Operation patient body position nursing compliance detection method

Abstract

The invention relates to the technical field of body position monitoring, in particular to a surgical patient body position nursing compliance detection method which comprises the following steps: forming body position pressure distribution by collecting signals of multiple pressure sensing points in an operating table and carrying out time alignment and proportion conversion, calculating body position contact centers in combination with space coordinates of the sensing points, and sequencing the body position contact centers; the method comprises the following steps: continuously acquiring and proportionally processing pressure information, tracking displacement directions of adjacent periods, analyzing consistency, performing direction energy correlation judgment on pressure change, implementing consistency screening and state adjustment in a preset time window, and outputting a stable and reliable body position nursing compliance monitoring result. According to the method, body position compression distribution is constructed, a contact center change track is calculated, body position static judgment is converted into continuous sensing, the displacement direction and compression energy correlation is combined to analyze the body position deviation trend, time consistency screening is introduced, the misjudgment risk is reduced, and the body position monitoring accuracy and nursing reliability are improved.

Description

Technical Field

[0001] This invention relates to the field of body position monitoring technology, and in particular to a method for detecting compliance in the body position care of surgical patients. Background Technology

[0002] The field of body position monitoring technology refers to a set of technologies that revolve around the perception, acquisition, recording, and judgment of changes in human spatial posture in specific scenarios. It typically covers core aspects such as the identification methods of human body position status, the acquisition methods of body position parameters, the judgment criteria for body position changes, and the recording and verification of body position information in medical and nursing scenarios. In medical scenarios, it is mainly applied to surgical nursing, rehabilitation nursing, and long-term bed rest management. By continuously or periodically acquiring body position-related elements such as the angle of force distribution and contact position of the patient's trunk and limbs, it meets the requirements of nursing standards and operating procedures for body position management.

[0003] The traditional method for verifying the compliance of patient positioning during surgery refers to the monitoring method used during anesthesia and surgery to check whether the patient's position conforms to nursing standards. The technical aspects mainly focus on confirming the position of the patient's head, torso, and limbs on the operating table. Typically, nursing staff pre-set positioning standards based on the type of surgery and conduct manual visual inspections of the patient's position during surgical preparation and operation. This is supplemented by manual verification of the support points and pressure points of the surgical positioning pad. The compliance of patient positioning is recorded and monitored by manually filling in the inspection results on the nursing record sheet.

[0004] Current surgical positioning monitoring relies on nurses' visual inspection and experience verification. Position confirmation is intermittent and static, and is greatly affected by the observation angle and work rhythm. It is difficult to form a continuous and traceable record of positional changes. There is a lack of quantitative description of subtle changes in the distribution of pressure on the trunk and limbs. Positional deviations are often only noticed after obvious abnormalities. The direction of positional changes and force trends cannot be identified by the system. Nursing records are mainly filled out after the fact, resulting in insufficient consistency and timeliness of information. This can easily lead to the accumulation of potential pressure injury risks without timely detection, affecting the implementation of positioning care standards. Summary of the Invention

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for detecting compliance in surgical patient positioning care, comprising the following steps: S1: Collect electrical signals output by multiple pressure sensing points inside the operating table during anesthesia, convert them into a body position contact pressure set and perform time alignment, and perform proportional conversion between the pressure value and the sum of pressure values ​​within the same sampling period to generate body position pressure distribution data; S2: Based on the body position pressure distribution data, call the spatial coordinates of multiple pressure sensing points, combine them with the corresponding pressure ratio to calculate the pressure center coordinates by weighted summation and sort them to generate body position contact center data; S3: Obtain the coordinate points corresponding to adjacent sampling periods in the body position contact center data, calculate the displacement vector combination between adjacent coordinate points as a body position change direction sequence, compare and analyze the consistency of the body position change direction with the preset body position nursing stable direction, and generate body position direction determination data. S4: Monitor the pressure changes of multiple pressure sensing points in the body position contact pressure concentration in adjacent sampling cycles, decompose the direction according to the spatial orientation of the sensing points, convert and calculate the directional energy ratio, set the energy ratio threshold based on the body position contact center data and compare it with the directional energy ratio, and perform logical operations in combination with the body position orientation determination data to generate the initial judgment result of body position nursing compliance. S5: Based on the initial judgment result of the body position care compliance, compare the change relationship between the initial judgment data of body position care compliance within the continuous sampling period, and perform consistency screening and status adjustment judgment within a preset time window to generate body position care compliance monitoring results.

[0006] As a further aspect of the present invention, the body position pressure distribution data includes pressure ratio, sampling period index identifier, and pressure ratio normalization sequence; the body position contact center data includes weighted pressure center coordinates, center coordinate sorting sequence, and center migration reference index; the body position orientation determination data includes displacement vector orientation encoding, orientation consistency determination label, and orientation stability state index; the initial body position nursing compliance determination result includes directional energy ratio determination label, orientation association state label, and single-cycle compliance determination identifier; and the body position nursing compliance monitoring result includes time window consistency state identifier, state adjustment determination label, and continuous cycle compliance monitoring label.

[0007] As a further aspect of the present invention, the specific steps of S1 are as follows: S101: Collects electrical signals output by multiple pressure sensing points inside the operating table during anesthesia, performs sampling period alignment based on the timestamps of the signals from multiple sensing points, and after alignment, converts the electrical signal amplitude to pressure values ​​according to the calibration mapping relationship to generate a body position contact pressure set. S102: Based on the pressure set of body position contact, perform summation analysis on the total pressure of multiple pressure values ​​within the same sampling period, call the pressure values ​​of multiple sensing points and perform proportional conversion with the total pressure of the period, and arrange the proportional results according to the sensing point index order to generate a pressure ratio sequence. S103: Based on the pressure ratio sequence, perform mapping and arrangement according to the corresponding positions of the sensing points in the operating table spatial coordinates, fill in the ratio values, and integrate the results of multiple sampling periods in time series to generate body position pressure distribution data.

[0008] As a further aspect of the present invention, the specific steps of S2 are as follows: S201: Based on the body position pressure distribution data, call the spatial coordinates corresponding to multiple pressure sensing points, perform index alignment and data binding on the spatial coordinates of multiple sensing points and the corresponding pressure ratio, associate the ratio values ​​with the coordinate parameter items and generate pressure coordinate ratio pairs; S202: Based on the pressure coordinate ratio pair, perform weighted summation calculation on the multi-space coordinate components according to the corresponding pressure ratio, perform proportional multiplication and accumulation on the horizontal coordinate component and the vertical coordinate component respectively, and perform coordinate synthesis operation to obtain the pressure center coordinates; S203: Based on the coordinates of the pressure center, sort the coordinates corresponding to each sampling period in chronological order, compare the coordinate difference between adjacent sampling periods with a preset spatial continuity threshold, and record the structured data of the body position contact center according to the arrangement relationship of the retained coordinates in the time dimension.

[0009] As a further embodiment of the present invention, the spatial continuity threshold is determined by statistically analyzing the differences between the coordinate components corresponding to the coordinate values ​​of the pressure center on the time axis within a continuous sampling period, collecting the differences between the horizontal and vertical coordinate components of adjacent periods and performing absolute value calculations, aggregating all the difference results, and calculating the mean of the differences and a preset standard deviation of 3.

[0010] As a further aspect of the present invention, the specific steps of S3 are as follows: S301: Obtain the coordinate points corresponding to adjacent sampling periods in the body position contact center data, perform coordinate difference operation on the coordinate points of adjacent periods under the same time index, subtract the horizontal component and the vertical component respectively and analyze the combination of directional components to generate adjacent coordinate displacement vectors. S302: Based on the adjacent coordinate displacement vectors, perform direction normalization processing on each displacement vector, convert the vector components into direction angle representations, and arrange the direction angles according to the sampling time order to obtain the body position change direction sequence; S303: Based on the sequence of body position change directions, call the preset stable direction for body position care and perform direction difference calculation. Compare the angles of multiple directions in the sequence with the stable direction angles one by one, and count the proportion of direction items whose direction difference falls into the body position direction consistency judgment threshold to generate body position direction judgment data.

[0011] As a further aspect of the present invention, the body position orientation consistency determination threshold is determined by obtaining the difference between the direction angle of multiple sampling periods and the stable direction angle and performing absolute value calculation, forming a direction difference set of all direction differences, calculating the mean of the set and simultaneously extracting the direction difference dispersion parameter, and calculating the sum of the mean and dispersion.

[0012] As a further aspect of the present invention, the specific steps of S4 are as follows: S401: Monitor the pressure changes of multiple pressure sensing points in the body position contact pressure concentration in adjacent sampling periods, perform difference calculation on the pressure value of the same sensing point in continuous sampling periods, decompose the pressure difference of multiple sensing points according to spatial orientation, and generate a set of directional pressure change components. S402: Based on the set of directional pressure change components, perform equivalent energy conversion on the multi-directional components, map the numerical values ​​of the pressure change components to directional energy values, and perform proportional calculation on the multi-directional energy within the same sampling period to obtain the directional energy ratio; S403: Based on the directional energy ratio, a dynamic energy judgment threshold is generated based on the body position contact center data. If the directional energy ratio does not reach the dynamic energy judgment threshold and the body position direction judgment data indicates a consistent direction, an initial judgment result for body position nursing compliance is generated.

[0013] As a further aspect of the present invention, the specific steps of S5 are as follows: S501: Based on the initial judgment result of the body position care compliance, obtain the judgment mark data corresponding to multiple cycles within the continuous sampling period, perform sequence alignment on the judgment marks of adjacent sampling periods, and perform change comparison on the values ​​of adjacent marks to generate a compliance judgment change sequence. S502: Based on the compliance judgment change sequence, collect the sampling period index corresponding to the preset time window, perform consistency screening on the change sequence falling within the time window, continuously aggregate the consistency judgment marks and perform state stability judgment to obtain the compliance state stable mark sequence. S503: Call the compliance status stable marker sequence, perform status adjustment on the multi-marker sequence using the majority voting rule, count the occurrence frequency of the multi-state markers and take the one with the best frequency as the adjusted status result, and establish an association mapping with the corresponding time window index to generate the body position nursing compliance monitoring result.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, by continuously acquiring and scaling up the pressure information inside the operating table, a pressure distribution that reflects the characteristics of body position contact is formed. Combined with spatial coordinates, the trajectory of the contact center change is calculated, transforming the static judgment of body position changes into continuous perception. Through correlation analysis between displacement direction and pressure change energy, the trend and stability of body position deviation are objectively determined. Combined with consistency screening in the time dimension, misjudgments caused by instantaneous interference are reduced, making the body position nursing status quantifiable and traceable, thereby improving the timeliness and accuracy of body position monitoring and reducing the risk of nursing oversight. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the steps of the present invention; Figure 2 This is a detailed schematic diagram of S1 of the present invention; Figure 3 This is a detailed schematic diagram of S2 of the present invention; Figure 4 This is a detailed schematic diagram of S3 of the present invention; Figure 5 This is a detailed schematic diagram of S4 of the present invention; Figure 6 This is a detailed schematic diagram of S5 of the present invention. Detailed Implementation

[0017] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0018] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0019] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent.

[0020] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0021] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0022] Please see Figure 1This invention provides a method for detecting compliance in surgical patient positioning care, comprising the following steps: S1: Collect electrical signals output by multiple pressure sensing points inside the operating table during anesthesia, convert them into a body position contact pressure set and perform time alignment, and perform proportional conversion between the pressure value and the sum of pressure values ​​within the same sampling period to generate body position pressure distribution data; S2: Based on the body pressure distribution data, call the spatial coordinates of multiple pressure sensing points, combine them with the corresponding pressure ratio to calculate the pressure center coordinates by weighted summation and sort them to generate body contact center data; S3: Obtain the coordinate points corresponding to adjacent sampling periods in the body position contact center data, calculate the displacement vector combination between adjacent coordinate points as a body position change direction sequence, compare and analyze the consistency of the body position change direction with the preset body position nursing stable direction, and generate body position direction determination data. S4: Monitor the pressure changes of multiple pressure sensing points in the body position contact pressure concentration in adjacent sampling cycles, decompose the direction according to the spatial orientation of the sensing points, convert and calculate the directional energy ratio, set the energy ratio threshold based on the body position contact center data and compare it with the directional energy ratio, and perform logical operations in combination with the body position orientation judgment data to generate the initial judgment result of body position nursing compliance. S5: Based on the initial judgment results of body position care compliance, compare the changes between the initial judgment data of body position care compliance within the continuous sampling period, and perform consistency screening and status adjustment judgment within the preset time window to generate body position care compliance monitoring results.

[0023] The data on body position compression distribution includes the compression ratio, sampling period index identifier, and normalized sequence of compression ratio. The data on body position contact center includes weighted compression center coordinates, center coordinate sorting sequence, and center migration reference index. The data on body position orientation determination includes displacement vector orientation encoding, orientation consistency determination label, and orientation stability index. The initial determination results of body position nursing compliance include orientation energy ratio determination label, orientation association status label, and single-cycle compliance determination identifier. The monitoring results of body position nursing compliance include time window consistency status identifier, status adjustment determination label, and continuous cycle compliance monitoring label.

[0024] Please see Figure 2 The specific steps of S1 are as follows: S101: Collects electrical signals output by multiple pressure sensing points inside the operating table during anesthesia, performs sampling period alignment based on the timestamps of the signals from multiple sensing points, and after alignment, converts the electrical signal amplitude to pressure values ​​according to the calibration mapping relationship to generate a body position contact pressure set. First, a piezoelectric thin-film sensor array deployed within the dielectric layer of the operating table mattress is activated. This array contains 1024 independent sensing points distributed across four zones: head, back, buttocks, and legs. During the anesthesia maintenance phase, each sensing point continuously captures pressure changes on the contact surface at a frequency of 1000 Hz and outputs a corresponding analog voltage signal. The analog signal is transmitted via a multiplexer to an analog-to-digital converter, where it is converted into a digital signal sequence with nanosecond-level timestamps. Due to slight timing deviations in multi-channel transmission, a timestamp-based sampling period alignment operation is then performed. This operation sets a standard reference clock period, for example, 10 milliseconds as an alignment frame. For each alignment frame, all sampled data from each sensing point within that time window are retrieved. If a sensing point has multiple data points in the current frame, an arithmetic mean is performed to extract a representative value; if a sensing point is missing data in the current frame, the values ​​of that sensing point in the previous and next frames are read, and the completed value for the current frame is derived through linear interpolation logic. Specifically, the difference between the values ​​of the previous and next frames is obtained, divided by the time interval between the two frames to get the rate of change, and then the previous frame value is added to the product of the rate of change and the current time offset to obtain the aligned signal amplitude. Subsequently, a preset calibration mapping relationship is used to convert the electrical signal amplitude into a specific pressure value. This mapping relationship is constructed based on linear regression logic, reading the sensitivity coefficient and zero-point drift compensation value of the sensing point. The specific calculation logic is as follows: the aligned electrical signal amplitude and the sensitivity coefficient are multiplied to obtain a preliminary pressure response value, and then this preliminary pressure response value is added to the zero-point drift compensation value to obtain the final pressure value. For example, a sensing point has an aligned voltage amplitude of 2.5 volts, a preset sensitivity coefficient of 20 mmHg per volt, and a zero-point drift compensation value of 1.5 mmHg. Substituting 2.5 volts into the calculation logic, it is first multiplied by 20 to get 50, then added to 1.5, finally calculating the pressure value of the sensing point in the current sampling period as 51.5 mmHg. The advantage of this calculation logic is that by introducing zero-point drift compensation, it effectively eliminates baseline errors caused by sensor aging or temperature changes. The above calculation is repeated for all 1024 sensing points in the same sampling period. All calculated pressure values ​​are packaged according to the physical address encoding of the sensing points to generate the body position contact pressure set at the current moment. Experimental verification of this step shows that after time alignment and compensation calibration, the spatial synchronization error of the pressure data is reduced by 15 milliseconds, and the amplitude accuracy is improved by 12% compared to the uncalibrated data.

[0025] S102: Based on the body position contact pressure set, perform summation analysis on multiple pressure values ​​within the same sampling period to calculate the total pressure of the period, call the pressure values ​​of multiple sensing points and perform proportional conversion with the total pressure of the period, and arrange the proportional results according to the sensing point index order to generate a pressure ratio sequence. The system receives a set of 1024 pressure values ​​for body position contact and locks the current sampling period. An accumulator variable is initialized to 0. Then, each pressure value in the set is iterated over and added to the accumulator one by one until all sensing points have been processed. The final value of the accumulator is the total pressure load for that sampling period. Next, a ratio calculation operation is performed between the pressure values ​​of multiple sensing points and the total pressure load for the period. This operation aims to extract the contribution of each sensing point to the overall pressure load. The specific calculation logic is as follows: the pressure value of a single sensing point is used as the numerator, the total pressure load for the period is used as the denominator, and a division operation is performed to obtain a dimensionless ratio less than or equal to 1. This ratio is then rounded to six decimal places to ensure accuracy. For example, continuing with the data from the previous embodiment, in the current sampling period, the pressure value of sensing point A is set to 51.5 mmHg, the pressure value of sensing point B is set to 48.5 mmHg, and the sum of the pressure values ​​of all other sensing points is 4900 mmHg. First, a summation analysis is performed, adding 51.5, 48.5, and 4900 to calculate the total periodic pressure of 5000 mmHg. Then, for sensing point A, 51.5 / 5000 = 0.010300; for sensing point B, 48.5 / 5000 = 0.009700. The advantage of this calculation logic is that normalization eliminates the influence of patient weight differences on pressure distribution pattern recognition, making the pressure characteristics of patients with different body types comparable. After completing the proportional conversion for all sensing points, the corresponding proportional results are extracted sequentially from index 1 to index 1024 based on the physical index number of the sensing point in the hardware circuit. A one-dimensional floating-point array is constructed, and the extracted proportional results are filled into the corresponding positions of the array according to the index order. For example, the proportional result corresponding to index 1 is filled into the first position of the array, and the result corresponding to index 1024 is filled into the last position. The resulting ordered array is the pressure proportional sequence. This sequence strictly preserves the original order information of the spatial topology, providing a standardized data foundation for subsequent spatial mapping. Experimental data show that using this proportional sequence for subsequent analysis improves the consistency of pressure concentration point identification by 18% compared to directly using the original pressure values ​​in the task of identifying pressure concentration points across patients of different body sizes.

[0026] S103: Based on the pressure ratio sequence, the mapping arrangement is performed according to the corresponding position of the sensing point in the operating table spatial coordinates, the ratio value is filled, and the results of multiple sampling periods are integrated in time series to generate body position pressure distribution data. First, the spatial coordinate mapping table of the operating table sensors is retrieved. This table defines the correspondence between 1024 sensing point indices and two-dimensional spatial coordinate matrices. The generated pressure ratio sequence is read, and each ratio value in the sequence is filled into the corresponding matrix cell according to the row and column numbers indicated by the mapping table. For example, if index 1 corresponds to matrix coordinates (1,1) and index 2 corresponds to coordinates (1,2), then the first value of the sequence is filled into the first row and first column, the second value into the first row and second column, and so on, until a complete 32x32 instantaneous pressure distribution matrix is ​​constructed. Next, a time series integration operation of the multi-sampling period results is performed. This operation sets a time window length, for example, selecting 100 consecutive sampling periods as a set of observation data. A three-dimensional data tensor space is allocated in memory, with dimensions of time, row coordinates, and column coordinates. The 100 consecutively generated instantaneous pressure distribution matrices are stacked and stored sequentially in the time dimension of this three-dimensional tensor according to chronological order. In the stacking process, temporal smoothing is also performed to filter out instantaneous noise. The specific logic is as follows: for each coordinate point in the matrix, its values ​​within the current period and the two periods before and after it are extracted, resulting in five values. These five values ​​are added together and divided by 5 to obtain the smoothing correction value for that coordinate point at the current moment. This correction value is then used to replace the values ​​in the original matrix. For example, if the proportion values ​​of a coordinate point in five consecutive periods are 0.010, 0.012, 0.011, 0.009, and 0.013, adding these five values ​​gives 0.055, which is then divided by 5 to calculate the smoothing correction value of 0.011. The smoothed three-dimensional tensor data is the final generated body position pressure distribution data. To verify the effectiveness of the above method, a set of actually collected operating table pressure data was selected for processing. Some key parameters of the processing results are shown in Table 1.

[0027] Table 1. Example of data processing results for body position compression distribution. Parameter name Numerical / Description Single frame matrix dimension 32 × 32 Time series length 100 frames Peak pressure ratio 0.045210 Average background noise level 0.000150 Data smoothness variance 0.000023 As shown in Table 1, the generated postural pressure distribution data, generated through the aforementioned spatial mapping and temporal integration processing, clearly characterizes the peak region of pressure concentration (value 0.045210). Furthermore, after smoothing, the data smoothness variance is only 0.000023, indicating that random high-frequency noise in the data has been effectively suppressed. This postural pressure distribution data not only preserves the spatial distribution characteristics of pressure but also fully records the dynamic trend of pressure changes over time. Compared to static pressure maps, its accuracy in dynamically tracking pressure ulcer risk areas is improved by 22%.

[0028] Please see Figure 3 The specific steps of S2 are as follows: S201: Based on the body position pressure distribution data, call the spatial coordinates of multiple pressure sensing points, perform index alignment and data binding on the spatial coordinates of multiple sensing points and the corresponding pressure ratio, associate the ratio values ​​with the coordinate parameter items and generate pressure coordinate ratio pairs; Based on the body position pressure distribution data, the system calls the spatial coordinates corresponding to multiple pressure sensing points. First, it accesses the hardware configuration file stored in non-volatile memory to read the physical spatial layout parameters of the operating table sensors. These layout parameters, based on a two-dimensional Cartesian coordinate system, define the geometric center positions of all 1024 sensing points on the operating table mattress plane. The origin is set to the upper left corner of the operating table head side, the horizontal X-axis represents the width of the bed, and the vertical Y-axis represents the length of the bed. The system extracts the pressure ratio sequence for the current time step from the body position pressure distribution data generated in the previous processing stage. This sequence contains 1024 dimensionless floating-point values. Subsequently, index alignment and data binding operations are initiated, constructing a temporary array containing 1024 structure units in memory. A loop counter from 1 to 1024 is set, and for each count value, it simultaneously serves as an index pointer pointing to a row of data in the spatial coordinate table and a value in the pressure ratio sequence. The horizontal and vertical coordinate values ​​corresponding to the current index are extracted, along with the corresponding pressure ratio value, and these three data items are encapsulated into an independent data unit. For example, when the index pointer points to sensing point number 512, the horizontal coordinate of this point in the hardware table is read as 250 mm and the vertical coordinate as 1000 mm, and the corresponding ratio value is read as 0.005000 from the pressure ratio sequence. These three values ​​are bound together to form a set of associated data containing physical location and weight information. The same reading and binding operation is performed on all remaining sensing points to ensure that each pressure ratio value can be accurately mapped to its specific location in physical space. After binding all points, the generated dataset is checked for integrity to see if there are any null values ​​or index out-of-bounds errors. If the data is complete, a set of 1024 data sets is output, each data set strictly corresponding to a pressure coordinate ratio pair. This operation transforms the abstract signal intensity distribution into spatial field data with geometric meaning, providing the necessary vector foundation for subsequent pressure center calculation. Experiments show that through this strong binding mechanism between hardware coordinates and software data, the spatial positioning matching error rate is controlled below 0.01% in the data fusion of large-scale sensor arrays, ensuring the geometric accuracy of subsequent center calculation.

[0029] S202: Based on the pressure coordinate ratio pair, the corresponding pressure ratio is called to perform weighted summation calculation on the multi-space coordinate components. The horizontal coordinate components and the vertical coordinate components are multiplied and accumulated respectively, and coordinate synthesis operation is performed to obtain the pressure center coordinates. The system receives a dataset containing 1024 pairs of compressive coordinate scales. It initializes two high-precision floating-point accumulator registers, named the transverse torque accumulator and the longitudinal torque accumulator, respectively, setting their initial values ​​to 0. It then iterates through each pair of coordinate scales in the dataset, performing a weighted summation calculation. For each pair of data, it first obtains the transverse coordinate component and the compressive scale value, performs a multiplication operation to obtain the transverse compressive torque component of that point, and adds this result to the transverse torque accumulator. Simultaneously, it obtains the longitudinal coordinate component and the same compressive scale value from the same pair of data, performs a multiplication operation to obtain the longitudinal compressive torque component of that point, and adds this result to the longitudinal torque accumulator. This operation logic is based on the centroid calculation principle, using the compressive scale as a normalization weight coefficient. For example, consider two simplified sets of data: the first set has a horizontal coordinate of 200 mm, a vertical coordinate of 500 mm, and a compression ratio of 0.3; the second set has a horizontal coordinate of 300 mm, a vertical coordinate of 600 mm, and a compression ratio of 0.7. Substituting the first set of data into the calculation logic, the horizontal moment component is 200 * 0.3 = 60, and the vertical moment component is 500 * 0.3 = 150; substituting the second set of data into the calculation logic, the horizontal moment component is 300 * 0.7 = 210, and the vertical moment component is 600 * 0.7 = 420. Summing these components, the horizontal sum is 60 + 210 = 270, and the vertical sum is 150 + 420 = 570. Since step S102 has already converted the pressure values ​​into a normalized proportional sequence, and the sum of all proportions is theoretically 1, the final value of the lateral torque accumulator is directly used as the abscissa of the synthesized pressure center, and the final value of the longitudinal torque accumulator is used as the ordinate of the pressure center. In the above example, the final pressure center coordinates are (270 mm, 570 mm). The calculated abscissa and ordinate are combined and output as the current pressure center coordinates. The advantage of this calculation logic is that, through weighted synthesis, the projected position of the patient's center of gravity can be accurately located. Even with limited sensor resolution, sub-millimeter-level positioning resolution can be obtained through interpolation. Experimental data shows that the pressure center coordinates calculated by this algorithm deviate from the physical center of gravity of the standard test dummy by only 1.2 mm, improving the positioning accuracy by 92% compared to the simple maximum value positioning method.

[0030] S203: Based on the coordinates of the pressure center, sort the coordinates of each sampling period in chronological order, compare the coordinate difference between adjacent sampling periods with a preset spatial continuity threshold, and record the structured data of the body position contact center according to the arrangement relationship of the retained coordinates in the time dimension. Based on the coordinates of the pressure center, a time-series buffer is first established to receive the coordinates of the pressure center calculated from multiple consecutive sampling periods in ascending order of timestamps. A spatial continuity threshold is set to identify and filter coordinate jumps caused by electromagnetic interference or momentary sensor malfunctions. The threshold is set based on the physiological movement limits of the patient under anesthesia. The maximum allowable movement speed parameter and the sampling time interval parameter are obtained, and a multiplication operation is performed to obtain the displacement limit value as the threshold. For example, when the maximum allowable movement speed is set to 500 mm / s and the sampling time interval is 0.01 seconds, the two are multiplied to calculate a spatial continuity threshold of 5 mm. The coordinates of the pressure center of the current sampling period and the coordinates of the pressure center of the previous sampling period are extracted from the buffer, and the coordinate difference between adjacent sampling periods is calculated. The specific calculation logic is as follows: the difference between the horizontal coordinates and the difference between the vertical coordinates are obtained respectively, and the two differences are squared respectively. The two squared results are added together and the square root is taken to obtain the Euclidean distance between the pressure centers at two time points. For example, if the current coordinates are (275 mm, 570 mm) and the previous coordinates are (270 mm, 570 mm), substituting them into the calculation, the horizontal difference is 5 and the vertical difference is 0, resulting in a distance of 5 mm. This calculated distance is then compared to the previously set threshold of 5 mm. If the calculated distance is less than or equal to the threshold, the current coordinate data is considered valid, and the coordinates are retained and appended to the structured record table in chronological order. If the calculated distance is greater than the threshold, the current data is considered an abnormal jump, and the valid coordinate values ​​from the previous time point are used to maintain and fill the current data, or the point is marked as needing repair. This judgment and recording operation is repeated for all data points in the time series, ultimately generating body contact center data containing timestamps and verified horizontal and vertical coordinates. To verify the effectiveness of the above processing, a section of test data containing simulated interference was selected for processing. The comparison and results of the data before and after processing are shown in Table 2.

[0031] Table 2. Results of Continuity Verification and Processing of Coordinates of the Center of Compression Time Index Original x-coordinate Original ordinate Adjacent displacement Determination threshold Verification results Output x-coordinate T0 270.0 570.0 - 5.0 Initial reference 270.0 T10 272.1 570.2 2.11 5.0 efficient 272.1 T20 274.0 570.5 1.92 5.0 efficient 274.0 T30 350.0 600.0 81.54 5.0 Abnormal jump 274.0 T40 275.5 570.8 1.53 5.0 efficient 275.5 As shown in Table 2, at time index T30, the original coordinates exhibited a drastic jump, with a calculated adjacent displacement of 81.54 mm, far exceeding the set threshold of 5.0 mm. After identifying this anomaly, the output at time T30 was corrected (or maintained) based on the temporal arrangement relationship, using the value of the previous valid time point T20 (274.0 mm), thus eliminating the disruption of trajectory continuity caused by sudden noise. The resulting body position contact center data not only ensured temporal causality but also significantly improved the signal-to-noise ratio. This processing mechanism reduced the false alarm rate for detecting minor patient positional slippage by 85%.

[0032] Please see Figure 4 The specific steps of S3 are as follows: S301: Obtain the coordinate points corresponding to adjacent sampling periods in the body position contact center data, perform coordinate difference operation on the coordinate points of adjacent periods under the same time index, subtract the horizontal component and the vertical component respectively and analyze the combination of directional components to generate adjacent coordinate displacement vectors. Based on the generated body position contact center data sequence, which is strictly ordered by timestamps, a traversal pointer is initialized, and the sequence is scanned item by item from the second time point to the end. Within each scan time step, the coordinate data of the current sampling period and the coordinate data of the previous sampling period are simultaneously locked and defined as the current point and reference point, respectively. The coordinate difference operation logic is then initiated, which is performed independently in the lateral and longitudinal channels. In the lateral channel, the x-coordinate value of the current point is used as the subtrahend, and the x-coordinate value of the reference point is used as the subtraction term; the subtraction operation is performed to obtain the lateral displacement component value. In the longitudinal channel, the y-coordinate value of the current point is used as the subtrahend, and the y-coordinate value of the reference point is used as the subtraction term; the subtraction operation is performed to obtain the longitudinal displacement component value. For example, when the x-coordinate of the current sampling period is 272.1 mm and the y-coordinate is 570.2 mm, while the x-coordinate of the previous sampling period was 270.0 mm and the y-coordinate was 570.0 mm, substituting 272.1 mm and 270.0 mm into the x-channel calculation yields a x-displacement component of 2.1 mm; substituting 570.2 mm and 570.0 mm into the y-channel calculation yields a y-displacement component of 0.2 mm. After calculating the two components, a direction component combination operation is performed to construct a one-dimensional array structure containing two floating-point values. The x-displacement component is placed at the beginning of the array, and the y-displacement component is placed at the second position, thus forming vector data describing the instantaneous movement characteristics within that time step. The above operation is repeated for each pair of adjacent time points in the sequence. If a missing data marker is encountered, all components of the corresponding displacement vector are forcibly set to zero. Finally, all the calculated vectors are stored in the displacement vector list according to their corresponding time index order, completing the conversion from static coordinate points to dynamic displacement vectors.

[0033] S302: Based on adjacent coordinate displacement vectors, perform direction normalization processing on each displacement vector, convert the vector components into direction angle representations, and arrange the direction angles according to the sampling time order to obtain the body position change direction sequence; The system reads the adjacent coordinate displacement vectors from the output and performs direction normalization and angle transformation on each vector element. First, it extracts the lateral and longitudinal displacement components of the current vector and calls the two-parameter arctangent function logic for calculation. This logic first calculates the ratio of the longitudinal to the lateral displacement components, then determines the quadrant position of the vector in the two-dimensional plane coordinate system based on the sign characteristics of the two components, thereby calculating the original radians or angles between -180 degrees and +180 degrees. To standardize the data, an angle mapping transformation is then performed. If the calculated angle value is less than 0 degrees, an addition operation is performed, adding it to 360 degrees, thus standardizing and mapping all angle values ​​to a closed interval between 0 and 360 degrees. For example, when using the lateral component of 2.1 mm and the longitudinal component of 0.2 mm from the previous example for calculation, since both are positive and the longitudinal component is smaller, the calculated angle of the vector is approximately 5.4 degrees. If both the lateral and longitudinal components are -2.0 mm, the original calculation result is -135 degrees, which is output as 225 degrees after mapping transformation. After completing the single-point angle transformation, the time series structure of the original data remains unchanged, and the calculated angle values ​​are sequentially filled into a new floating-point array. For stationary points with a displacement vector magnitude of zero, their angle values ​​are set to specific invalid marker values ​​(such as -1.0) to distinguish them from valid movement directions. Traversing the entire vector list, a sequence of body position change directions strictly arranged in time order is finally generated. This sequence reduces the two-dimensional spatial displacement information to a one-dimensional angle change stream, clarifying the patient's specific movement orientation at each instant.

[0034] S303: Based on the sequence of body position change directions, call the preset stable direction of body position care and perform direction difference calculation. Compare the angles of multiple directions in the sequence with the stable direction angles one by one, and count the proportion of direction items whose direction difference falls into the body position direction consistency judgment threshold to generate body position direction judgment data. Based on the sequence of body position changes, the preset stable direction parameters for the current surgical position are retrieved from the nursing configuration database. For example, the axial stable direction is set to 0 degrees. Simultaneously, the consistency threshold for body position direction is read, which is set as the maximum allowable angle range, such as 15 degrees. Each valid angle value in the sequence of body position changes is iterated, and the direction difference is calculated. The calculation logic is as follows: First, the absolute value of the difference between the current direction angle and the stable direction angle is calculated; then, a circular correction judgment is performed. If the absolute value is greater than 180 degrees, it is subtracted from 360 degrees to obtain the shortest angular distance between the two directions on the circumference. For example, when the stable direction is 0 degrees and the current direction angle is 355 degrees, the direct difference is 355 degrees. After circular correction, the actual deviation is 5 degrees. The calculated actual deviation value is compared with the judgment threshold of 15 degrees. If the actual deviation value is less than or equal to the threshold, the movement direction of the sampling point is determined to meet the stability requirements, and the consistency counter value is incremented by 1; if the deviation value is greater than the threshold, no counting is performed. After traversal, the total number of valid sampling points in the sequence is counted, and the final value of the consistency counter is divided by this total number to obtain the orientation consistency ratio. For example, if 920 out of a total of 1000 valid sampling points have a deviation of less than 15 degrees, then 920 is divided by 1000, resulting in an orientation consistency ratio of 0.92. Finally, this ratio value is encapsulated with statistical details, and the body orientation determination data is output. To verify the effectiveness of the above processing logic, a set of actual monitoring data was selected for experimental analysis, and the results are shown in Table 3.

[0035] Table 3 Results of Consistency Analysis of Body Movement Direction Parameter Indicators Set value / result Notes Preset stable direction 0.0 degrees Along the longitudinal axis of the bed Consistency determination threshold 15.0 degrees Maximum allowable deflection angle Total number of valid samples 1000 After removing stationary points Number of samples within the bias 920 In line with stable direction Number of out-of-biased samples 80 There is a tendency for lateral sliding. Calculate the consistency ratio 0.920 Judgment Result As shown in Table 3, under the condition of setting the stable direction to 0 degrees and the threshold to 15 degrees, the calculated consistency ratio is 0.920. These experimental results demonstrate that the above-mentioned direction difference and threshold determination logic can quantitatively assess the directional stability of patient positional movement. Compared to single displacement amplitude detection, this method improves the accuracy of identifying unexpected lateral slippage by approximately 28%.

[0036] Please see Figure 5 The specific steps of S4 are as follows: S401: Monitor the pressure changes of multiple pressure sensing points in the body position contact pressure concentration in adjacent sampling periods, perform difference calculation on the pressure value of the same sensing point in continuous sampling periods, decompose the pressure difference of multiple sensing points according to spatial orientation, and generate a set of directional pressure change components. To monitor pressure changes at multiple pressure sensing points in adjacent sampling cycles during body positioning contact, a communication connection is first established with a high-density piezoelectric thin-film sensor array. This array, distributed across the operating table contact surface, contains 1024 independent pressure sensing points. Pressure values ​​at each sensing point are read in real-time at a sampling frequency of 50 Hz and stored in a circular buffer. Pressure change analysis logic is then initiated, traversing each pressure sensing point in the array and reading the real-time pressure value corresponding to the timestamp of the current sampling cycle (denoted as the current pressure value) and the historical pressure value corresponding to the previous sampling cycle (denoted as the previous pressure value). A differential pressure calculation is performed, subtracting the previous pressure value from the current pressure value to extract the instantaneous pressure fluctuation at that point. For example, for sensing point ID-512, with a current pressure value of 45.5 mmHg and a previous pressure value of 42.0 mmHg, the subtraction operation yields a pressure difference of 3.5 mmHg. Subsequently, to accurately identify the physical migration path of pressure, an orthogonal vector decomposition operation is performed based on the spatial topological relationship of the sensing point array. The system pre-defines two orthogonal bases: a "longitudinal-traction safety axis" and a "lateral-shear risk axis." It acquires the pressure difference data of the sensing point at its lateral and longitudinal adjacent sensing points, calculates the local pressure change gradient at that point using the central difference method, and normalizes this gradient to obtain the lateral and longitudinal gradient coefficients representing the pressure migration trend. The innovation of this step lies in resolving the scalar pressure fluctuation into vector components: the calculated pressure difference is multiplied by the corresponding lateral and longitudinal gradient coefficients for that sensing point. Continuing with the previous example, if the gradient calculation shows that the pressure change at the ID-512 sensing point tends to spread to the right (laterally), the calculated normalized lateral gradient coefficient is 0.2 and the longitudinal gradient coefficient is 0.9. Multiplying 3.5 mmHg with 0.2 and 0.9 respectively, the calculated lateral pressure change component (characterizing potential sideslip force) is 0.7 mmHg and the longitudinal pressure change component (characterizing compliant traction force) is 3.15 mmHg. Repeating the above difference and decomposition operations for all active sensing points, the final result is a set of directional pressure change components containing the dynamic mechanical vector characteristics of the entire array.

[0037] S402: Based on the set of directional compressive variation components, perform equivalent energy conversion on the multi-directional components, map the numerical values ​​of the compressive variation components to directional energy values, and perform proportional calculation on the multi-directional energy within the same sampling period to obtain the directional energy ratio; The output set of directional pressure change components is read. To extract core indicators characterizing body positional stability from complex pressure fluctuations, signal power spectrum analysis is introduced. Equivalent energy conversion is performed on the lateral and longitudinal pressure change components of each sensing point in the set. This conversion maps the amplitude of pressure change to a non-negative "deformation energy" metric through squaring, effectively amplifying the influence of large signals and eliminating the interference of directional sign on total statistical analysis. The lateral and longitudinal component values ​​of each sensing point are multiplied by themselves (i.e., squared) to obtain the instantaneous pressure energy values ​​at that point in the lateral and longitudinal directions. For example, using the aforementioned lateral component 0.7 and longitudinal component 3.15, 0.7 × 0.7 = 0.49, and 3.15 × 3.15 = 9.9225. After completing the single-point conversion, a global energy accumulation operation is performed. The lateral energy values ​​of all sensing points within the same sampling period are summed to obtain the total lateral pressure energy (i.e., interference / risk energy); the longitudinal energy values ​​are summed to obtain the total longitudinal pressure energy (i.e., effective / compliant energy). Then, the total lateral and longitudinal pressure energy are added together to obtain the total global pressure energy. Based on this, a directional energy ratio calculation is performed to solve for the "mechanical signal-to-noise ratio": the total energy value corresponding to the pre-defined dominant stable direction (usually longitudinal) in surgical care is selected as the numerator, and the total global pressure energy is selected as the denominator, and a division operation is performed. For example, if the total lateral pressure energy at a certain moment is 50.0 and the total longitudinal pressure energy is 450.0, the sum of the two is 500.0. 450.0 / 500.0 = 0.9. This ratio is a dimensionless value between 0 and 1. Its physical meaning is that the closer the value is to 1, the more purely the current positional change follows the pre-set care path (such as longitudinal traction), while a lower value indicates the accumulation of unintended lateral shear or torsional energy at the contact surface. This ratio is calculated continuously over time, ultimately outputting a directional energy ratio sequence that reflects the stability of the mechanical distribution at the contact surface.

[0038] S403: Based on the directional energy ratio, a dynamic energy judgment threshold is generated based on the body position contact center data. If the directional energy ratio does not reach the dynamic energy judgment threshold and the body position direction judgment data indicates a consistent direction, an initial judgment result for the compliance of body position care is generated. This step innovatively constructs a "geometric-physical dual interlock" judgment logic. First, it retrieves the calculated directional energy ratio (e.g., 0.9) from the database, while simultaneously calling the generated body position orientation judgment data (including the directional consistency ratio) and the displacement velocity parameters from the body position contact center data. An adaptive dynamic energy judgment threshold model is constructed: considering that the tolerance of the contact surface to stray energy varies under different body position movement speeds, the system adjusts the energy threshold in real time based on the displacement velocity. When a relatively fast center displacement velocity is detected (i.e., active body position adjustment is underway), the system appropriately relaxes the threshold (e.g., set to 0.80) to accommodate operational fluctuations; when the displacement velocity approaches zero (i.e., during the static maintenance period), the system automatically tightens the threshold (e.g., set to 0.90) to sensitively capture minute lateral slippage trends. Dual interlock verification is performed: the first verification is a physical energy verification, determining whether the current directional energy ratio reaches the dynamic energy judgment threshold (e.g., whether it is greater than 0.90); the second verification is a geometric direction verification, determining whether the directional consistency ratio meets the preset standard. Only when the "geometric orientation is correct" but the "physical energy distribution is abnormal" (i.e., the directional energy ratio does not meet the standard) can the system penetrate appearances and accurately identify the "potential lateral slip risk." This situation usually corresponds to the naked eye seemingly indicating the correct direction of movement, but the patient's skin contact surface is actually bearing a huge lateral shear force. If both the directional energy ratio and the directional consistency ratio meet the standards, it is judged as "compliant and stable." Finally, the output includes a compliance status indicator (e.g., 0 represents compliance, 1 represents risk, and 2 represents violation) and a corresponding risk description, representing the initial judgment result of the positional care compliance. To verify the accuracy of this compliance judgment logic, positional monitoring data during surgery was collected in an actual clinical environment for testing. The specific experimental data and judgment results are shown in Table 4.

[0039] Table 4. Logic Verification Data Table for Compliance Judgment of Body Positioning Care Parameter Items Experimental Group A (Normal Traction) Experimental Group B (Hidden Sideslip) Parameter Description Directional energy ratio 0.94 0.62 Force distribution concentration (mechanical signal-to-noise ratio) Dynamic energy threshold 0.80 (due to displacement relaxation) 0.90 (due to static tightening) Adaptive reference based on displacement velocity Directional consistency ratio 0.95 0.88 Geometric orientation conformity Consistency threshold 0.90 0.90 Geometric determination baseline Decision logic output Compliance and stability Potential sideslip risk Double interlock verification result Clinical reality Stablize Subcutaneous shearing occurred Manual review results As shown in Table 4, in experimental group A, both the directional energy ratio (0.94) and the directional consistency ratio (0.95) were higher than the set thresholds, correctly identifying it as "compliant and stable." However, in experimental group B, although the directional consistency ratio (0.88) was close to the threshold (geometrically appearing close to compliance), the directional energy ratio (0.62) was significantly lower than the threshold of 0.85, indicating that the proportion of lateral shear energy was too high. The system successfully addressed the hidden risk through the energy dimension, identifying a "potential sideslip risk." These experimental results demonstrate that by introducing the correlation and interlocking judgment of the compressive energy distribution, this technical solution can effectively identify stress instability that is difficult to detect solely by geometric displacement. Compared to traditional single-parameter monitoring methods, its comprehensive detection rate for abnormal body positions is improved.

[0040] Please see Figure 6 The specific steps of S5 are as follows: S501: Based on the initial judgment result of the compliance of body position care, obtain the judgment mark data corresponding to multiple cycles within the continuous sampling period, perform sequence alignment on the judgment marks of adjacent sampling periods, and perform change comparison on the values ​​of adjacent marks to generate a compliance judgment change sequence. Based on the initial judgment result of body position care compliance, the system first accesses the status register in memory and reads the initial judgment result stream of body position care compliance output from the previous step S403 in ascending time order. A read pointer is set to extract a data segment containing the judgment markers of the most recent 100 sampling points. Sequence alignment and differential comparison logic is initiated, and a change marker array of the same length as the number of sampling points is initialized. This data segment is traversed, starting from the second sampling point, locking the time index t of the current sampling period and the time index t−1 of the previous sampling period. A numerical equivalence judgment operation is performed, comparing the item to be compared with the reference item. If the two values ​​are completely identical, it is determined that no state transition has occurred within this time step, and the change marker is assigned a value of 0; if the two values ​​are inconsistent, it is determined that a state transition has occurred, and the change marker is assigned a value of 1. This step essentially performs a "time-domain differentiation" on the judgment result, aiming to capture the edge transition characteristics of the state signal and provide basic data for subsequent elimination of high-frequency jitter noise. Finally, the compliance judgment change sequence is output in chronological order.

[0041] S502: Based on the compliance judgment change sequence, collect the sampling period index corresponding to the preset time window, perform consistency screening on the change sequence falling within the time window, continuously aggregate the consistency judgment marks and perform state stability judgment to obtain the compliance state stable mark sequence. Based on the compliance judgment change sequence, a time window length parameter is defined. According to the physiological inertia characteristics of human body position adjustment, the window length is set to 50 sampling periods (corresponding to 1 second). A sliding window algorithm is applied, sliding this time window across the compliance judgment change sequence. Consistency filtering and stability calculation are performed. This calculation logic first counts the number of change markers marked as 0 within the window, defining them as stable point counts. Then, the stable point count is divided by the total window length (50), and the division operation is performed to obtain the window stability coefficient. Simultaneously, a stability judgment threshold (e.g., 0.90) is introduced. This logic constitutes a digital filter: If the coefficient is greater than or equal to 0.90, it indicates that the judgment state remains highly consistent within this time period. In this case, a mode extraction operation is performed on the original judgment markers within the window, selecting the marker value with the highest frequency as the representative stable state of the window. If the coefficient is less than 0.90, the window is determined to be in a "transitional oscillation period," and a state preservation strategy is adopted, extending the stable state of the previous window to the current window, or marking it as an undetermined state. This continuous aggregation mechanism effectively filters out single-point false alarms caused by instantaneous sensor interference, generating a stable mark sequence of compliance states.

[0042] S503: Call the compliance status stable marker sequence, use the majority voting rule to perform status adjustment on the multi-marker sequence, count the occurrence frequency of the multi-state markers and take the one with the best frequency as the adjusted status result, and establish an association mapping with the corresponding time window index to generate the body position nursing compliance monitoring result; The compliance status stability marker sequence is invoked, and a status adjustment judgment strategy based on window period synchronization is applied. Since high-frequency noise filtering and stability aggregation have been completed through a 1-second time window, the status confirmation threshold is set to one complete stable time window (i.e., lasting 1 second). The sequence is traversed, and when the status marker is detected to change from "Compliance 0" to "Risk 1" and the stability coefficient of the window meets the requirements, the risk status is immediately confirmed as valid, and the output status is locked as "Risk Established"; if the subsequent window status falls back to 0, the risk lock is released. Compared with multi-window cumulative confirmation, this strategy controls the response delay to within 1.0 second (i.e., one window duration) while ensuring signal stability. After the status adjustment is completed, association mapping and merging processing are performed. Each confirmed status value is bound to its covered time start and end index (e.g., from start timestamp to end timestamp). Subsequently, adjacent time blocks are checked. If the status values ​​of adjacent time blocks are the same, a merging operation is performed to merge the two time blocks into a long-term event and update the end timestamp. For example, if time block A (indexes 100 to 150) is in a risky state, and time block B (indexes 151 to 200) is also in a risky state, they are merged into a single event (indexes 100 to 200, risky state). Finally, the merged event list is formatted and output to generate a positional care compliance monitoring result containing specific time periods, state types, and durations. To verify the effectiveness of the above multi-level judgment and state stability processing logic, a set of surgical positional monitoring data containing simulated interference was selected for verification. The specific data and processing results are shown in Table 5.

[0043] Table 5 Compliance Status Stability Processing Verification Data Table Verification parameters Original wave sequence (partial) Stable sequences after processing Notes Time window length 50 sampling points 50 sampling points Corresponding to 1 second duration Number of transitions from the original state 12 times 1 time Statistics within 10 seconds Stability determination threshold 0.90 0.90 Filtering high-frequency noise Number of false alarms 4 times 0 times Short-term interference was eliminated Real risk response delay 0.02 seconds 1.0 second Includes single-window confirmation cycle Final accuracy rate 82.5 % 98.0 % Comparison with expert annotations As shown in Table 5, when processing a data segment containing 12 initial state transitions, by setting a stability threshold of 0.90 and performing continuous aggregation, the number of transitions was successfully reduced to 1 (reflecting the true positional change), and all four false alarms caused by instantaneous vibrations were eliminated. Although the response time for the true risk increased by approximately 1.0 second due to the window aggregation logic, the final accuracy rate significantly improved from 82.5% to 98.0%. These experimental results demonstrate that this logical deduction effectively solves the signal jitter problem caused by high-sensitivity sensors, ensuring the clinical reliability of positional care monitoring results.

[0044] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for detecting surgical patient positioning compliance, the method comprising: Includes the following steps: S1: Collect electrical signals output by multiple pressure sensing points inside the operating table during anesthesia, convert them into pressure distribution data of body position contact, and perform proportional conversion between the pressure value and the sum of pressure values ​​within the same sampling period to generate body position pressure distribution data; S2: Based on the body position pressure distribution data, call the spatial coordinates of multiple pressure sensing points, combine them with the corresponding pressure ratio to calculate the pressure center coordinates by weighted summation and sort them to generate body position contact center data; S3: Obtain the coordinate points corresponding to adjacent sampling periods in the body position contact center data, calculate the displacement vector combination between adjacent coordinate points as a body position change direction sequence, compare and analyze the consistency of the body position change direction with the preset body position nursing stable direction, and generate body position direction determination data. S4: Monitor the pressure changes of multiple pressure sensing points in the body position contact pressure concentration in adjacent sampling cycles, decompose the direction according to the spatial orientation of the sensing points, convert and calculate the directional energy ratio, set the energy ratio threshold based on the body position contact center data and compare it with the directional energy ratio, and perform logical operations in combination with the body position orientation determination data to generate an initial judgment result of body position nursing compliance.

2. The surgical patient position care compliance detection method of claim 1, wherein, The body position pressure distribution data includes the pressure ratio, sampling period index identifier, and pressure ratio normalization sequence; the body position contact center data includes weighted pressure center coordinates, center coordinate sorting sequence, and center migration reference index; the body position orientation determination data includes displacement vector orientation encoding, orientation consistency determination mark, and orientation stability state index; and the initial body position nursing compliance determination result includes orientation energy ratio determination mark, orientation association state label, and single-cycle compliance determination identifier.

3. The method for detecting compliance of surgical patient positioning care according to claim 1, characterized in that, The specific steps of S1 are as follows: S101: Collects electrical signals output by multiple pressure sensing points inside the operating table during anesthesia, performs sampling period alignment based on the timestamps of the signals from multiple sensing points, and after alignment, converts the electrical signal amplitude to pressure values ​​according to the calibration mapping relationship to generate a body position contact pressure set. S102: Based on the pressure set of body position contact, perform summation analysis on the total pressure of multiple pressure values ​​within the same sampling period, call the pressure values ​​of multiple sensing points and perform proportional conversion with the total pressure of the period, and arrange the proportional results according to the sensing point index order to generate a pressure ratio sequence. S103: Based on the pressure ratio sequence, perform mapping and arrangement according to the corresponding positions of the sensing points in the operating table spatial coordinates, fill the ratio values ​​into the spatial grid structure, and integrate the results of multiple sampling periods in time series to generate body position pressure distribution data.

4. The method for detecting compliance of surgical patient positioning care according to claim 1, characterized in that, The specific steps of S2 are as follows: S201: Based on the body position pressure distribution data, call the spatial coordinates corresponding to multiple pressure sensing points, perform index alignment and data binding on the spatial coordinates of multiple sensing points and the corresponding pressure ratio, associate the ratio values ​​with the coordinate parameter items and generate pressure coordinate ratio pairs; S202: Based on the pressure coordinate ratio pair, perform weighted summation calculation on the multi-space coordinate components according to the corresponding pressure ratio, perform proportional multiplication and accumulation on the horizontal coordinate component and the vertical coordinate component respectively, and perform coordinate synthesis operation to obtain the pressure center coordinates; S203: Based on the coordinates of the pressure center, sort the coordinates corresponding to each sampling period in chronological order, compare the coordinate difference between adjacent sampling periods with a preset spatial continuity threshold, and record the structured data of the body position contact center according to the arrangement relationship of the retained coordinates in the time dimension.

5. The method for detecting compliance of surgical patient positioning care according to claim 4, characterized in that, The spatial continuity threshold is determined by statistically analyzing the differences between the coordinate components of the pressure center coordinates on the time axis within a continuous sampling period, collecting the differences between the horizontal and vertical coordinate components of adjacent periods and performing absolute value calculations, aggregating all the difference results, and calculating the mean of the differences and a preset standard deviation of 3.

6. The method for detecting compliance of surgical patient positioning care according to claim 1, characterized in that, The specific steps for S3 are as follows: S301: Obtain the coordinate points corresponding to adjacent sampling periods in the body position contact center data, perform coordinate difference operation on the coordinate points of adjacent periods under the same time index, subtract the horizontal component and the vertical component respectively and analyze the combination of directional components to generate adjacent coordinate displacement vectors. S302: Based on the adjacent coordinate displacement vectors, perform direction normalization processing on each displacement vector, convert the vector components into direction angle representations, and arrange the direction angles according to the sampling time order to obtain the body position change direction sequence; S303: Based on the sequence of body position change directions, call the preset stable direction for body position care and perform direction difference calculation. Compare the angles of multiple directions in the sequence with the stable direction angles one by one, and count the proportion of direction items whose direction difference falls into the body position direction consistency judgment threshold to generate body position direction judgment data.

7. The method for detecting compliance of surgical patient positioning care according to claim 6, characterized in that, The body position orientation consistency determination threshold is determined by obtaining the difference between the direction angle of multiple sampling periods and the stable direction angle and performing absolute value calculation, forming a direction difference set of all direction differences, calculating the mean of the set and simultaneously extracting the direction difference dispersion parameter, and calculating the sum of the mean and dispersion.

8. The method for detecting compliance of surgical patient positioning care according to claim 1, characterized in that, The specific steps of S4 are as follows: S401: Monitor the pressure changes of multiple pressure sensing points in the body position contact pressure concentration in adjacent sampling periods, perform difference calculation on the pressure value of the same sensing point in continuous sampling periods, decompose the pressure difference of multiple sensing points according to spatial orientation, and generate a set of directional pressure change components. S402: Based on the set of directional pressure change components, perform equivalent energy conversion on the multi-directional components, map the numerical values ​​of the pressure change components to directional energy values, and perform proportional calculation on the multi-directional energy within the same sampling period to obtain the directional energy ratio; S403: Based on the directional energy ratio, a dynamic energy judgment threshold is generated based on the body position contact center data. If the directional energy ratio does not reach the dynamic energy judgment threshold and the body position direction judgment data indicates a consistent direction, an initial judgment result for body position nursing compliance is generated.

9. The method for detecting compliance of surgical patient positioning care according to claim 1, characterized in that, The method further includes: S5: Based on the initial judgment result of the body position care compliance, compare the change relationship between the initial judgment data of body position care compliance within the continuous sampling period, and perform consistency screening and status adjustment judgment within a preset time window to generate body position care compliance monitoring results; The compliance monitoring results for body positioning care include a time window consistency status indicator, a status adjustment judgment label, and a continuous cycle compliance monitoring mark.

10. The method for detecting compliance of surgical patient positioning care according to claim 9, characterized in that, The specific steps of S5 are as follows: S501: Based on the initial judgment result of the body position care compliance, obtain the judgment mark data corresponding to multiple cycles within the continuous sampling period, perform sequence alignment on the judgment marks of adjacent sampling periods, and perform change comparison on the values ​​of adjacent marks to generate a compliance judgment change sequence. S502: Based on the compliance judgment change sequence, collect the sampling period index corresponding to the preset time window, perform consistency screening on the change sequence falling within the time window, continuously aggregate the consistency judgment marks and perform state stability judgment to obtain the compliance state stable mark sequence. S503: Call the compliance status stable marker sequence, perform status adjustment on the multi-marker sequence using the majority voting rule, count the occurrence frequency of the multi-state markers and take the one with the best frequency as the adjusted status result, and establish an association mapping with the corresponding time window index to generate the body position nursing compliance monitoring result.

Images