Sleep monitoring system and method for neurology department

Abstract

The invention discloses a sleep monitoring system and method for the neurology department, relates to the technical field of sleep monitoring, and solves the problems that when changes of a whole waveform in different sleep sections are not comprehensively considered, some waveform fluctuations and sleep fluctuations are caused, and sleep fault misjudgment is caused. According to the method, waveform monitoring is carried out through neurology sleep monitoring personnel, the corresponding brain wave waveform is confirmed, wave band division is carried out on the corresponding brain wave waveform based on the set preset feature interval to lock the related feature segments of the same kind, feature analysis is carried out on the divided feature segments of the same kind, and the accuracy of the brain wave waveform is improved. According to the method, the electric wave waveform can be effectively subjected to correlation division and feature segment fusion processing, so that a better wave band fusion processing effect is achieved, a plurality of different wave bands can be effectively combined to generate a complete corresponding wave band, and a better electric wave waveform processing effect can be achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of sleep monitoring, and in particular to a neurology sleep monitoring system and method. Background Art

[0002] Neurological sleep monitoring is an important examination method used to evaluate sleep quality and diagnose sleep disorders. By monitoring various physiological parameters during sleep, such as brain waves, eye movements, muscle activity, breathing, heart rate, etc., it can understand the structure and depth of sleep and determine whether sleep is sufficient and effective.

[0003] The application with publication number CN117100222A discloses a neurological sleep monitoring method, which obtains first-category sensor data, which is behavioral data of the subject, reflecting the sleep process; second-category sensor data is used to judge the changes in facial expression and morphology of the subject during sleep; third-category sensor data is obtained to judge the medical data of the subject; and the sleep process of the subject is evaluated based on the fuzzy reasoning method based on expert experience by combining the first-category sensor data, the second-category sensor data, and the third-category sensor data, to judge problems that occur during the sleep process of the subject.

[0004] During the monitoring process of neurology-related sleep parameters, the abnormal sleep conditions of the corresponding personnel are generally identified based on the specific changes in the monitoring waveform. However, the original analysis method does not take into account the changes in the entire waveform in different sleep segments, which will lead to certain waveform fluctuations and sleep fluctuations and cause misjudgment of sleep failures. The original monitoring and analysis method is not comprehensive and cannot achieve better sleep monitoring and analysis results. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a neurological sleep monitoring system and method, which solves the problem that failure to take into account the changes in the entire waveform in different sleep segments will lead to certain waveform fluctuations and sleep fluctuations, resulting in misjudgment of sleep failure.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: A neurological sleep monitoring method, comprising the following steps:

[0007] Step 1: Perform brain wave monitoring on the sleep monitors participating in the neurology department, and generate brain wave waveforms belonging to the sleep monitors based on the monitoring data of the sleep monitors throughout the sleep process. Based on the preset waveform standard features, the brain wave waveforms are divided into multiple similar feature segments. The specific method is as follows:

[0008] S11. Based on the confirmed brain wave waveform, a plurality of peak points existing in the brain wave waveform are confirmed in sequence, wherein the trend of the waveform segments before and after the peak point is opposite, and the trend of the waveform segment before the peak point is upward, while the trend of the waveform segment after the peak point is downward;

[0009] S12, determine the waveform segments of adjacent peak points, and calibrate the wave speed of the corresponding waveform segment generation process as V i , and mark the vertical distance between adjacent peak points as L i , where i represents the waveform segment between different adjacent peak points, using F i =V i ÷L i Confirm the band frequency F of this waveform segment i ;

[0010] S13, according to the preset waveform standard characteristics, the band frequency F i The waveform standard feature is marked as a characteristic waveform segment, and the characteristic waveform segments that appear continuously and belong to the same waveform standard feature are marked as similar characteristic segments, and several waveform segments generated by the brain wave waveform are analyzed, and similar characteristic segments belonging to different waveform standard features are confirmed in turn, and their different waveform standard features are preset interval features, which correspond to different sleep stages respectively;

[0011] Step 2: Based on the multiple similar feature segments confirmed in succession, lock the feature segment set column, and then perform feature analysis on the similar feature segments arranged in sequence in the feature segment set column, evaluate whether the corresponding similar feature segments can be merged into new similar feature segments, and re-calibrate them into the feature segment set column. The evaluation processing method is:

[0012] S21. Identify whether there are similar feature segments associated with the same sleep stage in the feature segment set:

[0013] If they exist, other similar feature segments contained between two groups of similar feature segments or multiple groups of similar feature segments are marked as subsidiary feature segments, and the total length of the line segments of the subsidiary feature segments is confirmed and marked as D1, and then the total length of the line segments between two groups of similar feature segments or multiple groups of similar feature segments is marked as D2, and the total proportion ZB of its subsidiary feature segments is determined by D1÷D2=ZB. If ZB>Y1, where Y1 is a preset value, no fusion processing is performed. If ZB≤Y1, other similar feature segments contained between two groups of similar feature segments or multiple groups of similar feature segments are marked as abnormal bands of this sleep stage. This sleep stage is the sleep stage corresponding to the two groups of similar feature segments or multiple groups of similar feature segments. After the abnormal bands are marked, the two groups of similar feature segments or multiple groups of similar feature segments are fused into new similar feature segments;

[0014] If there is no similar feature segment associated with the same sleep stage in the feature segment set, no processing is performed;

[0015] Step 3: Based on the processed feature segment set sequence, confirm the sleep stages corresponding to the feature segment set sequence from front to back, and check the sleep stage sequence confirmed in sequence with the set standard sleep segment to assess whether the two are consistent. If they are consistent, analyze and process the abnormal segment. If they are not consistent, directly generate abnormal signals for relevant display, where the standard sleep segment is a preset arrangement segment; the specific method is:

[0016] S31, based on the determined feature segment set sequence, confirm the sleep stages associated with the same feature segments in the feature segment set sequence from front to back, so as to lock the sleep stage sequence, compare the sleep stage sequence with the set standard sleep segment, and identify whether the two are completely consistent: if they are consistent, execute step S32 to analyze and process the abnormal segment, if not, directly generate an abnormal signal display;

[0017] S32, determining the abnormal band marked by the same characteristic segment in the sleep stage sequence, and confirming the minimum waveform frequency B of the waveform standard feature based on the waveform standard feature corresponding to the same characteristic segment k min and maximum waveform frequency B k max, where k represents different waveform standard characteristics, and then based on the band frequency F confirmed by the corresponding waveform segment i , determine the multiple F associated with this abnormal band i , and then from multiple F i Select the minimum value F i min and maximum value F i max, determine the frequency interval associated with the abnormal band corresponding to this similar characteristic segment;

[0018] S33, compare the difference between the waveform standard feature and the wave frequency interval: determine the associated characteristic value of this abnormal band: Tz = |B k max-F i max|+|B k min-F i min|Lock the associated characteristic value Tz, then confirm the associated characteristic values ​​of the abnormal bands associated with other similar characteristic segments, sum the confirmed groups of associated characteristic values ​​Tz, and determine the total value ZH;

[0019] S34. Compare the total value ZH with the preset value Y2, where Y2 is the preset value. If ZH>Y2, a violent fluctuation signal is generated and the generated violent fluctuation signal is displayed; if ZH≤Y2, a normal fluctuation signal is generated and the generated normal fluctuation signal is displayed.

[0020] Preferably, a neurology sleep monitoring system comprises:

[0021] The waveform segment division end divides the brain wave waveform generated during the brain wave monitoring process into a plurality of similar feature segments based on preset waveform standard features;

[0022] The feature segment processing end locks the feature segment set column based on the multiple similar feature segments confirmed in succession, and then performs feature analysis on the similar feature segments arranged in sequence in the feature segment set column, assesses whether the corresponding similar feature segments can be merged into new similar feature segments, and re-calibrates them into the feature segment set column;

[0023] The sleep stage checking end, based on the processed feature segment set sequence, confirms the sleep stages corresponding to the feature segment set sequence from front to back, checks the confirmed sleep stage sequence with the set standard sleep segment, and evaluates whether the two are consistent. If they are consistent, the abnormal segment is analyzed and processed and a violent fluctuation signal or a normal fluctuation signal is generated. If they are not consistent, an abnormal signal is directly generated for display;

[0024] The display end displays the generated violent fluctuation signal, normal fluctuation signal, or abnormal signal.

[0025] The present invention provides a neurology sleep monitoring system and method. Compared with the prior art, it has the following beneficial effects:

[0026] The present invention uses a neurology sleep monitoring personnel to perform waveform monitoring, confirm the corresponding brain wave waveform, and then divide the corresponding brain wave waveform into bands based on the preset characteristic interval to lock the related similar characteristic segments, and then perform feature analysis on the divided similar characteristic segments, which can effectively make the electric wave waveform be associated and divided, and perform fusion processing of the characteristic segments, so as to achieve a better band fusion processing effect, and can effectively combine multiple different bands to generate a complete corresponding band, which can achieve a better electric wave waveform processing effect;

[0027] Based on the processed electric wave waveform, the verification difference between it and the original preset sleep segment is identified, and then the correlation analysis of the abnormal bands is performed according to the verification result. Based on the correlation analysis result of the abnormal bands, it is confirmed whether the monitored sleep quality meets the standard, and the relevant signals are generated for display, which can achieve better electric wave waveform analysis effect, improve the accuracy of its waveform analysis, and achieve better signal confirmation effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a schematic diagram of the process of the present invention;

[0029] Figure 2 It is a schematic diagram of the principle framework of the present invention. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] First embodiment

[0032] See also Figure 1 , the present application provides a neurological sleep monitoring method, comprising the following steps:

[0033] Step 1: Perform brain wave monitoring on the sleep monitors participating in the neurology department, and generate a brain wave waveform belonging to the sleep monitor based on the monitoring data of the sleep monitor throughout the sleep process (the corresponding monitoring instrument will generate the corresponding waveform based on the real-time monitoring results), and divide the brain wave waveform into multiple similar feature segments based on the preset waveform standard features. Specifically, the waveform standard feature is the frequency interval associated with the corresponding waveform, and different sleep stages correspond to different waveform frequency intervals, which are prepared in advance by relevant operators based on experience;

[0034] The specific method of dividing the brain wave waveform into multiple feature segments is as follows:

[0035] S11. Based on the confirmed brain wave waveform, a number of peak points existing in the brain wave waveform are confirmed in sequence, and the trend of the waveform segments before and after the peak point is opposite, and the trend of the waveform segment before the peak point is upward, and the trend of the waveform segment after the peak point is downward. Specifically, there is a corresponding peak point in the waveform segment, and the waveform intensity corresponding to the line segment before the peak point gradually increases with time. When this waveform segment reaches the peak point, the waveform intensity corresponding to the line segment at the rear end gradually decreases with time. Therefore, the point between the waveform segments on both sides belongs to the corresponding peak point;

[0036] S12, determine the waveform segments of adjacent peak points, and calibrate the wave speed of the corresponding waveform segment generation process as V i , and mark the vertical distance between adjacent peak points as L i , where i represents the waveform segment between different adjacent peak points, using F i =V i ÷L i Confirm the band frequency F of this waveform segment i ;

[0037] S13, according to the preset waveform standard characteristics, the band frequency Fi The waveform standard feature is marked as a characteristic waveform segment, and the characteristic waveform segments that appear continuously and belong to the same waveform standard feature are marked as similar characteristic segments, and several waveform segments generated by the brain wave waveform are analyzed, and similar characteristic segments belonging to different waveform standard features are confirmed in turn, and its different waveform standard features are preset interval features, which correspond to different sleep stages respectively. Specifically, its waveform standard features are preset characteristic intervals, which are prepared in advance by relevant operators based on experience, obtained from different sleep stages, and confirmed based on the practical experience of relevant operators. Its different sleep stages correspond to different preset characteristic intervals, and different preset characteristic intervals can divide different waveform segments. According to the continuous state of the brain wave waveform, the waveform segments can be confirmed in succession, so as to lock similar characteristic segments in turn;

[0038] Step 2: Based on the multiple similar feature segments confirmed in succession, lock the feature segment set column, and then perform feature analysis on the similar feature segments arranged in sequence in the feature segment set column, evaluate whether the corresponding similar feature segments can be merged into new similar feature segments, and re-calibrate them into the feature segment set column. Specifically, in the process of wave frequency confirmation, there will be some changes in waveform intensity, resulting in the waveform belonging to the current sleep stage being divided into other sleep stages. In this case, the waveform of this stage belongs to the sleep characteristics of the previous and next waveform segments, which is only caused by some abnormal fluctuations. Therefore, relevant fusion can be performed to determine new similar feature segments, so as to achieve a better collection effect.

[0039] The evaluation and processing method of the feature segment set column is as follows:

[0040] S21. Identify whether there are similar feature segments associated with the same sleep stage in the feature segment set:

[0041] If it exists, the other similar feature segments contained between two groups of similar feature segments or multiple groups of similar feature segments are marked as subsidiary feature segments, and the total length of the line segments of the subsidiary feature segments is confirmed and marked as D1, and then the total length of the line segments between two groups of similar feature segments or multiple groups of similar feature segments is marked as D2, and D1÷D2=ZB is used to determine the total proportion ZB of its subsidiary feature segments (its feature segment is the corresponding waveform segment, and the line length of its waveform segment can be directly determined, so that the length of the line segment associated with it can be directly locked). If ZB>Y1, where Y1 is the predicted Set a value, the specific value is determined by the operator based on experience, and no fusion processing is performed. If ZB≤Y1, the other similar feature segments contained between the two groups of similar feature segments or multiple groups of similar feature segments are marked as abnormal bands of this sleep stage. This sleep stage is the sleep stage corresponding to the two groups of similar feature segments or multiple groups of similar feature segments. After the abnormal band is calibrated, the two groups of similar feature segments or multiple groups of similar feature segments are fused into a new similar feature segment (that is, multiple groups of similar feature segments are fused into one group of similar feature segments to facilitate subsequent specific verification);

[0042] If it does not exist, no processing is performed.

[0043] Specifically, here we analyze the relevant characteristics of the waveform to fuse the feature segments, so as to achieve a better band fusion processing effect, which can effectively combine multiple different bands to generate a complete corresponding band, and its complete corresponding band is associated with the corresponding sleep stage.

[0044] Second embodiment

[0045] Based on the feature segment set sequence after the feature processing, the feature segment set sequence is reprocessed to complete the specific abnormality confirmation process.

[0046] Step 3: Based on the processed feature segment set column, confirm the sleep stages corresponding to the feature segment set column from front to back, check the confirmed sleep stage sequence with the set standard sleep segment, and evaluate whether the two are consistent. If they are consistent, analyze and process the abnormal segment. If they are not consistent, directly generate abnormal signals for relevant display. The standard sleep segment is a preset arrangement segment. Specifically, its feature segment set column is the relevant set column after the corresponding feature segment fusion processing. Each different feature segment corresponds to a different sleep stage. The sleep stages under normal circumstances are: falling asleep, light sleep, deep sleep and deep sleep. Each different sleep stage corresponds to a different frequency range, so the arranged standard sleep segment is: falling asleep-light sleep-deep sleep-deep sleep. The specific way to evaluate it is:

[0047] S31, based on the determined feature segment set column, confirm the sleep stages associated with the same feature segments in the feature segment set column from front to back, so as to lock the sleep stage sequence, compare the sleep stage sequence with the set standard sleep segment, and identify whether the two are completely consistent: if they are consistent, execute step S32 to analyze and process the abnormal segment; if they are not consistent, directly generate an abnormal signal display for external personnel to view;

[0048] S32, determining the abnormal band marked by the same characteristic segment in the sleep stage sequence, and confirming the minimum waveform frequency B of the waveform standard feature based on the waveform standard feature corresponding to the same characteristic segment. k min and maximum waveform frequency B k max, where k represents different waveform standard characteristics, and then based on the band frequency F confirmed by the corresponding waveform segment i , determine the multiple F associated with this abnormal band i , and then from multiple F i Select the minimum value F i min and maximum value F i max, determine the frequency interval associated with the abnormal band corresponding to this similar characteristic segment;

[0049] S33, compare the difference between the waveform standard feature and the wave frequency interval: determine the associated characteristic value of this abnormal band: Tz = |B k max-F i max|+|B k min-F i min|Lock the associated characteristic value Tz, then confirm the associated characteristic values ​​of the abnormal bands associated with other similar characteristic segments, sum the confirmed groups of associated characteristic values ​​Tz, and determine the total value ZH;

[0050] S34, the sum value ZH is checked against a preset value Y2, wherein Y2 is a preset value, and its specific value is determined by the operator based on experience. If ZH>Y2, a violent fluctuation signal is generated, and the generated violent fluctuation signal is displayed; if ZH≤Y2, a normal fluctuation signal is generated, and the generated normal fluctuation signal is displayed;

[0051] Specifically, the generation of violent fluctuation signals or normal fluctuation signals depends on the changes in the fluctuation segment. When the generated fluctuation segment has a large fluctuation but a short fluctuation length, there will be a merger of the same characteristic segments. For such cases, it is necessary to conduct relevant analysis of the fluctuation characteristics to assess whether the fluctuation characteristics meet the standards, so as to make relevant judgments on whether the fluctuations are normal or abnormal, so as to achieve better monitoring and analysis results. External operators can make specific execution judgments based on the specific signals displayed to achieve better judgment accuracy.

[0052] Third embodiment

[0053] A neurology sleep monitoring system, comprising:

[0054] The waveform segment division end divides the brain wave waveform generated during the brain wave monitoring process into a plurality of similar feature segments based on preset waveform standard features;

[0055] The feature segment processing end locks the feature segment set column based on the multiple similar feature segments confirmed in succession, and then performs feature analysis on the similar feature segments arranged in sequence in the feature segment set column, assesses whether the corresponding similar feature segments can be merged into new similar feature segments, and re-calibrates them into the feature segment set column;

[0056] The sleep stage checking end, based on the processed feature segment set sequence, confirms the sleep stages corresponding to the feature segment set sequence from front to back, checks the confirmed sleep stage sequence with the set standard sleep segment, and evaluates whether the two are consistent. If they are consistent, the abnormal segment is analyzed and processed and a violent fluctuation signal or a normal fluctuation signal is generated. If they are not consistent, an abnormal signal is directly generated for display;

[0057] The display end displays the generated violent fluctuation signal, normal fluctuation signal, or abnormal signal.

[0058] Fourth embodiment

[0059] The specific implementation process of this embodiment includes all the implementation processes of the above three groups of embodiments.

[0060] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0061] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A neurological sleep monitoring method, characterized in that: The following steps are involved: Step 1: Perform brain wave monitoring on the sleep monitor participating in the neurology department, and generate a brain wave waveform belonging to the sleep monitor based on the monitoring data of the sleep monitor throughout the sleep process, and divide the brain wave waveform into multiple similar feature segments based on preset waveform standard features; Step 2: based on the multiple similar feature segments confirmed in succession, lock the feature segment set column, and then perform feature analysis on the similar feature segments arranged in sequence in the feature segment set column, evaluate whether the corresponding similar feature segments can be merged into new similar feature segments, and re-mark them into the feature segment set column; Step three, based on the processed feature segment set column, confirm the sleep stages corresponding to this feature segment set column from front to back, and compare the sleep stage sequence confirmed in sequence with the set standard sleep segment to assess whether the two are consistent. If they are consistent, analyze and process the abnormal segment. If not, directly generate an abnormal signal for relevant display, where the standard sleep segment is a preset arrangement segment.

2. A neurological sleep monitoring method according to claim 1, characterized in that: In step 1, the specific method of dividing the brain wave waveform into multiple feature segments is: S11. Based on the confirmed brain wave waveform, a plurality of peak points existing in the brain wave waveform are confirmed in sequence, wherein the trend of the waveform segments before and after the peak point is opposite, and the trend of the waveform segment before the peak point is upward, while the trend of the waveform segment after the peak point is downward; S12, determine the waveform segments of adjacent peak points, and calibrate the wave speed of the corresponding waveform segment generation process as V i , and mark the vertical distance between adjacent peak points as L i , where i represents the waveform segment between different adjacent peak points, using F i =V i ÷L i Confirm the band frequency F of this waveform segment i ; S13, according to the preset waveform standard characteristics, the band frequency F i The waveform standard feature is marked as a characteristic waveform segment, and the characteristic waveform segments that appear continuously and belong to the same waveform standard feature are marked as similar characteristic segments. Several waveform segments generated by the EEG waveform are analyzed, and similar characteristic segments belonging to different waveform standard features are confirmed in turn. The different waveform standard features are preset interval features, which correspond to different sleep stages respectively.

3. A neurological sleep monitoring method according to claim 1, characterized in that: In step 2, the evaluation and processing method of the feature segment set sequence is: S21. Identify whether there are similar feature segments associated with the same sleep stage in the feature segment set: If they exist, other similar feature segments contained between two groups of similar feature segments or multiple groups of similar feature segments are marked as subsidiary feature segments, and the total length of the subsidiary feature segments is confirmed and marked as D1. Then the total length of the line segments between two groups of similar feature segments or multiple groups of similar feature segments is marked as D2, and D1÷D2=ZB is used to determine the total proportion ZB of its subsidiary feature segments. If ZB≤Y1, other similar feature segments contained between two groups of similar feature segments or multiple groups of similar feature segments are marked as abnormal bands of this sleep stage. This sleep stage is the sleep stage corresponding to the two groups of similar feature segments or multiple groups of similar feature segments. After the abnormal bands are marked, the two groups of similar feature segments or multiple groups of similar feature segments are merged into new similar feature segments.

4. A neurological sleep monitoring method according to claim 3, characterized in that: If ZB>Y1, where Y1 is a preset value, no fusion processing is performed.

5. A neurological sleep monitoring method according to claim 3, characterized in that: If the feature segment set does not contain the same type of feature segment associated with the same sleep stage, no processing is performed.

6. A neurological sleep monitoring method according to claim 1, characterized in that: In step 3, the specific method of checking the confirmed sleep stage sequence with the set standard sleep segment is: S31, based on the determined feature segment set sequence, confirm the sleep stages associated with the same feature segments in the feature segment set sequence from front to back, so as to lock the sleep stage sequence, compare the sleep stage sequence with the set standard sleep segment, and identify whether the two are completely consistent: if they are consistent, execute step S32 to analyze and process the abnormal segment, if not, directly generate an abnormal signal display; S32, determining the abnormal band marked by the same characteristic segment in the sleep stage sequence, and confirming the minimum waveform frequency B of the waveform standard feature based on the waveform standard feature corresponding to the same characteristic segment k min and maximum waveform frequency B k max, where k represents different waveform standard characteristics, and then based on the band frequency F confirmed by the corresponding waveform segment i , determine the multiple F associated with this abnormal band i , and then from multiple F i Select the minimum value F i min and maximum value F i max, determine the frequency interval associated with the abnormal band corresponding to this similar characteristic segment; S33, compare the difference between the waveform standard feature and the wave frequency interval: determine the associated characteristic value of this abnormal band: Tz = |B k max-F i max|+|B k min-F i min|Lock the associated characteristic value Tz, then confirm the associated characteristic values ​​of the abnormal bands associated with other similar characteristic segments, sum the confirmed groups of associated characteristic values ​​Tz, and determine the total value ZH; S34, comparing the total value ZH with the preset value Y2, where Y2 is the preset value. If ZH>Y2, a violent fluctuation signal is generated, and the generated violent fluctuation signal is displayed.

7. A neurological sleep monitoring method according to claim 6, characterized in that: In the step S34, if ZH≤Y2, a normal fluctuation signal is generated and the generated normal fluctuation signal is displayed.

8. A neurology sleep monitoring system, the monitoring system being used to execute the neurology sleep monitoring method according to any one of claims 1 to 7, characterized in that: include: The waveform segment division end divides the brain wave waveform generated during the brain wave monitoring process into a plurality of similar feature segments based on preset waveform standard features; The feature segment processing end locks the feature segment set column based on the multiple similar feature segments confirmed in succession, and then performs feature analysis on the similar feature segments arranged in sequence in the feature segment set column, assesses whether the corresponding similar feature segments can be merged into new similar feature segments, and re-calibrates them into the feature segment set column; The sleep stage checking end, based on the processed feature segment set sequence, confirms the sleep stages corresponding to the feature segment set sequence from front to back, checks the confirmed sleep stage sequence with the set standard sleep segment, and evaluates whether the two are consistent. If they are consistent, the abnormal segment is analyzed and processed and a violent fluctuation signal or a normal fluctuation signal is generated. If they are not consistent, an abnormal signal is directly generated for display; The display end displays the generated violent fluctuation signal, normal fluctuation signal, or abnormal signal.

Images