Smart ward data monitoring method based on wireless communication technology

Abstract

The present invention relates to the field of ward data monitoring technology, and in particular to a smart ward data monitoring method based on wireless communication technology. The method includes: obtaining text data and decibel values ​​corresponding to the voice signals of patients in the smart ward; segmenting the text data, combining the decibel value of each character, the overall time interval of the characters in the segment where each character is located, and the time interval between the segment where each character is located and its adjacent segment, to obtain initial important indicators; hierarchically clustering the segments according to the time intervals between adjacent segments, and obtaining interval uniformity values ​​based on the clustering results; analyzing the time intervals between the characters in the nodes of each character in each layer to be analyzed, and obtaining an overall stability value; combining the initial important indicators, the interval uniformity values, and the overall stability values, to determine the importance of each character, and then encoding and transmitting the text data. The present invention improves the reliability of the smart ward data monitoring results.

Description

Technical Field

[0001] The present invention relates to the technical field of ward data monitoring, and in particular to a smart ward data monitoring method based on wireless communication technology. Background Art

[0002] In existing smart wards, when a patient presses a call button, the call system of the nurse station or central nursing desk will be triggered. The nurse or medical staff will receive the call signal and then ask the patient about the urgency or needs through the receiver. This process transmits and presents the patient's voice data through the display screen or speaker on the receiver. The existing transmission process transmits compressed data through compression coding, but when the number of people calling at the same time in the hospital is large, it is easy to cause channel interference, which in turn makes the transmission effect poor, thereby affecting the reliability and accuracy of the smart ward data monitoring results. Summary of the invention

[0003] In order to solve the problem of low reliability of monitoring results in existing methods when monitoring smart ward data, the purpose of the present invention is to provide a smart ward data monitoring method based on wireless communication technology, and the technical solution adopted is as follows:

[0004] The present invention provides a smart ward data monitoring method based on wireless communication technology, the method comprising the following steps:

[0005] Obtain the text data and decibel value corresponding to the voice signal of the patient in the smart ward;

[0006] The text data is segmented using the time intervals between adjacent characters in the text data; the initial important index of each character is obtained by combining the decibel value of each character, the overall time interval of the characters in the segment where each character is located, and the time interval between the segment where each character is located and its adjacent segment;

[0007] According to the time intervals between adjacent segments, hierarchical clustering is performed on the segments to obtain a hierarchical clustering tree; the variation trend of the number of nodes in the hierarchical clustering tree, the overall discrete degree of the time intervals between adjacent segments in each layer to be analyzed, and the discrete degree of the time intervals between adjacent characters in the node are analyzed to obtain an interval uniformity value; the layer to be analyzed is any layer except the first layer and the last layer in the hierarchical clustering tree;

[0008] Analyze the time interval between characters in the nodes of each layer to be analyzed to obtain the overall stability value of the ward data; comprehensively consider the initial important index, the interval average value and the overall stability value to determine the importance of each character;

[0009] The text data is encoded and transmitted based on the importance.

[0010] Preferably, the initial important index of each character is obtained by combining the decibel value of each character, the overall time interval of the characters in the segment where each character is located, and the time interval between the segment where each character is located and its adjacent segment, including:

[0011] According to the normalized result of the decibel value of the candidate character, the average level of the time intervals of all adjacent characters in the segment where the candidate character is located, and the time interval between the segment where the candidate character is located and its adjacent segments, the initial important index of the candidate character is obtained, wherein the normalized result is positively correlated with the initial important index, and the average level and the time interval between the segment where the candidate character is located and its adjacent segments are both negatively correlated with the initial important index;

[0012] The candidate character is any character in the text data.

[0013] Preferably, the analysis of the variation trend of the number of nodes in the hierarchical clustering tree, the overall discrete degree of the time intervals between adjacent segments in each layer to be analyzed, and the discrete degree of the time intervals between adjacent characters in the node to obtain the interval uniformity value includes:

[0014] Count the number of nodes in each layer to be analyzed in the hierarchical clustering tree, and sort the number of nodes in all layers to be analyzed to obtain a node number sequence;

[0015] Analyzing the change characteristics of the elements in the node quantity sequence to obtain a trend characteristic value;

[0016] According to the discrete degree of the characteristic distance between all adjacent nodes in the candidate layer and the discrete degree of the time interval between all adjacent characters in the candidate layer, the inter-class difference of the candidate layer is obtained, wherein the characteristic distance is obtained based on the time interval between all adjacent characters in the node; the candidate layer is any layer to be analyzed;

[0017] The PCA algorithm is used to process the inter-class differences of all layers to be analyzed in the hierarchical clustering tree to obtain the trend angle value;

[0018] The trend characteristic value and the trend angle value are combined to obtain an interval uniform value.

[0019] Preferably, analyzing the change characteristics of the elements in the node quantity sequence to obtain the trend characteristic value includes:

[0020] Use the order value of the elements in the node quantity sequence as the horizontal coordinate and the elements as the vertical coordinate to obtain the coordinate point;

[0021] The coordinate points are used as the input of PCA to obtain the projection length value corresponding to each projection vector, the projection vector corresponding to the maximum projection length value is used as the trend vector, the arc tangent value of the ratio of the ordinate to the abscissa of the trend vector is recorded as the trend angle, and the normalized result of the ratio of the trend angle to 90° is used as the trend characteristic value.

[0022] Preferably, the combining the trend characteristic value and the trend angle value to obtain the interval uniform value includes:

[0023] The ratio between the trend characteristic value and the trend angle value is used as the interval uniform value.

[0024] Preferably, the step of analyzing the time interval between characters in each node of each layer to be analyzed to obtain the overall stability value of the ward data includes:

[0025] The time intervals between all adjacent characters in each node in each layer to be analyzed constitute an interval value sequence corresponding to each node in each layer to be analyzed; according to the overall difference between each element and all other elements in the interval value sequence corresponding to each node in each layer to be analyzed, the comprehensive difference value of each layer to be analyzed is obtained;

[0026] According to the inter-class difference and the comprehensive difference value of each layer to be analyzed, the overall stability value of the ward data is obtained, and the inter-class difference and the comprehensive difference value are both negatively correlated with the overall stability value.

[0027] Preferably, the method of obtaining the comprehensive difference value of each layer to be analyzed according to the overall difference between each element and all other elements in the interval value sequence corresponding to each node in each layer to be analyzed includes:

[0028] The overall difference between each element in the interval value sequence corresponding to the candidate node and all other elements is used as the center distance of each element in the interval value sequence corresponding to the candidate node; the maximum value of the center distances of all elements in the interval value sequence corresponding to the candidate node is determined as the interval representation value of the candidate node; the candidate node is any node in any layer to be analyzed;

[0029] According to the difference in the interval characterization values ​​between every two nodes in each layer to be analyzed, the comprehensive difference value of each layer to be analyzed is determined.

[0030] Preferably, the comprehensive initial importance index, the interval average value and the overall stability value are used to determine the importance of each character, including:

[0031] The importance of the candidate characters is obtained according to the initial importance index, spacing uniformity and overall stability value of the candidate characters, wherein the initial importance index is positively correlated with the importance, and the spacing uniformity and the overall stability value are negatively correlated with the importance.

[0032] Preferably, the segmenting of the text data by using the time intervals between adjacent characters in the text data includes:

[0033] Threshold segmentation is performed on the time intervals between all adjacent characters in the text data to obtain a number of segmentation segments.

[0034] Preferably, encoding and transmitting text data based on the importance level includes:

[0035] The MTF encoding method is used to encode and transmit the characters in the text data according to the order of importance.

[0036] The present invention has at least the following beneficial effects:

[0037] The present invention takes into account that under normal circumstances, when the patient's needs are more urgent, the speaking speed will often be faster and the speaking voice will be higher. First, the text data is divided by the time interval between adjacent characters in the text data corresponding to the voice signal of the patient in the smart ward. The time interval between adjacent characters in the same segment is relatively close, and the time interval between adjacent characters in different segments is relatively different. Combined with the decibel value of each character, the overall time interval of the characters in the segment where each character is located, and the time interval between the segment where each character is located and its adjacent segment, the importance of each character is preliminarily evaluated, and the initial important index is obtained; hierarchical clustering is performed on the segment based on the time interval between adjacent segments, and combined with hierarchical clustering, the segment is clustered. The changing trend of the number of nodes in the clustering tree, the overall discreteness of the time intervals between adjacent segments in each layer to be analyzed, and the discreteness of the time intervals between adjacent characters in the node were used to obtain the interval uniformity value. The overall stability value of the ward data was obtained for the time interval between characters in each node in each layer to be analyzed, which is used to reflect the overall stability of the patient's speech. Furthermore, the initial important indicators, the interval uniformity value and the overall stability value were combined to obtain the accurate importance of each character. The more important the character, the more features it can reflect, and the more attention it should be paid when encoding. Therefore, the text data is encoded and transmitted based on the importance, which improves the reliability and accuracy of the smart ward data monitoring results. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 A flow chart of a smart ward data monitoring method based on wireless communication technology provided by an embodiment of the present invention;

[0040] Figure 2 A flowchart of a method for obtaining initial important indicators of a character provided by an embodiment of the present invention;

[0041] Figure 3 A flowchart of a method for obtaining an interval uniform value provided by an embodiment of the present invention;

[0042] Figure 4 A flowchart of a method for obtaining the importance of a character provided by an embodiment of the present invention;

[0043] Figure 5 This is a structural block diagram of a smart ward data monitoring system based on wireless communication technology provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the smart ward data monitoring method based on wireless communication technology proposed by the present invention is described in detail below in combination with the accompanying drawings and preferred embodiments.

[0045] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0046] The specific scheme of the smart ward data monitoring method based on wireless communication technology provided by the present invention is described in detail below with reference to the accompanying drawings.

[0047] Embodiment of the smart ward data monitoring method based on wireless communication technology:

[0048] The specific scenario targeted by this embodiment is: in a smart ward, when a patient calls through the calling system, a large amount of monitoring data will be generated, and the generated monitoring data needs to be encoded and transmitted. Since the urgency or needs of different calls may be different, it is necessary to combine the urgency of the call and the needs of the patient, encode them, and then complete the transmission.

[0049] This embodiment proposes a smart ward data monitoring method based on wireless communication technology, such as Figure 1 As shown, the smart ward data monitoring method based on wireless communication technology of this embodiment includes the following steps:

[0050] Step S1, obtaining text data and decibel values ​​corresponding to the voice signals of patients in the smart ward.

[0051] In existing smart wards, patients use call buttons to communicate with medical staff. After receiving the call, the medical staff will ask the patient about the urgency or needs through the receiver. This process is carried out through the display and speaker on the receiver. The call button of the smart ward has a built-in microphone to collect the patient's voice, ensuring that the quality and position of the microphone can accurately capture the patient's voice and collect the patient's voice data. At the same time, the microphone has a built-in processing chip that can transmit the collected data. First, the patient's voice signal collected in the ward is split into two parts through the built-in chip of the sensor, one part is the voice decibel data, and the other part is the text data obtained by voice conversion, that is, the text data corresponding to the voice signal and the decibel value of each character in the text data are obtained.

[0052] At this point, the text data corresponding to the voice signal of the patient in the smart ward and the decibel value of each character in the text data have been obtained.

[0053] Step S2, segmenting the text data using the time intervals between adjacent characters in the text data; combining the decibel value of each character, the overall time interval of the characters in the segment where each character is located, and the time interval between the segment where each character is located and its adjacent segments, to obtain the initial importance index of each character.

[0054] In the smart ward, the intensity (decibel) of the sound emitted by the patient through the call system and the time interval between calls are important clues for assessing the urgency of the call. Generally speaking, a high decibel means that the patient feels extreme discomfort or pain and needs a quick response. Usually, when the patient's needs are more urgent, he speaks faster, that is, the time interval between the language he emits is often shorter, and the patient needs immediate medical assistance. Therefore, the time interval between characters in the patient's voice signal and the decibel value can be combined to make a preliminary evaluation of the patient's call and obtain the initial important indicators of each character.

[0055] Step S21, dividing the text data into a plurality of segments according to the time intervals between adjacent characters in the text data.

[0056] First, the absolute value of the difference in the timestamps between every two adjacent characters in the text data is obtained, recorded as the character interval between every two adjacent characters, and the ratio between each character interval and the maximum character interval is taken as the time interval, that is, the time interval between every two adjacent characters in the text data is obtained. It should be noted that there is a time interval between every two adjacent characters in the text data.

[0057] All time intervals are sorted in order from small to large to obtain a time interval sequence. The Otsu threshold segmentation algorithm is used for processing, that is, multiple threshold segmentation is performed on the elements in the time interval sequence to obtain multiple sub-intervals. According to the division results of the time interval sequence, the characters in the text data are divided to obtain multiple segmentation segments. The time intervals between adjacent characters in the same segmentation segment are located in the same sub-interval, and the time intervals between adjacent characters in different segmentation segments are located in different sub-intervals.

[0058] At this point, the division of the text data is completed and multiple segments are obtained.

[0059] Step S22, combining the decibel value of each character, the overall time interval of the characters in the segment where each character is located, and the time interval between the segment where each character is located and its adjacent segment, to obtain the initial importance index of each character.

[0060] Generally speaking, when patients' needs are more urgent, they speak faster and louder, so we first use this feature to make a preliminary evaluation of the importance of each character in the text data and obtain the initial importance index of each character.

[0061] Next, a character in text data is taken as an example for explanation. Other characters can also be processed using the method provided in this embodiment.

[0062] Specifically, any character in the text data is recorded as a candidate character.

[0063] The initial importance index of the candidate character is obtained based on the normalized result of the decibel value of the candidate character, the average level of the time intervals of all adjacent characters in the segment where the candidate character is located, and the time interval between the segment where the candidate character is located and its adjacent segments. The normalized result is positively correlated with the initial importance index, and the average level and the time interval between the segment where the candidate character is located and its adjacent segments are negatively correlated with the initial importance index.

[0064] Among them, a positive correlation means that the dependent variable will increase as the independent variable increases, and the dependent variable will decrease as the independent variable decreases. The specific relationship can be a multiplication relationship, an addition relationship, or the power of an exponential function; a negative correlation means that the dependent variable will decrease as the independent variable increases, and the dependent variable will increase as the independent variable decreases. It can be a subtraction relationship, a division relationship, etc., which is determined by actual application.

[0065] The average level of time intervals of all adjacent characters in the segment where the candidate character is located can be expressed by indicators such as the mode of the time intervals of all adjacent characters in the segment where the candidate character is located, the median of the time intervals of all adjacent characters in the segment where the candidate character is located, and the average level of the time intervals of all adjacent characters in the segment where the candidate character is located.

[0066] In this embodiment, a specific calculation formula for the initial importance index is given, and the initial importance index of the candidate character can be expressed as:

[0067]

[0068] Among them, A represents the initial important index of the candidate character, f represents the decibel value of the candidate character, and f max represents the maximum decibel value of all characters in the text data, e represents a natural constant, d1 represents the time interval between the segment where the candidate character is located and its adjacent segment on the left, d2 represents the time interval between the segment where the candidate character is located and its adjacent segment on the right, Indicates the average value of the time intervals between all adjacent characters in the segment where the candidate character is located.

[0069] The method for obtaining the time interval between two segments is: taking the time interval between the first character in the latter of the two segments and the last character in the former of the two segments as the time interval between the two segments.

[0070] It should be noted that: if the segment where the candidate character is located has only one adjacent segment, that is, the segment where the candidate character is located is the first segment or the last segment, then the time interval between the adjacent segment where the candidate character exists and the segment where the candidate character is located is used to replace the initial important index calculation formula

[0071] The normalized decibel value used to characterize the candidate character. It is used to reflect the average level of time intervals between all adjacent characters in the segment where the candidate character is located. The larger the value, the larger the time interval between the characters in the segment where the candidate character is located. It indicates the average value of the time interval between the segment where the candidate character is located and its adjacent segment. The larger the value, the larger the time interval between the segment where the candidate character is located and its adjacent segment. When the normalized decibel value of the candidate character is larger, the average value of the time intervals of all adjacent characters in the segment where the candidate character is located is smaller, and the average value of the time intervals between the segment where the candidate character is located and its adjacent segment is smaller, it means that when the patient speaks this speech, the louder the voice is and the faster the speech speed is, the greater the importance of this paragraph of text is, and therefore the initial importance index of the candidate character is larger.

[0072] By using the above method, the initial important index of each character in the text data can be obtained, that is, the importance of the characters is preliminarily evaluated, such as Figure 2 As shown, this figure is a flow chart of the method for obtaining the initial important indicators of characters.

[0073] Step S3, hierarchical clustering is performed on the segments according to the time intervals between adjacent segments to obtain a hierarchical clustering tree; the changing trend of the number of nodes in the hierarchical clustering tree, the overall discrete degree of the time intervals between adjacent segments in each layer to be analyzed, and the discrete degree of the time intervals between adjacent characters in the node are analyzed to obtain a uniform interval value; the layer to be analyzed is any layer except the first layer and the last layer in the hierarchical clustering tree.

[0074] In a hospital environment, not all calls from patients require urgent medical care. For example, although the decibel level of a patient's call is high and the time interval is short, this may only indicate that they have questions or need help with daily affairs, such as asking about family visitation policies. In fact, when patients do not have urgent medical needs, their language usually exhibits a stable speech rate and decibel level, which helps medical staff distinguish between true emergencies and general inquiries. Therefore, the patient's call content, speech rate, decibel level, and behavioral performance should be comprehensively evaluated to accurately respond to their needs. This embodiment will first hierarchically cluster all segments according to the time interval between adjacent segments, and then evaluate the inter-class differences and intra-class differences after the distance to obtain the interval uniformity value.

[0075] Step S31: cluster the segments using a hierarchical clustering algorithm according to the time intervals between adjacent segments to obtain a hierarchical clustering tree.

[0076] According to the time intervals between adjacent segments in the text data, a hierarchical clustering algorithm is used to perform bottom-up hierarchical clustering on the segments. The hierarchical clustering algorithm is a prior art and will not be described in detail here.

[0077] At this point, a hierarchical clustering tree is obtained.

[0078] Step S32, obtaining the interval uniformity value according to the variation trend of the number of nodes in the hierarchical clustering tree, the overall discrete degree of the time intervals between adjacent segments in each layer to be analyzed, and the discrete degree of the time intervals between adjacent characters in the node.

[0079] Different layers in the hierarchical clustering tree can represent the uniformity of intervals at different similarity scales, and the number of nodes in each layer can represent the difference in uniformity of categories corresponding to different nodes. Therefore, the hierarchical clustering results can be used to accurately represent the uniformity of character intervals.

[0080] For the hierarchical clustering tree, from bottom to top, that is, from the low layer to the high layer of the hierarchical clustering tree structure, the number of nodes gradually decreases and the difference between the layers increases; in the process from bottom to top, if the similarity of the time interval at different scales is greater, it means that the distinction between different scales is not obvious, resulting in fewer layers in the tree structure, that is, the number of nodes decreases faster.

[0081] Each layer except the first and last layer in the hierarchical clustering tree is recorded as a layer to be analyzed, that is, multiple layers to be analyzed are obtained. The number of nodes in each layer to be analyzed in the hierarchical clustering tree is counted, and the number of nodes in all layers to be analyzed is sorted from bottom to top in the hierarchical clustering tree to obtain a node number sequence.

[0082] The order value of the elements in the node quantity sequence is used as the horizontal coordinate, and the elements are used as the vertical coordinate to obtain multiple coordinate points; the coordinate points are used as the input of the principal component analysis algorithm (PCA) to obtain the projection length value corresponding to each projection vector, and the projection vector corresponding to the maximum projection length value is used as the trend vector, and the arc tangent value of the ratio of the vertical coordinate to the horizontal coordinate of the trend vector is recorded as the trend angle, and the normalized result of the ratio of the trend angle to 90° is used as the trend characteristic value. Normalization can be processed by the linear normalization method, which is a prior art and will not be described in detail here.

[0083] Next, any layer to be analyzed is taken as an example for description, and other layers to be analyzed can be processed by the method provided in this embodiment.

[0084] Any layer to be analyzed is recorded as a candidate layer, and the difference between the discrete degree of the feature distance between all adjacent nodes in the candidate layer and the discrete degree of the time interval between all adjacent characters in the candidate layer is taken as the inter-class difference of the candidate layer. It should be noted that the difference between the discrete degree of the feature distance between all adjacent nodes in the candidate layer and the discrete degree of the time interval between all adjacent characters in the candidate layer is obtained by subtracting the discrete degree of the time interval between all adjacent characters in the candidate layer from the discrete degree of the feature distance between all adjacent nodes in the candidate layer.

[0085] The specific method for obtaining the characteristic distance is as follows: for any node, the average value of the time intervals between all adjacent characters in the node is calculated based on the time intervals between each two adjacent characters in the node, and the average value is recorded as the average interval corresponding to the node; using this method, the average interval corresponding to each node can be obtained. For any two adjacent nodes in the candidate layer, the average value of the average intervals corresponding to the two nodes is taken as the characteristic distance between the corresponding two nodes.

[0086] The degree of discreteness of feature distance can be represented by various indicators such as standard deviation and variance of feature distance. In the present embodiment, the standard deviation of feature distance is used to characterize the degree of discreteness of feature distance, that is, the standard deviation of feature distance between all adjacent nodes in the candidate layer is used as the degree of discreteness of feature distance between all adjacent nodes in the candidate layer. Similarly, the degree of discreteness of time interval can be represented by various indicators such as standard deviation and variance of time interval. In the present embodiment, the standard deviation of time interval is used to characterize the degree of discreteness of time interval, that is, the standard deviation of time interval between all adjacent characters in the candidate layer is used as the degree of discreteness of time interval between all adjacent characters in the candidate layer.

[0087] For all layers to be analyzed in the hierarchical clustering tree: the inter-class differences of all layers to be analyzed from bottom to top constitute an inter-class difference sequence. The PCA algorithm is used to process the inter-class difference sequence to obtain the trend angle value, which is the value after normalization. There are many normalization methods, and the linear normalization method can be used for processing.

[0088] In the process of the hierarchical clustering tree from bottom to top, the faster the number of nodes decreases and the slower the difference increases, the greater the interval uniformity; the larger the trend angle value, the more uneven the interval. Based on this, the ratio between the trend feature value and the trend angle value is used as the interval uniformity value.

[0089] At this point, the interval uniform value is obtained, such as Figure 3 As shown, the figure is a flow chart of the method for obtaining the interval uniform value.

[0090] Step S4, analyzing the time interval between characters in each node of each layer to be analyzed to obtain the overall stability value of the ward data; and determining the importance of each character by combining the initial important index, the interval uniformity value and the overall stability value.

[0091] For call decibels, in the clustering layer with larger differences in character intervals, the comprehensive difference values ​​of character categories corresponding to different nodes are smaller, which means that the decibel stability is stronger, the patient's current urgency is lower, and it is closer to normal daily affairs inquiries.

[0092] In the process of encoding speech signals, in order to ensure that important information is not lost, it is necessary to pay attention to the importance of the data, and pay different attention to data of different importance to ensure that important features are not lost. For any layer to be analyzed in the hierarchical clustering tree, it contains multiple nodes, each node contains multiple time interval values, and the time interval between characters in the nodes of each layer to be analyzed, the initial important index and the average value of the interval are comprehensively evaluated to evaluate the importance of each character.

[0093] Step S41, analyzing the time interval between characters in each node in each layer to be analyzed, and obtaining the overall stability value of the ward data.

[0094] Considering that in general, the more uniform the time intervals between characters in text data, the more likely the patient is to be in a stable state, that is, without emergency needs, this embodiment will analyze the time intervals between characters in each node in each layer to be analyzed to evaluate the overall stability of the ward data.

[0095] The time intervals between all adjacent characters in each node in each layer to be analyzed constitute an interval value sequence corresponding to each node in each layer to be analyzed; that is, for any layer to be analyzed, each node in the layer to be analyzed corresponds to an interval value sequence.

[0096] Any node of any layer to be analyzed is recorded as a candidate node, and the overall difference between each element and all other elements in the interval value sequence corresponding to the candidate node is taken as the center distance of each element in the interval value sequence corresponding to the candidate node; the overall difference between each element and all other elements in the interval value sequence corresponding to the candidate node is obtained as follows: in the interval value sequence corresponding to the candidate node, the absolute value of the difference between each element and each other element is taken as the difference between each element and each other element; according to the difference between each element and other elements, the average value of the difference between each element and all other elements is calculated, and the average value is taken as the overall difference between each element and all other elements. The maximum value of the center distances of all elements in the interval value sequence corresponding to the candidate node is determined as the interval representation value of the candidate node.

[0097] By adopting the above method, the interval characterization value of each node in each layer to be analyzed can be obtained.

[0098] According to the difference of the interval characterization value between every two nodes in each layer to be analyzed, the comprehensive difference value of each layer to be analyzed is determined. Specifically: for any layer to be analyzed: the absolute value of the difference of the interval characterization value between every two nodes in the layer to be analyzed is calculated respectively, and the absolute value is used as the difference of the interval characterization value of the corresponding two nodes; according to the difference of the interval characterization value between every two nodes in the layer to be analyzed, the average value of the difference of all the interval characterization values ​​in the layer to be analyzed is calculated, and the average value is used as the comprehensive difference value of the layer to be analyzed. By adopting this method, the comprehensive difference value of each layer to be analyzed can be obtained.

[0099] Next, the overall stability value of the ward data is obtained according to the inter-class difference and the comprehensive difference value of each layer to be analyzed, and the inter-class difference and the comprehensive difference value are both negatively correlated with the overall stability value.

[0100] A negative correlation means that the dependent variable will decrease as the independent variable increases, and the dependent variable will increase as the independent variable decreases. It can be a subtraction relationship, a division relationship, etc., which is determined by actual application.

[0101] In this embodiment, a specific calculation formula for the overall stability value of the ward data is given, and the overall stability value of the ward data can be expressed as:

[0102]

[0103] Among them, W represents the overall stability value of the ward data, n represents the number of layers to be analyzed, and a i represents the inter-class difference of the i-th layer to be analyzed, b i represents the comprehensive difference value of the i-th layer to be analyzed, and e represents a natural constant.

[0104] When the inter-class difference of the layer to be analyzed in the hierarchical clustering tree is smaller and the comprehensive difference value is smaller, it means that the overall speech signal is more stable, that is, the overall stability value of the ward data is larger. When the inter-class difference of the layer to be analyzed in the hierarchical clustering tree is larger and the comprehensive difference value is larger, it means that the overall speech signal is more unstable, that is, the overall stability value of the ward data is smaller.

[0105] Using the above method, the overall stability value of the ward data was obtained.

[0106] Step S42, comprehensively considering the initial importance index, the interval uniformity value and the overall stability value to determine the importance of each character.

[0107] In order to improve the accuracy of the calculation results of character importance, the initial important index, the interval average value and the overall stability value of the ward data were further combined to conduct a comprehensive evaluation of the importance of the characters.

[0108] Specifically, the candidate characters are still used as an example for analysis, and other characters can be processed using the method provided in this embodiment.

[0109] The importance of the candidate characters is obtained according to the initial importance index, spacing uniformity and overall stability value of the candidate characters, wherein the initial importance index is positively correlated with the importance, and the spacing uniformity and the overall stability value are negatively correlated with the importance.

[0110] Among them, a positive correlation means that the dependent variable will increase as the independent variable increases, and the dependent variable will decrease as the independent variable decreases. The specific relationship can be a multiplication relationship, an addition relationship, or the power of an exponential function; a negative correlation means that the dependent variable will decrease as the independent variable increases, and the dependent variable will increase as the independent variable decreases. It can be a subtraction relationship, a division relationship, etc., which is determined by actual application.

[0111] In this embodiment, a specific calculation process of the importance is given. The specific calculation process of the importance of the candidate characters is: calculate the product of the interval average value and the overall stability value of the ward data, and take the ratio of the importance of the candidate character to the product as the importance of the candidate character.

[0112] So far, using the above method, the importance of each character can be obtained, such as Figure 4 As shown, the figure is a flowchart of a method for obtaining the importance of characters.

[0113] Step S5, encoding and transmitting the text data based on the importance.

[0114] The MTF (Move-to-Front) encoding method is an existing encoding method that requires the initial order of characters to be given before encoding. This embodiment adjusts the encoding strategy according to the importance of the characters. The more important the characters are, the smaller their corresponding index values ​​are, so they are encoded more frequently during the encoding process. Therefore, this embodiment uses the MTF encoding method to encode the characters in the text data according to the order of importance.

[0115] Specifically, all characters in the text data are sorted in order of importance from large to small to obtain a character sequence, and then the MTF encoding method is used to encode the characters in the character sequence to obtain an encoding result, the encoding result is compressed, and the decibel value of the text data is compressed at the same time, and the compressed result is transmitted.

[0116] So far, the compression and transmission of the patient's voice signal have been completed using the method provided in this embodiment.

[0117] This embodiment takes into account that under normal circumstances, when the patient's needs are more urgent, the speaking speed will often be faster and the speaking voice will be higher. First, the text data is divided by the time interval between adjacent characters in the text data corresponding to the voice signal of the patient in the smart ward. The time interval between adjacent characters in the same segment is relatively close, and the time interval between adjacent characters in different segments is relatively different. Combined with the decibel value of each character, the overall time interval of the characters in the segment where each character is located, and the time interval between the segment where each character is located and its adjacent segment, the importance of each character is preliminarily evaluated, and the initial important index is obtained; hierarchical clustering is performed on the segment based on the time interval between adjacent segments, and combined with the hierarchical The changing trend of the number of nodes in the clustering tree, the overall discreteness of the time intervals between adjacent segments in each layer to be analyzed, and the discreteness of the time intervals between adjacent characters in the node were used to obtain the interval uniformity value. The overall stability value of the ward data was obtained for the time interval between characters in each node in each layer to be analyzed, which is used to reflect the overall stability of the patient's speech. Furthermore, the initial important indicators, the interval uniformity value and the overall stability value were combined to obtain the accurate importance of each character. The more important the character, the more features it can reflect, and the more attention it should be paid when encoding. Therefore, the text data is encoded and transmitted based on the importance, which improves the reliability and accuracy of the smart ward data monitoring results.

[0118] Example of smart ward data monitoring system based on wireless communication technology:

[0119] See also Figure 5 , which shows a structural block diagram of a smart ward data monitoring system based on wireless communication technology provided by an embodiment of the present invention. The system may include a data acquisition module, an initial important indicator calculation module, an interval average value acquisition module, an importance determination module and a text data encoding module.

[0120] Among them, the data acquisition module is used to obtain the text data and decibel value corresponding to the voice signal of the patient in the smart ward;

[0121] The initial important index calculation module is used to segment the text data using the time intervals between adjacent characters in the text data; the initial important index of each character is obtained by combining the decibel value of each character, the overall time interval of the characters in the segment where each character is located, and the time interval between the segment where each character is located and its adjacent segment;

[0122] The interval average value acquisition module is used to perform hierarchical clustering on the segmented segments according to the time intervals between adjacent segmented segments to obtain a hierarchical clustering tree; analyze the change trend of the number of nodes in the hierarchical clustering tree, the overall discrete degree of the time intervals between adjacent segmented segments in each layer to be analyzed, and the discrete degree of the time intervals between adjacent characters in the node to obtain the interval average value; the layer to be analyzed is any layer except the first layer and the last layer in the hierarchical clustering tree;

[0123] The importance determination module is used to analyze the time interval between characters in each node of each layer to be analyzed to obtain the overall stability value of the ward data; the importance of each character is determined by combining the initial important index, the interval average value and the overall stability value;

[0124] The text data encoding module is used to encode and transmit the text data based on the importance level.

[0125] It should be understood that Figure 5 The structural block diagram of the smart ward data monitoring system based on wireless communication technology and its modules shown can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented by hardware, software, or a combination of software and hardware. Among them, the hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or a dedicated design hardware. Those skilled in the art can understand that the above methods and systems can be implemented using computer executable instructions and / or included in processor control codes, such as carrier media such as disks, CDs or DVD-ROMs, programmable memories such as read-only memories (firmware), or data carriers such as optical or electronic signal carriers. Such codes are provided on. The system and its modules of this specification can not only be implemented by hardware circuits such as ultra-large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but can also be implemented by software executed by various types of processors, and can also be implemented by a combination of the above hardware circuits and software (e.g., firmware).

[0126] For more details about the above modules, please refer to other places in this manual (for example Figure 1 , Figure 2 , Figure 3 , Figure 4 part and its related description), which will not be repeated here.

[0127] In other embodiments, a device is also provided, including a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, so that the device executes the above-mentioned smart ward data monitoring method based on wireless communication technology. The device can be specifically a chip, a component or a module, and the chip may include a connected processor and a memory; wherein the memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute the smart ward data monitoring method based on wireless communication technology provided in the above embodiment.

[0128] In other embodiments, a computer program product is also provided. When the computer program product is run on a computer, the computer executes the above-mentioned related steps to implement the smart ward data monitoring method based on wireless communication technology provided in the above embodiment.

[0129] In other embodiments, a computer-readable storage medium is also provided, in which a computer program code is stored. When the computer program code is run on a computer, the computer executes the above-mentioned related method steps to implement the smart ward data monitoring method based on wireless communication technology provided in the above-mentioned embodiment.

[0130] Among them, the provided systems, devices, computer program products, and computer-readable storage media are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0131] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A smart ward data monitoring method based on wireless communication technology, characterized in that: The method comprises the following steps: Obtain the text data and decibel value corresponding to the voice signal of the patient in the smart ward; The text data is segmented using the time intervals between adjacent characters in the text data; the initial important index of each character is obtained by combining the decibel value of each character, the overall time interval of the characters in the segment where each character is located, and the time interval between the segment where each character is located and its adjacent segment; According to the time intervals between adjacent segments, hierarchical clustering is performed on the segments to obtain a hierarchical clustering tree; the variation trend of the number of nodes in the hierarchical clustering tree, the overall discrete degree of the time intervals between adjacent segments in each layer to be analyzed, and the discrete degree of the time intervals between adjacent characters in the node are analyzed to obtain an interval uniformity value; the layer to be analyzed is any layer except the first layer and the last layer in the hierarchical clustering tree; Analyze the time interval between characters in the nodes of each layer to be analyzed to obtain the overall stability value of the ward data; comprehensively consider the initial important index, the interval average value and the overall stability value to determine the importance of each character; encoding and transmitting text data based on the importance; The encoding of the text data based on the importance level includes: All characters in the text data are sorted in descending order of importance to obtain a character sequence, and the characters in the character sequence are encoded using the MTF encoding method to obtain an encoding result.

2. The smart ward data monitoring method based on wireless communication technology according to claim 1 is characterized in that: The initial important index of each character is obtained by combining the decibel value of each character, the overall time interval of the characters in the segment where each character is located, and the time interval between the segment where each character is located and its adjacent segment, including: According to the normalized result of the decibel value of the candidate character, the average level of the time intervals of all adjacent characters in the segment where the candidate character is located, and the time interval between the segment where the candidate character is located and its adjacent segments, the initial important index of the candidate character is obtained, wherein the normalized result is positively correlated with the initial important index, and the average level and the time interval between the segment where the candidate character is located and its adjacent segments are both negatively correlated with the initial important index; The candidate character is any character in the text data.

3. The smart ward data monitoring method based on wireless communication technology according to claim 1 is characterized in that: The method of analyzing the variation trend of the number of nodes in the hierarchical clustering tree, the overall discrete degree of the time intervals between adjacent segments in each layer to be analyzed, and the discrete degree of the time intervals between adjacent characters in the node to obtain the interval uniformity value includes: Count the number of nodes in each layer to be analyzed in the hierarchical clustering tree, and sort the number of nodes in all layers to be analyzed to obtain a node number sequence; Analyzing the change characteristics of the elements in the node quantity sequence to obtain a trend characteristic value; According to the discrete degree of the characteristic distance between all adjacent nodes in the candidate layer and the discrete degree of the time interval between all adjacent characters in the candidate layer, the inter-class difference of the candidate layer is obtained, wherein the characteristic distance is obtained based on the time interval between all adjacent characters in the node; the candidate layer is any layer to be analyzed; The PCA algorithm is used to process the inter-class differences of all layers to be analyzed in the hierarchical clustering tree to obtain the trend angle value; The trend characteristic value and the trend angle value are combined to obtain an interval uniform value.

4. The smart ward data monitoring method based on wireless communication technology according to claim 3 is characterized in that: The step of analyzing the change characteristics of the elements in the node quantity sequence to obtain the trend characteristic value includes: Use the order value of the elements in the node quantity sequence as the horizontal coordinate and the elements as the vertical coordinate to obtain the coordinate point; The coordinate points are used as the input of PCA to obtain the projection length value corresponding to each projection vector, the projection vector corresponding to the maximum projection length value is used as the trend vector, the arc tangent value of the ratio of the ordinate to the abscissa of the trend vector is recorded as the trend angle, and the normalized result of the ratio of the trend angle to 90° is used as the trend characteristic value.

5. The smart ward data monitoring method based on wireless communication technology according to claim 3 is characterized in that: The step of combining the trend characteristic value and the trend angle value to obtain an interval uniform value includes: The ratio between the trend characteristic value and the trend angle value is used as the interval uniform value.

6. The smart ward data monitoring method based on wireless communication technology according to claim 3 is characterized in that: The analyzing the time interval between characters in each node of each layer to be analyzed to obtain the overall stability value of the ward data includes: The time intervals between all adjacent characters in each node in each layer to be analyzed constitute an interval value sequence corresponding to each node in each layer to be analyzed; according to the overall difference between each element and all other elements in the interval value sequence corresponding to each node in each layer to be analyzed, the comprehensive difference value of each layer to be analyzed is obtained; According to the inter-class difference and the comprehensive difference value of each layer to be analyzed, the overall stability value of the ward data is obtained, and the inter-class difference and the comprehensive difference value are both negatively correlated with the overall stability value.

7. The smart ward data monitoring method based on wireless communication technology according to claim 6 is characterized in that: The comprehensive difference value of each layer to be analyzed is obtained according to the overall difference between each element and all other elements in the interval value sequence corresponding to each node in each layer to be analyzed, including: The overall difference between each element in the interval value sequence corresponding to the candidate node and all other elements is used as the center distance of each element in the interval value sequence corresponding to the candidate node; the maximum value of the center distances of all elements in the interval value sequence corresponding to the candidate node is determined as the interval representation value of the candidate node; the candidate node is any node in any layer to be analyzed; According to the difference in the interval characterization values ​​between every two nodes in each layer to be analyzed, the comprehensive difference value of each layer to be analyzed is determined.

8. The smart ward data monitoring method based on wireless communication technology according to claim 2 is characterized in that: The comprehensive initial importance index, the interval average value and the overall stability value are used to determine the importance of each character, including: The importance of the candidate characters is obtained according to the initial importance index, spacing uniformity and overall stability value of the candidate characters, wherein the initial importance index is positively correlated with the importance, and the spacing uniformity and the overall stability value are negatively correlated with the importance.

9. The smart ward data monitoring method based on wireless communication technology according to claim 1 is characterized in that: The method of segmenting the text data by using the time intervals between adjacent characters in the text data includes: Threshold segmentation is performed on the time intervals between all adjacent characters in the text data to obtain a number of segmentation segments.

Images