. The cumulative ultrafiltration change is calculated based on the cumulative ultrafiltration volume at adjacent collection times. The cumulative ultrafiltration change and ultrafiltration rate are correlated with changes in facial expression, cold sweat, sudden drop in body surface temperature, and muscle state within the same vital signs observation segment. When both the cumulative ultrafiltration change and ultrafiltration rate reach the preset ultrafiltration conditions, and at least one of the changes in facial expression, cold sweat, sudden drop in body surface temperature, and muscle state persists, an ultrafiltration risk warning is generated.

[0086] For patients with chest tightness, chest tightness is differentiated based on heart rate, 12-lead ECG information, serum potassium information, forced sitting up, and changes in facial expression during the occurrence of chest tightness. The process of chest tightness differentiation continues to use risk warnings and clinical auxiliary judgment methods, and disease diagnosis results are not formed based on a single patient's physical signs.

[0087] The risk set is formed by identifying the risk of insufficient brain perfusion, insufficient peripheral tissue perfusion, muscle spasm, ultrafiltration, and chest tightness according to the occurrence time. The warning level is then classified according to the risk type, combination of patient signs, duration of patient signs, and preset nursing rules. This allows for risk warnings to be provided even when tissue oxygen saturation monitoring is not available, by utilizing the temporal relationship of multimodal patient signs and nursing knowledge rules.

[0088] The preset time correlation range is 0 to 5 minutes. When the time interval between the occurrence of vital signs of different patients participating in the same risk assessment is less than or equal to 5 minutes, the corresponding patient vital signs are determined to be correlated in time.

[0089] The preset duration conditions are set according to different patient signs. The preset duration condition for staring is that the patient's gaze remains fixed for 5 seconds. The preset duration condition for yawning is that the patient yawns 3 times within 5 minutes. The preset duration condition for slow reaction is that the patient's response time to complete the preset interactive command is longer than 5 seconds compared to the individual's normal response state, or the patient is unable to complete the preset interactive command within 10 seconds but maintains spontaneous breathing and open eyes. The preset duration condition for sudden drop in body surface temperature is that the body surface temperature drops by more than 0.5°C within 5 minutes. The preset duration condition for forced sitting up is that the patient changes from a supine or semi-recumbent position to a sitting position and remains there for 30 seconds. The preset duration condition for muscle status is that abnormal muscle contraction of the left or right lower limb lasts for 3 seconds.

[0090] The preset ultrafiltration conditions include preset ultrafiltration conditions corresponding to cumulative ultrafiltration changes and preset ultrafiltration conditions corresponding to ultrafiltration rates. The preset ultrafiltration conditions corresponding to cumulative ultrafiltration changes are 110% of the ultrafiltration volume of the medical order within the observation period of vital signs specified in the current effective medical order, and the preset ultrafiltration conditions corresponding to ultrafiltration rates are 110% of the ultrafiltration rates specified in the current effective medical order.

[0091] The preset time-related range, preset duration conditions, and preset ultrafiltration conditions are used as initial judgment parameters in this embodiment to generate changes in patient status and risk warnings. The preset time-related range, preset duration conditions, and preset ultrafiltration conditions can be adjusted based on the experience of nursing experts, the patient's historical dialysis status, and subsequent clinical validation results. The specific parameters are used for risk warnings and decision support, and are not used as disease diagnostic criteria.

[0092] S3.2: The heart rate state during the occurrence of chest tightness is defined as the heart rate state of chest tightness, which includes the state of slowed heart rate and the state of accelerated heart rate.

[0093] When the chest tightness and heart rate status are in a bradycardia state, check the doctor's order authorization. If the doctor's order authorization is indicated, collect the 12-lead ECG information and blood potassium information. Compare the 12-lead ECG information and blood potassium information with the preset hyperkalemia conditions to obtain the hyperkalemia judgment result.

[0094] The preset hyperkalemia condition adopts the hyperkalemia identification conditions determined in advance before hemodialysis. The preset hyperkalemia condition includes the hyperkalemia comparison condition corresponding to blood potassium information and the hyperkalemia ECG comparison condition corresponding to twelve-lead ECG information. Based on the hyperkalemia judgment result, the hyperkalemia exclusion result is given.

[0095] In one embodiment, the hyperkalemia comparison condition corresponding to serum potassium information is a serum potassium concentration of 5.5 mmol / L, and the hyperkalemia ECG comparison condition corresponding to 12-lead ECG information is at least one of the following: the presence of tall peaked T waves, a PR interval exceeding 200 ms, or a QRS duration exceeding 120 ms.

[0096] When the serum potassium concentration reaches 5.5 mmol / L, the hyperkalemia assessment result is marked as an abnormal serum potassium state. When the 12-lead ECG information shows at least one of the following: tall peaked T waves, PR interval exceeding 200 ms, or QRS duration reaching 120 ms, the hyperkalemia assessment result is marked as an abnormal ECG state. When both abnormal serum potassium state and abnormal ECG state exist simultaneously, the hyperkalemia assessment result is marked as a hyperkalemia risk state.

[0097] When the result of the hyperkalemia assessment does not show any abnormal blood potassium or electrocardiogram (ECG) status, the result of the hyperkalemia exclusion is marked as excluded hyperkalemia. When the result of the hyperkalemia assessment shows either an abnormal blood potassium or an ECG status, the result of the hyperkalemia exclusion is marked as not excluded hyperkalemia.

[0098] When an abnormal blood potassium level exists but an abnormal electrocardiogram (ECG) level does not exist, the abnormal blood potassium level is recorded but the ECG is normal. When an abnormal ECG level exists but an abnormal blood potassium level does not exist, the abnormal ECG level is recorded but the abnormal blood potassium level is normal. Neither an abnormal blood potassium level nor an abnormal ECG level will result in a result excluding hyperkalemia.

[0099] When the result of ruling out hyperkalemia indicates that hyperkalemia has been ruled out, examine the changes in tissue oxygenation, facial appearance, and cold sweats after the formation of the hyperkalemia exclusion result from the physical examination. When the tissue oxygenation change is decreased, and at least one of pallor and cold sweats is in the same physical observation segment as chest tightness, the first chest tightness result is marked as highly suspected myocardial ischemia.

[0100] When chest tightness is accompanied by an increased heart rate, and when forced sitting up and chest tightness are observed in the same symptom observation segment, a heart failure diagnosis is obtained, and a heart failure exclusion result is given based on the heart failure diagnosis result.

[0101] When the heart failure exclusion result indicates that heart failure has been ruled out, check the doctor's authorization. If the doctor's authorization has been obtained, compare the ST segment changes in the 12-lead ECG information with the tissue oxygenation changes. If ST segment changes are present in the 12-lead ECG information and the tissue oxygenation changes are decreasing, mark the second chest tightness result as highly suspected myocardial ischemia.

[0102] The first and second chest tightness results were used as differential chest tightness results. The risks of insufficient cerebral perfusion, insufficient peripheral tissue perfusion, muscle spasm, ultrafiltration, and chest tightness differential results were recorded according to the time of occurrence to form a risk set.

[0103] S4: Classify the early warning level according to the risk set, match safety measures from the preset nursing rules according to the early warning level, nursing authority and medical order authority, form nursing matters and medical order matters, and record the actual execution results and execution time of the nursing matters to form a treatment record.

[0104] S4.1: The risk of insufficient brain perfusion, the risk of insufficient peripheral tissue perfusion, the risk of muscle spasm, the risk of ultrafiltration, and the results of chest tightness identification are respectively regarded as risk types.

[0105] When the risk profile indicates that the patient has a risk of muscle spasm or ultrafiltration, but no risk of insufficient brain perfusion or peripheral tissue perfusion, and the chest tightness is not marked as highly suspected myocardial ischemia, the warning level is determined to be moderate.

[0106] When the risk cluster indicates that the patient is at risk of insufficient cerebral perfusion or insufficient peripheral tissue perfusion, and the chest tightness differential diagnosis result is not marked as highly suspected myocardial ischemia, the warning level is determined to be high warning.

[0107] When chest tightness is identified as highly suspected of being a myocardial ischemia, the warning level is set to the highest level alert.

[0108] When multiple risk types exist simultaneously in a risk cluster, the warning levels corresponding to each risk type are compared in descending order of highest level alert, high level warning, and medium level warning. The highest level warning is taken as the current warning level, and the multiple risk types corresponding to the risk cluster are retained as inputs for subsequent safety measure matching and risk verification.

[0109] S4.2: Based on the current warning level and risk type, search for safety measures that correspond to both the current warning level and risk type in the preset nursing rules, and use the found safety measures as candidate measures.

[0110] In one embodiment, candidate measures corresponding to a moderate warning include reducing the current ultrafiltration rate to the ultrafiltration rate specified in the current effective medical order and strengthening monitoring; candidate measures corresponding to a high warning include suspending ultrafiltration, slowing blood flow, adjusting body position, emergency blood refill and disembarking, and strengthening monitoring; and candidate measures corresponding to the highest level alarm include immediately stopping the pump, adjusting body position, emergency blood refill and disembarking, and calling 120.

[0111] It should also be noted that calling 120 is only applicable to scenarios outside of hospitals or home dialysis. In hospital scenarios, the responsible medical staff will initiate the departmental or hospital emergency procedures as needed.

[0112] S4.3: Check the nursing authority for each candidate measure, compare the nursing execution requirements corresponding to the candidate measure with the nursing authority, and mark the candidate measure as a nursing matter when the nursing execution requirements corresponding to the candidate measure indicate that the nursing staff can execute it directly, and the nursing authority indicates that the nursing staff can execute it directly.

[0113] In one embodiment, the nursing execution requirements corresponding to pausing ultrafiltration, slowing blood flow, immediately stopping the pump, and adjusting the body position are to allow nursing staff to execute them first. The execution authorization status corresponding to pausing ultrafiltration, slowing blood flow, immediately stopping the pump, and adjusting the body position is recorded as allowing nursing staff to execute them directly. When candidate measures include pausing ultrafiltration, slowing blood flow, immediately stopping the pump, and adjusting the body position, the candidate measures that meet the nursing authority are marked as nursing matters.

[0114] S4.4: Check the candidate measures that are not marked as nursing matters according to the medical order execution requirements in the preset nursing rules. When the medical order execution requirement corresponding to the candidate measure indicates that it needs to be confirmed by a doctor before execution, check the medical order authority corresponding to the candidate measure.

[0115] When the doctor's authorization is indicated, the candidate measures that have been authorized by the doctor are marked as doctor's orders. When the doctor's authorization is indicated, the candidate measures that require the doctor's confirmation before execution are marked as doctor's orders, and the doctor's authorization corresponding to the doctor's orders is recorded as not having been authorized by the doctor.

[0116] In one embodiment, the execution requirement for enhanced monitoring is that it requires physician confirmation before execution. Enhanced monitoring includes 12-lead electrocardiogram monitoring and enhanced observation of tissue oxygen saturation. When enhanced monitoring is included as a candidate measure, the authorization of the medical order is determined based on the authorization record in the currently effective medical order, and enhanced monitoring is marked as a medical order item.

[0117] S4.5: Send nursing tasks to the corresponding nursing staff for execution according to the current warning level, and record the actual execution results and execution time for each nursing task.

[0118] For nursing tasks corresponding to pausing ultrafiltration, slowing blood flow, and immediately stopping the pump, when the hemodialysis machine completes the corresponding operation, the actual execution result is recorded as successful, and the time when the hemodialysis machine completes the corresponding operation is recorded as the execution time. When the hemodialysis machine fails to complete the corresponding operation, the actual execution result is recorded as failed, and the time when the corresponding operation fails is recorded as the execution time.

[0119] For nursing matters related to adjusting body position, the nursing staff shall complete the adjustment and submit an execution confirmation. The execution confirmation submitted by the nursing staff shall be regarded as the actual execution result, and the time corresponding to the execution confirmation shall be recorded as the execution time.

[0120] For nursing procedures related to emergency blood resuscitation and discharging, when the nursing staff initiates and completes the emergency blood resuscitation and discharging operation, the actual execution result is recorded as successful, and the time of completion of the blood resuscitation and discharging operation is recorded as the execution time. When the emergency blood resuscitation and discharging is not completed, the actual execution result is recorded as failed, and the time when the emergency blood resuscitation and discharging operation stops is recorded as the execution time.

[0121] The start time of the risk management period is the time when the risk concentration occurs, and the end time of the risk management period is the actual completion time of all nursing matters corresponding to the risk. Nursing matters and medical orders formed during the risk management period are classified into the corresponding risk types. Nursing matters are arranged in order of execution time, and the actual execution results and execution time of the nursing matters are recorded under the corresponding nursing matters to form a management record.

[0122] In one embodiment, when only sluggishness is present in the vital signs set, sluggishness is used as a sign of decreased level of consciousness, and the corresponding safety measures for moderate warning are matched, while outputting the prompt message "Decreased level of consciousness, please conduct bedside assessment immediately".

[0123] When the only physical sign is a gaze and the gaze lasts for 5 seconds, the gaze will be used as a high-risk warning entry point, and the corresponding safety measures for the high-risk warning will be matched. At the same time, the prompt message "The level of consciousness is declining. Please judge the risk of insufficient brain tissue perfusion in conjunction with changes in tissue oxygenation" will be output.

[0124] When staring and slowed reaction coexist within the same physical sign observation period, the change in consciousness is marked as a progressive decline, and the message "Progressive decline in consciousness level, suspected cerebral insufficiency or hypoglycemia" is output. At the same time, the risk of cerebral insufficiency is further judged by combining the condition that the prefrontal tissue oxygen decline lasts for more than 30 seconds and the decline reaches 15% of the prefrontal tissue oxygen line.

[0125] When a staring or slow reaction is accompanied by cold sweats, trembling hands, or hunger, the system outputs a message: "Suspected hypoglycemia, please measure your blood sugar immediately."

[0126] When a staring or slow reaction is accompanied by paleness, cold sweats, or a drop in body temperature of more than 0.5°C within 5 minutes, a prompt message "Suspected peripheral circulation perfusion abnormality" is output.

[0127] When staring is accompanied by a heart rate exceeding 100 beats per minute and chest tightness, the system will output the message "Suspected serious circulatory abnormality, please take immediate emergency measures."

[0128] It should also be noted that the risk set, warning level, nursing care, and medical orders are all used to provide risk warnings and clinical assistance during hemodialysis and do not constitute a disease diagnosis.

[0129] The risk cluster includes risks of insufficient brain perfusion, insufficient peripheral tissue perfusion, muscle spasm, ultrafiltration, and chest tightness. The differential diagnosis results indicate that the patient may have the corresponding abnormal condition, but do not mean that the patient has been diagnosed with the corresponding disease.

[0130] The output prompts are used to assist medical staff in assessing the patient's condition. The final diagnosis and treatment decision are determined by a licensed physician based on the patient's actual situation.

[0131] S5: Use the treatment record to determine the tissue oxygenation recovery status of the forehead, left lower limb, and right lower limb, as well as the exit status of corresponding vital signs, and at the same time verify the corresponding risks of the risk concentration to form a verification record.

[0132] S5.1: The end time of the risk management period is taken as the starting point of response observation. From the starting point of response observation, tissue oxygen saturation of the forehead, left lower limb and right lower limb is collected. Staring, yawning, changes in facial expression, cold sweat, body surface temperature, chest tightness, heart rate, forced sitting up and muscle status are recorded. The time of tissue oxygen saturation collection of the forehead, left lower limb and right lower limb and the time of occurrence of the patient's vital signs are also recorded.

[0133] When there are nursing matters that are not completed or fail to be completed during the risk management period, the time of the last actual execution record shall be taken as the end time of the risk management period, and the corresponding execution results shall continue to be recorded.

[0134] According to the preset response observation time, the oxygen saturation of the forehead, left lower limb, and right lower limb tissues, patient signs, cumulative ultrafiltration volume, and ultrafiltration rate are extracted after the start of the response observation. The data are then arranged from first to last according to the collection time and the occurrence time to form a response sequence.

[0135] In one embodiment, the preset response observation time is 5 minutes.

[0136] S5.2: The difference between the individual oxygen line in the frontal lobe and the oxygen saturation of the frontal tissue at each acquisition time in the response sequence is taken as the amount of frontal oxygen decrease, and the ratio of the amount of frontal oxygen decrease to the individual oxygen line in the frontal lobe is taken as the magnitude of frontal oxygen decrease during the response observation period.

[0137] When the forehead descent during the response observation period is lower than the preset descent condition and the preset recovery time is maintained continuously, the forehead tissue oxygen recovery status is marked as recovered; when the forehead descent is not lower than the preset descent condition, the forehead tissue oxygen recovery status is marked as not recovered.

[0138] The tissue oxygen saturation of the left and right lower limbs in the response sequence was divided into continuous observation segments according to the preset observation duration. The median of tissue oxygen saturation in each continuous observation segment was calculated, and the median tissue oxygen saturation of the left and right lower limbs was arranged in the time order of the continuous observation segments.

[0139] When the decrease in median oxygen saturation of the left lower limb tissue relative to the individual oxygen line of the left lower limb is less than the preset decrease condition, and the median oxygen saturation of the left lower limb tissue no longer meets the condition of continuous decrease in the lower limb oxygen sequence, the oxygen recovery status of the left lower limb tissue is marked as recovered. If the aforementioned conditions are not met simultaneously, the oxygen recovery status of the left lower limb tissue is marked as not recovered.

[0140] When the decrease in median oxygen saturation of the right lower limb tissue relative to the individual oxygen line of the right lower limb is less than the preset decrease condition, and the median oxygen saturation of the right lower limb tissue no longer meets the condition of continuous decrease in the lower limb oxygen sequence, the oxygen recovery status of the right lower limb tissue is marked as recovered. If the aforementioned conditions are not met simultaneously, the oxygen recovery status of the right lower limb tissue is marked as not recovered.

[0141] The oxygen recovery status of the forehead tissue, the left lower limb tissue, and the right lower limb tissue is considered as the tissue oxygen recovery status.

[0142] In one embodiment, the preset decrease condition is a 15% decrease in tissue oxygen saturation relative to the corresponding individual oxygen line, the preset recovery time is 1 minute, the preset observation time is 1 minute, and the condition for a continuous decrease in the lower limb oxygen sequence is that the median tissue oxygen saturation decreases sequentially across three consecutive observation segments.

[0143] S5.3: Based on the risk type in the risk cluster, examine the patient's vital signs corresponding to the risk type from the response sequence. The risk of insufficient brain perfusion corresponds to staring and yawning. The risk of insufficient peripheral tissue perfusion corresponds to changes in facial expression, cold sweats, and body surface temperature. The risk of muscle spasm corresponds to the muscle state of the lower limbs at risk of muscle spasm. The first chest tightness result corresponds to chest tightness, heart rate, changes in facial expression, and cold sweats. The second chest tightness result corresponds to chest tightness, heart rate, and forced sitting up.

[0144] When the gaze no longer meets the gaze condition and the yawning no longer meets the yawning condition during the response observation period, the corresponding signs of insufficient brain perfusion risk are marked as exited. When the gaze meets the gaze condition or the yawning meets the yawning condition during the response observation period, the corresponding signs of insufficient brain perfusion risk are marked as persistent.

[0145] When the facial color returns to normal, cold sweats cease, and the body surface temperature no longer meets the condition for a sudden drop in body surface temperature during the response observation period, the corresponding signs of peripheral tissue insufficiency risk are marked as exited. When the facial color remains pale, cold sweats continue to occur, or the body surface temperature meets the condition for a sudden drop in body surface temperature during the response observation period, the corresponding signs of peripheral tissue insufficiency risk are marked as persistent.

[0146] When the muscle status of the lower limb at risk of muscle spasm no longer meets the abnormal contraction condition during the response observation period, the corresponding sign of muscle spasm risk is marked as exited. When the muscle status of the lower limb at risk of muscle spasm meets the abnormal contraction condition during the response observation period, the corresponding sign of muscle spasm risk is marked as persistent.

[0147] When the chest tightness assessment result is the first chest tightness result, check the changes in chest tightness, heart rate, facial expression and cold sweat during the response observation period. When the chest tightness stops, the heart rate is within the preset normal heart rate range, the facial color returns to normal and the cold sweat stops, mark the corresponding signs of the first chest tightness result as exit.

[0148] When chest tightness persists, heart rate falls below the lower limit of the preset normal heart rate range, heart rate rises above the upper limit of the preset normal heart rate range, or the face remains pale or cold sweats continue to occur, the corresponding physical signs of the first chest tightness result are marked as persistent.

[0149] When the chest tightness assessment result is the second chest tightness result, check the chest tightness, heart rate, and forced sitting up during the response observation period. When the chest tightness stops, the heart rate is within the preset normal heart rate range, and forced sitting up stops, mark the corresponding signs of the second chest tightness result as exited. When the chest tightness continues to occur, the heart rate is lower than the lower limit of the preset normal heart rate range, the heart rate is higher than the upper limit of the preset normal heart rate range, or forced sitting up persists, mark the corresponding signs of the second chest tightness result as persistent.

[0150] The exit status of the signs corresponding to the risk of insufficient brain perfusion, the signs corresponding to the risk of insufficient peripheral tissue perfusion, the signs corresponding to the risk of muscle spasm, the signs corresponding to the first chest tightness result, and the signs corresponding to the second chest tightness result is used as the exit status of the signs.

[0151] In one embodiment, the gazing condition is that the patient's gaze is fixed for 5 seconds, the yawning condition is that the patient yawns 3 times within 5 minutes, the sudden drop in body surface temperature condition is that the body surface temperature drops by more than 0.5°C within 5 minutes, the abnormal contraction condition is that the muscles of the left or right lower limb contract abnormally for 3 seconds, and the preset normal heart rate range is 60 beats / minute to 100 beats / minute.

[0152] S5.4: Based on the risk type of the risk cluster, the tissue oxygenation recovery status and the exit of vital signs are checked against the identification results of insufficient brain tissue perfusion risk, insufficient peripheral tissue perfusion risk, muscle spasm risk, ultrafiltration risk and chest tightness, respectively, to form the risk response status of each risk type.

[0153] When the risk type is insufficient brain perfusion risk, check the prefrontal tissue oxygen recovery status and the exit status of the corresponding signs of insufficient brain perfusion risk. If the prefrontal tissue oxygen recovery status is recovered and the corresponding signs of insufficient brain perfusion risk are exited, mark the risk response status of insufficient brain perfusion risk as consistent response. If the prefrontal tissue oxygen recovery status is not recovered, or the corresponding signs of insufficient brain perfusion risk are persistent, mark the risk response status of insufficient brain perfusion risk as inconsistent response.

[0154] When the risk type is peripheral tissue hypoperfusion risk, the oxygen recovery status of the left or right lower limb tissue is selected based on the location of the lower limb descent when the peripheral tissue hypoperfusion risk occurs. If the tissue oxygen recovery status at the location of the lower limb descent is recovered and the corresponding signs of peripheral tissue hypoperfusion risk are withdrawn, the risk response status of peripheral tissue hypoperfusion risk is marked as consistent response. If the tissue oxygen recovery status at the location of the lower limb descent is not recovered, or the corresponding signs of peripheral tissue hypoperfusion risk are persistent, the risk response status of peripheral tissue hypoperfusion risk is marked as inconsistent response.

[0155] When the risk type is muscle spasm risk, the oxygen recovery status of the left lower limb or the right lower limb is selected based on the location where the muscle spasm risk occurs. When the oxygen recovery status of the tissue at the location where the muscle spasm risk occurs is recovered and the corresponding sign of the muscle spasm risk is withdrawn, the risk response status of the muscle spasm risk is marked as consistent. When the oxygen recovery status of the tissue at the location where the muscle spasm risk occurs is not recovered, or the corresponding sign of the muscle spasm risk is persistent, the risk response status of the muscle spasm risk is marked as inconsistent.

[0156] When the risk type is ultrafiltration risk, the tissue oxygen recovery status at the lower limb descent location and the ultrafiltration rate at the end of the response observation are selected from the response sequence. The ultrafiltration rate at the end of the response observation is compared with the ultrafiltration rate specified in the current effective medical order. When the tissue oxygen recovery status at the lower limb descent location is recovered and the ultrafiltration rate at the end of the response observation is not higher than the ultrafiltration rate specified in the current effective medical order, the risk response status of ultrafiltration risk is marked as consistent response.

[0157] If tissue oxygenation at the lower limb descent site is not restored, or if the ultrafiltration rate at the end of the response observation is higher than the ultrafiltration rate specified in the current effective medical order, the risk response status of the ultrafiltration risk will be marked as inconsistent response.

[0158] When the chest tightness assessment result is the first chest tightness result, select the tissue oxygen recovery status at the location where the tissue oxygen change occurred during the formation of the first chest tightness result and the exit status of the corresponding sign of the first chest tightness result. When the tissue oxygen recovery status is recovered and the corresponding sign of the first chest tightness result is exited, mark the risk response status of the first chest tightness result as consistent response. When the tissue oxygen recovery status is not recovered, or the corresponding sign of the first chest tightness result is persistent, mark the risk response status of the first chest tightness result as inconsistent response.

[0159] When the chest tightness assessment result is the second chest tightness result, select the tissue oxygen recovery status at the location where the tissue oxygen change occurred during the formation of the second chest tightness result and the exit status of the corresponding sign of the second chest tightness result. When the tissue oxygen recovery status is recovered and the corresponding sign of the second chest tightness result is exited, mark the risk response status of the second chest tightness result as consistent response. When the tissue oxygen recovery status is not recovered, or the corresponding sign of the second chest tightness result is persistent, mark the risk response status of the second chest tightness result as inconsistent response.

[0160] The risk response statuses of the risks of insufficient brain perfusion, insufficient peripheral tissue perfusion, muscle spasm, ultrafiltration, first chest tightness outcome, and second chest tightness outcome are arranged according to the occurrence time of the risk cluster to obtain the risk response status.

[0161] Search for the risk management period of the same risk type in the management records according to the time of occurrence of the risk in the risk concentration, write the risk response status into the management record corresponding to the same risk type, and record the start point of response observation, tissue oxygen recovery status, vital sign exit status and risk response status.

[0162] The risk type, risk management period, response observation start point, tissue oxygen recovery status, vital sign exit status, and risk response status are arranged in chronological order according to the time of risk occurrence to form a verification record.

[0163] It should also be noted that the specific values ​​involved in this embodiment are all initial preset parameters formed based on the experience of hemodialysis nursing experts, the patient's historical dialysis status, clinical treatment standards, and verification data. The correspondence between risk type and warning level, the correspondence between warning level and safety measures, the correspondence between patient vital sign combinations and risk warnings, and other nursing knowledge mapping relationships are all initial mapping rules formed based on the experience of nursing experts. The initial preset parameters and initial mapping rules can be adjusted or updated according to the patient's historical dialysis status, medical institution nursing standards, and subsequent clinical research and verification results. The adjusted or updated initial preset parameters and initial mapping rules continue to be used for the judgment and warning processing of the relationship between corresponding patient vital signs, tissue oxygenation changes, dialysis process parameters, and risk status.

[0164] The risk of insufficient brain perfusion, insufficient peripheral tissue perfusion, muscle spasm, ultrafiltration, chest tightness identification results, warning levels, and corresponding prompts in this embodiment are all used for risk warnings and clinical decision support during hemodialysis. They are not used as independent bases for disease diagnosis, disease severity determination, or treatment effect evaluation. The final diagnosis and treatment decision shall be determined by medical personnel with corresponding professional qualifications in combination with the patient's actual situation.

[0165] This embodiment can generate preset nursing rule optimization suggestions by combining background data analysis, and after review by nursing experts, the preset nursing rules can be manually updated by authorized personnel.

[0166] This embodiment also provides a hemodialysis complication early warning system based on multimodal signs, including: The baseline generation module continuously collects patient vital signs, tissue oxygen saturation, cumulative ultrafiltration volume, ultrafiltration rate, nursing permissions, and medical order permissions, and establishes an individual oxygen baseline based on the tissue oxygen saturation at the beginning of hemodialysis. The vital signs association module compares individual oxygen lines and tissue oxygen saturation to obtain tissue oxygen changes, and associates patient vital signs with tissue oxygen changes according to the time of occurrence to form a vital signs set. The risk assessment module determines the risk of insufficient brain perfusion, insufficient peripheral tissue perfusion, and muscle spasm based on the occurrence time of physical signs and tissue oxygenation changes. It also determines the risk of ultrafiltration based on the changes in lower limb tissue oxygenation, cumulative ultrafiltration volume, and ultrafiltration rate in the physical signs, and identifies chest tightness based on the physical signs to form a risk set. The tiered treatment module classifies warning levels based on risk sets, matches safety measures from preset nursing rules based on warning level, nursing authority, and medical order authority, generates nursing and medical order items, and records the actual execution results and execution time of the nursing items to form a treatment record; The response verification module uses treatment records to determine the tissue oxygenation recovery status of the forehead, left lower limb, and right lower limb, as well as the exit status of corresponding vital signs. At the same time, it verifies the corresponding risks in the risk cluster and generates verification records.

[0167] In summary, this invention, by continuously observing the changes in tissue oxygenation and the occurrence and duration of patient signs, establishes a continuous temporal correlation between forehead drooping, lower limb drooping, and corresponding clinical signs. This provides a sequential basis for determining the risks of insufficient brain perfusion, insufficient peripheral tissue perfusion, muscle spasm, and ultrafiltration, ensuring that early tissue oxygenation changes and subsequent clinical manifestations correspond within the same temporal information chain. Furthermore, by continuously verifying tissue oxygenation recovery, sign withdrawal, and ultrafiltration status, the risk formation, treatment execution, and post-treatment response are sequentially linked, and a risk response status is established based on the tissue oxygenation recovery and sign withdrawal status corresponding to different risks, thus enabling post-treatment verification of the risk assessment results.

[0168] 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 early warning of hemodialysis complications based on multimodal signs, characterized in that: include, Continuously collect patient vital signs, tissue oxygen saturation, cumulative ultrafiltration volume, ultrafiltration rate, nursing authority and medical order authority, and establish individual oxygen saturation lines based on tissue oxygen saturation at the beginning of hemodialysis; Individual oxygen line and tissue oxygen saturation are compared to obtain tissue oxygen changes. Patient signs are correlated with tissue oxygen changes according to the time of occurrence to form a sign set. The risk of insufficient brain perfusion, insufficient peripheral tissue perfusion, and muscle spasm is determined based on the timing of the occurrence of physical signs and tissue oxygenation changes. The risk of ultrafiltration is determined based on the changes in lower limb tissue oxygenation, cumulative ultrafiltration volume, and ultrafiltration rate in the physical signs. Chest tightness is also identified based on the physical signs, thus forming a risk set. The warning level is divided according to the risk set. Based on the warning level, nursing authority and medical order authority, safety measures are matched from the preset nursing rules to form nursing matters and medical order matters. The actual execution results and execution time of the nursing matters are recorded to form a treatment record. The treatment records are used to assess the tissue oxygenation recovery in the forehead, left lower limb, and right lower limb, as well as the exit of corresponding vital signs. At the same time, the corresponding risks of the risk concentration are verified, and verification records are generated.

2. The method for early warning of hemodialysis complications based on multimodal signs as described in claim 1, characterized in that: The establishment of individual oxygen lines, include, The patient's physical signs include staring, slow reaction, yawning, changes in facial expression, cold sweat, body surface temperature, chest tightness, heart rate, forced sitting up, and muscle condition. Tissue oxygen saturation of the forehead, left lower limb, and right lower limb at the beginning of hemodialysis was collected to form three initial oxygen sequences; After removing the tissue oxygen saturation corresponding to sensor detachment, contact abnormalities, and invalid signals from the three initial oxygen sequences, three effective oxygen sequences were obtained. The median of the three effective oxygen sequences was then used as the individual oxygen line for the frontal limb, left lower limb, and right lower limb.

3. The method for early warning of hemodialysis complications based on multimodal signs as described in claim 1, characterized in that: The obtained tissue oxygenation changes include, The difference between the prefrontal tissue oxygen saturation and the individual prefrontal oxygen line is compared to obtain the prefrontal oxygen change. When the prefrontal oxygen change reaches the preset decrease condition and continues for the preset duration, the prefrontal oxygen change is marked as a prefrontal decrease. The tissue oxygen saturation of the left and right lower limbs was divided into continuous observation segments, and the median tissue oxygen saturation of each continuous observation segment was calculated to obtain the lower limb oxygen sequence. When the oxygen sequence of the ipsilateral lower limb continues to decrease and reaches the preset decrease condition, the corresponding lower limb oxygen change is marked as lower limb decrease, and the decrease in forehead and lower limb oxygen is regarded as tissue oxygen change.

4. The method for early warning of hemodialysis complications based on multimodal signs as described in claim 1, characterized in that: The method of correlating patient vital signs with changes in tissue oxygenation by time of occurrence includes, Record changes in tissue oxygenation and the occurrence time of patient signs to form a sign time series. Divide the sign time series into multiple sign observation segments according to the preset observation duration. Changes in tissue oxygenation occurring within the same observation segment, along with staring, slowed reflexes, yawning, changes in facial expression, cold sweats, body surface temperature, chest tightness, heart rate, forced sitting up, and muscle condition, are grouped into observation segment sign groups and arranged in chronological order to form sign sets.

5. The method for early warning of hemodialysis complications based on multimodal signs as described in claim 1, characterized in that: The assessment of the risk of insufficient brain perfusion, insufficient peripheral tissue perfusion, and muscle spasm includes, From the set of vital signs, forehead descent and gaze and yawning within the same observation segment were selected, and the first time interval between the occurrence of forehead descent and gaze and the second time interval between the occurrence of forehead descent and yawning were calculated respectively. When at least one of the first and second time intervals is within a preset time correlation range, frontal descent persists, and at least one of the following—gazing, slowed reaction, and yawning—occurs within the observation period of signs where frontal descent occurs, the patient is deemed to be at risk of insufficient brain perfusion. The third time interval between the time of lower limb descent and the time of facial changes, the fourth time interval between the time of lower limb descent and the time of cold sweating, and the fifth time interval between the time of lower limb descent and the time of sudden drop in body surface temperature were calculated separately. When at least one of the third, fourth, and fifth time intervals is within a preset time correlation range, the lower limb descent persists, and at least two of the following—facial changes, cold sweats, and a sudden drop in body surface temperature—occur within the observation period of the signs of lower limb descent, the patient is deemed to be at risk of peripheral tissue insufficiency. If the descent of the lower limb persists and abnormal contraction of the muscles on the same side occurs, it is determined that there is a risk of muscle spasm in the corresponding lower limb.

6. The method for early warning of hemodialysis complications based on multimodal signs as described in claim 1, characterized in that: The assessment of ultrafiltration risk include, The cumulative ultrafiltration volume and ultrafiltration rate were extracted according to the physical signs observed in the lower limbs to obtain the ultrafiltration observation group; The cumulative ultrafiltration change is calculated based on the cumulative ultrafiltration amount at adjacent time points in the ultrafiltration observation group, and the cumulative ultrafiltration change and ultrafiltration rate are arranged according to the corresponding time points to obtain the ultrafiltration change sequence; By comparing the recording time of the ultrafiltration change sequence with the occurrence time of lower limb descent, the cumulative ultrafiltration change, ultrafiltration rate, and lower limb descent status falling within the same sign observation segment are merged into an ultrafiltration oxygen change group. When the cumulative ultrafiltration change and ultrafiltration rate in the ultrafiltration oxygen change group both reach the preset ultrafiltration conditions and the lower limb descent persists, the patient is determined to have ultrafiltration risk.

7. The method for early warning of hemodialysis complications based on multimodal signs as described in claim 1, characterized in that: The process of identifying chest tightness includes... The heart rate state during the occurrence of chest tightness is divided into a slow heart rate state and a fast heart rate state, which is the chest tightness heart rate state. When the chest tightness and heart rate state is a bradycardia state, obtain the 12-lead ECG information and blood potassium information, compare the 12-lead ECG information and blood potassium information with the preset hyperkalemia conditions to obtain the hyperkalemia judgment result, confirm the hyperkalemia exclusion result based on the doctor's diagnosis information, and correlate the hyperkalemia exclusion result with tissue oxygenation changes, facial appearance changes and cold sweats over time to obtain the first chest tightness result. When the chest tightness and heart rate state is an tachycardia state, correlate the forced sitting up with the chest tightness over time to obtain the heart failure judgment result. Based on the doctor's diagnosis information, the heart failure exclusion result is confirmed. The heart failure exclusion result is correlated with the ST segment changes and tissue oxygenation changes in the twelve-lead electrocardiogram information over time to obtain the second chest tightness result. The first chest tightness result and the second chest tightness result are used as the chest tightness differential result. Together with the risks of insufficient brain tissue perfusion, insufficient peripheral tissue perfusion, muscle spasm, and ultrafiltration, a risk set is formed.

8. The method for early warning of hemodialysis complications based on multimodal signs as described in claim 7, characterized in that: The matching security measures include, According to the warning level, the corresponding safety measures are selected from the preset nursing rules to obtain candidate measures. The nursing authority is used to verify the authority of the candidate measures. The candidate measures that can be executed by the nursing authority are marked as nursing matters. The doctor's order authority is used to verify the authority of the candidate measures that are not marked as nursing matters. The candidate measures that require doctor's confirmation are marked as doctor's order matters. Record the actual results and execution time of nursing matters, and combine the actual results, execution time, nursing matters and medical orders into a treatment record.

9. The method for early warning of hemodialysis complications based on multimodal signs as described in claim 1, characterized in that: The corresponding risks in the verification risk set include, The execution time in the treatment record is used as the starting point for response observation. The tissue oxygen saturation, corresponding signs, cumulative ultrafiltration volume and ultrafiltration rate of the forehead, left lower limb and right lower limb after the starting point of response observation are used to form a response sequence. The tissue oxygen saturation of the forehead, left lower limb and right lower limb in the response sequence is compared with the individual oxygen line of the forehead, the individual oxygen line of the left lower limb and the individual oxygen line of the right lower limb, respectively, to obtain the tissue oxygen recovery status. According to the response sequence, check the persistence status of the corresponding signs, mark the corresponding signs that stop appearing as exit signs, and obtain the sign exit status. Then, check the tissue oxygenation recovery status and the sign exit status with the corresponding risks in the risk set to obtain the risk response status. When tissue oxygen saturation recovers to the corresponding individual oxygen line and the corresponding signs disappear, the risk response status is marked as consistent response; when tissue oxygen saturation continues to decrease and the corresponding signs persist, the risk response status is marked as inconsistent response. Corresponding the risk response status with the handling records over time creates a verification record.

10. A hemodialysis complication early warning system based on multimodal signs, based on the hemodialysis complication early warning method based on multimodal signs according to any one of claims 1 to 9, characterized in that: include, The baseline generation module continuously collects patient vital signs, tissue oxygen saturation, cumulative ultrafiltration volume, ultrafiltration rate, nursing permissions, and medical order permissions, and establishes an individual oxygen baseline based on the tissue oxygen saturation at the beginning of hemodialysis. The vital signs association module compares individual oxygen lines and tissue oxygen saturation to obtain tissue oxygen changes, and associates patient vital signs with tissue oxygen changes according to the time of occurrence to form a vital signs set. The risk assessment module determines the risk of insufficient brain perfusion, insufficient peripheral tissue perfusion, and muscle spasm based on the occurrence time of physical signs and tissue oxygenation changes. It also determines the risk of ultrafiltration based on the changes in lower limb tissue oxygenation, cumulative ultrafiltration volume, and ultrafiltration rate in the physical signs, and identifies chest tightness based on the physical signs to form a risk set. The tiered treatment module classifies warning levels based on risk sets, matches safety measures from preset nursing rules based on warning level, nursing authority, and medical order authority, generates nursing and medical order items, and records the actual execution results and execution time of the nursing items to form a treatment record; The response verification module uses treatment records to determine the tissue oxygenation recovery status of the forehead, left lower limb, and right lower limb, as well as the exit status of corresponding vital signs. At the same time, it verifies the corresponding risks in the risk cluster and generates verification records.

Images