Home textile recommendation method and system based on child health data monitoring and analysis

Abstract

The invention relates to the technical field of biomechanical measurement and analysis, in particular to a home textile recommendation method and system based on child health data monitoring and analysis. The method comprises the following steps: collecting an original point cloud coordinate set of a sleeping posture of a child; extracting a cervical curvature parameter group according to the extracted cervical curvature parameter group of the skeleton key point spatial distribution diagram; performing cervical vertebra and body position analysis according to the cervical vertebra curvature parameter group to obtain cervical vertebra curvature-body position holographic data; collecting a static pressure baseline data set according to the cervical curvature-body position holographic data; performing body position-pressure correlation analysis according to the static pressure baseline data set to obtain a body position-pressure mapping relation table; and generating a pressure distribution index map according to the body position-pressure mapping relation table. According to the method, the physical sign data, the pressure distribution and the physiological curvature change of the children in the natural sleep state are comprehensively collected and analyzed, so that high-precision, whole-process and personalized healthy home textile recommendation is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of biomechanical measurement and analysis, and in particular to a home textile recommendation method and system based on children's health data monitoring and analysis. Background Art

[0002] Existing home textile products often use uniform sizes or crude age categories, ignoring the individual differences in children's growth and development. Recommendations often rely on parents' subjective experience, questionnaires, or single, static parameters like height and weight, making it difficult to achieve dynamic, detailed analysis of children's physical signs and health conditions, and accurate matching.

[0003] Most home textile recommendations still use two-dimensional measurements or plane data, or even rely solely on visual observation. They lack high-precision data collection methods such as three-dimensional scanning and bone modeling, and are unable to fully reflect children's actual sleeping position, cervical curvature and related health indicators, resulting in insufficient accuracy and scientificity in recommendations.

[0004] Existing methods generally only perform single-point, static pressure measurements, which cannot dynamically reflect the pressure migration and force changes caused by children's changes in body position during sleep. They are prone to missing health risks such as excessive local pressure and insufficient support, and cannot provide children with a scientific and reasonable basis for home textile selection.

[0005] In summary, existing technologies lack personalized, dynamic, and refined health data support for home textile recommendations, making it difficult to achieve scientific and accurate matching, a problem that needs to be urgently addressed. Summary of the Invention

[0006] Based on this, it is necessary to provide a home textile recommendation method and system based on children's health data monitoring and analysis to solve at least one of the above technical problems.

[0007] To achieve the above objectives, a home textile recommendation method based on children's health data monitoring and analysis includes the following steps: Step S1: collecting the original point cloud coordinate set of the child's sleeping posture; extracting the cervical curvature parameter group based on the cervical curvature parameter group of the extracted skeletal key point spatial distribution map; performing cervical vertebra and body position analysis based on the cervical curvature parameter group to obtain cervical curvature-body position holographic data; Step S2: collecting a static pressure baseline data set based on the cervical vertebra curvature-posture holographic data; performing posture-pressure correlation analysis based on the static pressure baseline data set to obtain a posture-pressure mapping relationship table; generating a pressure distribution index map based on the posture-pressure mapping relationship table; Step S3: quantifying the curvature-support effect according to the pressure distribution index map to obtain a curvature deviation index set; analyzing the pressure-curvature response relationship between the curvature deviation index set and the pressure distribution index map to obtain a curvature recovery adaptation factor; Step S4: Obtain pillow material performance data; calculate a biomechanical support index table based on the pillow material performance data and the curvature recovery adaptation factor; make personalized pillow support recommendations based on the biomechanical support index table to obtain a recommended support parameter set.

[0008] The present invention collects the original point cloud coordinate set of the child's sleeping position, extracts the cervical curvature parameter group, and performs cervical spine and body position analysis to obtain cervical curvature-body position holographic data. Its beneficial effect is that: through multi-angle optical scanning, three-dimensional point cloud data of the child in a natural sleeping position is obtained, avoiding the subjectivity and limitations of traditional two-dimensional measurement, and achieving objective, comprehensive, and high-precision measurement of the child's head and neck morphology. Extracting bone key points and constructing a cervical curvature parameter group can quantify the physiological curvature of the cervical spine, providing a basis for subsequent curvature analysis and evaluation. Cervical spine and body position analysis can establish the relationship between cervical curvature and different sleeping positions, which helps to understand the impact of different body positions on cervical health. The cervical curvature-body position holographic data finally generated comprehensively and accurately reflects the child's cervical health status and sleeping habits, providing a reliable data basis for subsequent stress analysis and personalized recommendations. Based on the holographic cervical curvature-position data, a static pressure baseline dataset is collected, and a position-pressure correlation analysis is performed to generate a pressure distribution index map. This beneficial effect is that, through the pressure sensor array, the pressure distribution between the child's head and neck and the bedding is monitored in real time, enabling a dynamic and detailed analysis of the force situation, avoiding the limitations of traditional methods that only perform single-point static measurements. The static pressure baseline dataset provides pressure data for different contact areas, which helps to assess the force situation in different areas. The position-pressure correlation analysis establishes a mapping relationship between position changes and pressure distribution, which can understand the changing patterns of pressure distribution under different sleeping positions and identify potential pressure concentration points and areas of insufficient pressure. The generated pressure distribution index map visualizes the pressure distribution information and intuitively displays the pressure conditions in each area. It provides key input for subsequent quantification of curvature-support effects and personalized recommendations, and facilitates a comprehensive assessment of children's sleeping posture and force situation. The curvature-support effect is quantified based on the pressure distribution index map, and the pressure-curvature response relationship is analyzed to generate a curvature recovery adaptation factor. This has the following benefits: By establishing age-stage criteria and an ideal curvature model, it enables an objective assessment of cervical curvature recovery in children, avoiding subjective bias. The measurement of dynamic curvature change trajectories can reflect the actual impact of pillows / bedding on cervical curvature recovery. The calculation of a curvature deviation index quantifies the difference between actual cervical curvature and ideal curvature, helping to evaluate support effectiveness. The analysis of the pressure-curvature response relationship reveals the intrinsic connection between pressure distribution and cervical curvature changes, providing a scientific basis for understanding the impact of pillows / bedding on cervical health. The resulting curvature recovery adaptation factor comprehensively considers curvature recovery, pressure-curvature response characteristics, and age-stage characteristics. It quantifies the degree of match between pillows / bedding and children's cervical health needs, providing a key basis for personalized recommendations, helping to select the most suitable pillows / bedding for children and promote healthy cervical development.By obtaining pillow material performance data and calculating a biomechanical support index, personalized pillow support recommendations are made, resulting in a recommended support parameter set. This beneficial effect lies in: By establishing a pillow material performance database, a comprehensive understanding of the mechanical properties of various materials is provided, providing a foundation for appropriate material selection. The biomechanical support index calculation comprehensively considers factors such as material properties, head and neck biomechanical parameters, support force, sleeping position, and stability, quantifying the support effect of different pillows on children's cervical spine and providing a scientific basis for personalized recommendations. Pillow morphology parameter optimization allows for the design of optimal pillow height, curvature, edge transition angle, and support zone width based on individual child differences, improving pillow comfort and support effectiveness. Material combination design allows for the selection of appropriate materials based on the support needs of different areas, achieving differentiated support and enhancing overall pillow performance. The resulting recommended support parameter set includes pillow height ranges, material types and combination ratios, hardness zoning, morphology design parameters, and usage recommendations, providing users with a comprehensive, personalized, and actionable pillow selection solution that helps improve children's sleep quality and promote cervical spine health. Therefore, the present invention provides a home textile recommendation method based on children's health data monitoring and analysis. By introducing the process of three-dimensional health sign stereo modeling, zoned pressure dynamic mapping, cervical curvature recovery modeling and multi-parameter intelligent recommendation, the physical sign data, pressure distribution and physiological curvature changes of children in natural sleep state are comprehensively collected and analyzed, thereby achieving high-precision, full-process and personalized healthy home textile recommendations, effectively solving the prominent problems in the existing technology such as insufficient personalization, rough data collection and unscientific pressure analysis, and improving the health adaptability and scientificity of children's home textile products.

[0009] Preferably, the present invention further provides a home textile recommendation system based on child health data monitoring and analysis, which is used to execute the above-mentioned home textile recommendation method based on child health data monitoring and analysis. The home textile recommendation system based on child health data monitoring and analysis includes: The physical sign three-dimensional modeling module is used to collect the original point cloud coordinate set of the child's sleeping posture; extract the cervical curvature parameter group based on the cervical curvature parameter group of the extracted skeletal key point spatial distribution map; perform cervical vertebra and body position analysis based on the cervical curvature parameter group to obtain cervical curvature-body position holographic data; The partitioned pressure mapping module is used to collect a static pressure baseline data set based on the cervical curvature-posture holographic data; perform posture-pressure correlation analysis based on the static pressure baseline data set to obtain a posture-pressure mapping relationship table; and generate a pressure distribution index map based on the posture-pressure mapping relationship table; The curvature recovery modeling module is used to quantify the curvature-support effect based on the pressure distribution index map to obtain the curvature deviation index set; the pressure-curvature response relationship between the curvature deviation index set and the pressure distribution index map is analyzed to obtain the curvature recovery adaptation factor; The support matching reasoning module is used to obtain pillow material performance data; calculate the biomechanical support index table based on the pillow material performance data and the curvature recovery adaptation factor; and make personalized pillow support recommendations based on the biomechanical support index table to obtain a recommended support parameter set.

[0010] The home textile recommendation system, based on child health data monitoring and analysis, provides a comprehensive solution from data collection and analysis to recommendations, utilizing modules such as 3D physical sign modeling, zonal pressure mapping, curvature recovery modeling, and support matching reasoning. Focusing on child health data, the system conducts multi-dimensional health assessments and scientific support matching, enabling personalized and precise home textile product recommendations. This system improves children's sleep quality and promotes healthy cervical spine development. Highly integrated, automated, and scalable, it ultimately provides users with comprehensive, easy-to-understand recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A flowchart of the steps of a home textile recommendation method based on children's health data monitoring and analysis; Figure 2 Schematic diagram of the detailed implementation steps of step S1 in the present invention.

[0012] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0013] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. It is obvious that the embodiments described are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts are within the scope of protection of the present invention.

[0014] In addition, the accompanying drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor and / or microcontroller approaches.

[0015] It should be understood that although the terms "first," "second," and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are used solely to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of the exemplary embodiments. The term "and / or" as used herein includes any and all combinations of one or more of the listed associated items.

[0016] In the embodiment of the present invention, reference Figure 1 FIG. 1 is a flow chart of the steps of the home textile recommendation method based on child health data monitoring and analysis of the present invention. In this example, the home textile recommendation method based on child health data monitoring and analysis includes the following steps: Step S1: collecting the original point cloud coordinate set of the child's sleeping posture; extracting the cervical curvature parameter group based on the cervical curvature parameter group of the extracted skeletal key point spatial distribution map; performing cervical vertebra and body position analysis based on the cervical curvature parameter group to obtain cervical curvature-body position holographic data; In this embodiment of the present invention, a raw point cloud coordinate set of a child's sleeping posture is collected, a cervical curvature parameter set is extracted, and cervical and body position analysis is performed to ultimately generate cervical curvature-position holographic data. First, a 3D optical scanner is used to scan the child's natural sleeping position from multiple angles to obtain a raw point cloud coordinate set. After filtering and head and neck extraction, a pre-trained convolutional neural network (CNN) model, combined with manual correction, automatically identifies and labels key skeletal points, generating a spatial distribution map of skeletal key points. Based on these key points, the cervical centerline trajectory is extracted, and a three-plane curvature distribution map, key segment angle values, curve morphological feature sets, curvature standard deviation values, cervical stress distribution maps, and curvature dynamic change characteristics are calculated and integrated to generate a cervical curvature parameter set. Finally, combining the skeletal key point spatial distribution map and the cervical curvature parameter set, head-neck ratio and multi-dimensional posture angle measurement and analysis are performed. All parameters are fused and standardized, and grouped by age to establish age-related reference standards, generating complete cervical curvature-position holographic data.

[0017] Step S2: collecting a static pressure baseline data set based on the cervical vertebra curvature-posture holographic data; performing posture-pressure correlation analysis based on the static pressure baseline data set to obtain a posture-pressure mapping relationship table; generating a pressure distribution index map based on the posture-pressure mapping relationship table; In an embodiment of the present invention, a static pressure baseline data set is collected based on the cervical curvature-posture holographic data, and a posture-pressure correlation analysis is performed to ultimately generate a pressure distribution index map. First, based on the child's sleeping position in the cervical curvature-posture holographic data, the child is placed on a pressure sensor test bed, and the pressure distribution of the head and neck in contact with the bedding is measured to obtain a contact area division map. Then, in different postures, a static pressure baseline data set is collected, and the pressure value, contact area, and pressure center position are recorded. At the same time, the dynamic pressure change sequence is monitored, and the pressure changes during the posture change process are recorded. The pressure uniformity is calculated to obtain a pressure uniformity distribution matrix. Next, the dynamic pressure change sequence is time-synchronized and spatially aligned with the cervical curvature-posture holographic data to establish a posture-pressure mapping relationship. The characteristic patterns of pressure distribution in different postures are analyzed, the pressure transfer trajectory and pressure response delay are calculated, and a posture-pressure mapping relationship table is constructed. Finally, the pressure peak is identified, the correspondence between the pressure peak and the key anatomical structure of the cervical spine is analyzed, and a pressure peak distribution characteristic diagram is generated. The pressure uniformity, posture-pressure mapping relationship and pressure peak distribution are comprehensively considered to calculate the comprehensive pressure distribution index and generate a pressure distribution index map.

[0018] Step S3: quantifying the curvature-support effect according to the pressure distribution index map to obtain a curvature deviation index set; analyzing the pressure-curvature response relationship between the curvature deviation index set and the pressure distribution index map to obtain a curvature recovery adaptation factor; In an embodiment of the present invention, the curvature-support effect is quantified based on the pressure distribution index map, and the pressure-curvature response relationship is analyzed, and finally the curvature recovery adaptation factor is obtained. First, based on the pressure distribution index map and the cervical curvature-posture holographic data, a reference standard for cervical development stages is established according to age groups, and an ideal curvature model is constructed as a benchmark for evaluating the recovery of cervical curvature. Next, let the children use the pillow / bedding to be tested, record the changes in cervical curvature in real time, and measure the corresponding pressure distribution changes to obtain the dynamic curvature change trajectory. Through numerical comparison, the deviation value between the actual cervical curvature and the ideal curvature model is calculated to generate a curvature deviation index set. Then, the pressure and cervical curvature data are collected synchronously at high frequency, and time window segmentation processing is performed to analyze the influence of pressure on cervical curvature, establish a pressure-curvature mapping function, and perform time delay characteristic analysis to generate a pressure-curvature response matrix. Finally, based on the curvature deviation index set and the pressure-curvature response matrix, a comprehensive scoring system for cervical curvature recovery was established, and a curvature recovery score table was calculated. Taking into account the curvature recovery score, pressure-curvature response characteristics, and age stage characteristics, a multidimensional adaptation factor calculation model was constructed to generate the curvature recovery adaptation factor.

[0019] Step S4: Obtain pillow material performance data; calculate a biomechanical support index table based on the pillow material performance data and the curvature recovery adaptation factor; perform personalized pillow support recommendations based on the biomechanical support index table to obtain a recommended support parameter set; In an embodiment of the present invention, pillow material performance data is obtained, a biomechanical support index table is calculated, and personalized pillow support recommendations are made, ultimately generating a recommended support parameter set. First, a pillow material performance database is constructed, and mechanical property parameters of different materials are collected. Material rebound response analysis is performed to generate a material rebound property evaluation table. Next, biomechanical parameters of the child's head and neck are collected to calculate a regional ideal support force table. Next, a material support force curve set is measured, a regional support matching index is calculated, and a posture difference analysis is performed to obtain a posture support correction factor. Combined with sleep posture ratio monitoring and analysis, a sleep posture time percentage table is generated to evaluate the support stability coefficient. Finally, a comprehensive biomechanical support index is calculated to generate a biomechanical support index table. Based on the biomechanical support index table and the curvature recovery adaptation factor, pillow morphological parameters are optimized to obtain a morphological parameter combination. Based on the morphological parameter combination and the biomechanical support index table, pillow material combinations are determined to obtain a material combination scheme, and a personalized adjustment parameter table is calculated. Finally, the morphological parameter combination, material combination scheme, and personalized adjustment parameter table are integrated to generate a recommended support parameter set.

[0020] As an example of the present invention, refer to Figure 2 As shown, in this example, step S1 includes: Step S11: capturing the child's natural sleeping posture through a high-speed camera and performing multi-angle optical scanning on the child's natural sleeping posture to obtain an original point cloud coordinate set; In an embodiment of the present invention, to obtain a child's natural sleeping position, it is necessary to ensure that the child is relaxed and avoids excessive tension or discomfort. The operator should guide the child to adopt their usual sleeping position and record the sleeping position. To reduce interference from environmental factors, the scanning environment should maintain stable lighting, a suitable temperature, and provide a quiet scanning space. Multi-angle optical scanning is performed on the child's natural sleeping position to obtain a set of raw point cloud coordinates. A 3D optical scanner equipped with structured light or laser scanning technology is used to scan the child from four orthogonal directions (front, back, left, and right) and from the top. The minimum measurement accuracy of the scanner must reach 0.5mm. Each scan generates raw point cloud data containing X, Y, and Z three-dimensional coordinates. To improve data integrity, the relative position between the scanner and the child must be stable during the scanning process to avoid data loss due to minor movements of the child. After the scan is completed, the point cloud data in the four directions are aligned and fused to generate complete point cloud data of the child's head and neck. The original point cloud coordinate set is represented as P = {(xᵢ, yᵢ, zᵢ)|i = 1, 2, …, N}, where N is the total number of point clouds.

[0021] Step S12: positioning the skeleton key points of the original point cloud coordinate set and drawing a spatial distribution map of the skeleton key points; In an embodiment of the present invention, the skeleton key points are located on the original point cloud coordinate set to obtain a skeleton key point spatial distribution map. An automatic skeleton key point positioning algorithm based on computer vision is adopted. First, the point cloud data is filtered to remove noise points. Then, the point cloud segmentation algorithm is used to identify and extract the point cloud data of the head and neck. Next, a pre-trained convolutional neural network (CNN) model is applied to perform key point detection on the head and neck point cloud data. The key points include: C1 vertebra, C2 vertebra, C3 vertebra, C4 vertebra, C5 vertebra, C6 vertebra, C7 vertebra, skull base, mandibular angle, acromion, etc. The CNN model needs to be trained on a large amount of children's head and neck point cloud data to ensure positioning accuracy. In order to improve positioning accuracy, manual auxiliary correction can be combined. The skeleton key point spatial distribution map can be expressed as K={(x k ,y k , z k , k)|k=1, 2, …, M}, where M is the total number of key points and k represents the key point type.

[0022] Step S13: extracting a cervical curvature parameter group from the spatial distribution map of the skeleton key points; In an embodiment of the present invention, a cervical vertebra curvature parameter group of the skeletal key point spatial distribution map is extracted. According to the skeletal key point spatial distribution map, the cervical vertebra centerline trajectory is extracted. The key points of the C1 to C7 vertebrae are connected to obtain a three-dimensional coordinate sequence of the cervical vertebra centerline. The sequence is smoothed to eliminate noise. The spline interpolation algorithm is used to perform curve fitting on the cervical vertebra centerline to obtain a continuous cervical vertebra centerline trajectory. Multi-plane curvature calculation is performed on the cervical vertebra centerline trajectory to obtain a three-plane curvature distribution map. The curvature of the cervical vertebra centerline is calculated in the coronal plane, sagittal plane and horizontal plane respectively. The curvature calculation formula is κ=∣r″(t)×r′(t)∣ / ∣r′(t)∣³, where r(t) represents the parametric equation of the cervical vertebra centerline trajectory, and r′(t) and r″(t) represent its first-order and second-order derivatives, respectively. The key segment angle values ​​of the three-plane curvature distribution map are measured. The sagittal curvature angles of the C2-C7 cervical vertebrae, the C1-C2 segment angle, and the C6-C7 segment angle were measured. Curve length and morphology were measured based on the cervical centerline trajectory and key segment angles, generating a curve morphology feature set. The total length of the cervical centerline, the lengths of each segment, and the symmetry of the cervical curve were calculated. Standard curve deviation analysis was performed on the curve morphology feature set to generate the curvature standard deviation. The actual cervical curve was compared with the standard cervical curve, and the Euclidean distance between corresponding points was calculated. The root mean square error (RMSE) was then calculated as the curvature standard deviation. Stress distribution was estimated based on the curvature standard deviation to generate a cervical stress distribution map. A three-dimensional model of the bones and muscles of the child's head and neck was constructed using finite element analysis (FEA). Cervical curvature parameters were used as boundary conditions in the FEA model to simulate cervical stress conditions and output a cervical stress distribution map. Dynamic characteristics of the curvature were measured based on the cervical centerline trajectory and the curve morphology feature set to generate the dynamic characteristics of the curvature. Repeat the above steps in different sleeping positions to obtain the cervical curvature parameters under different sleeping positions. A cervical curvature parameter set is constructed based on the three-plane curvature distribution map, key segment angle values, curve morphology feature set, curvature standard deviation value, cervical stress distribution map, and curvature dynamic change characteristics.

[0023] Step S14: performing head-neck ratio measurement and analysis based on the spatial distribution map of the key bone points and the cervical curvature parameter group to obtain head-neck mechanical ratio data; In an embodiment of the present invention, head-neck ratio measurement and analysis is performed based on the spatial distribution diagram of skeletal key points and the cervical vertebra curvature parameter group to obtain head-neck mechanical ratio data. The relative position of the head weight distribution center and the cervical vertebra support point is measured. According to the spatial distribution diagram of skeletal key points, the position of the head center of mass is determined, and the coordinates of the head center of gravity are calculated. The cervical vertebra support point is measured and the coordinates of the cervical vertebra support center are calculated. The horizontal and vertical distances between the head center of gravity and the cervical vertebra support center are calculated. According to the cervical vertebra curvature parameter group, the support area of ​​the cervical vertebra is calculated. The proportional relationship between the head weight and the cervical vertebra support force is calculated. Based on the age grouping of children, a growth and development reference value of the head-neck ratio is established. The head-neck mechanical ratio data include: the coordinates of the head center of gravity, the coordinates of the cervical vertebra support center, the horizontal distance, the vertical distance, the support area, and the weight ratio.

[0024] Step S15: Based on the cervical curvature parameter group and the head-neck mechanical ratio data, multi-dimensional angle parameters of the angle between the head and the bed surface, the angle between the cervical spine and the shoulder, and the angle between the cervical spine axis and the horizontal plane are measured to construct a multi-dimensional posture angle matrix; In an embodiment of the present invention, based on the cervical curvature parameter group and the head-neck mechanical ratio data, the multi-dimensional angle parameters of the angle between the head and the bed surface, the angle of the line connecting the cervical spine and the shoulder, and the angle between the cervical spine axis and the horizontal plane are measured to construct a multi-dimensional posture angle matrix. Measure the angle between the head and the bed surface. According to the spatial distribution map of the key points of the bones, determine the highest point of the head and the contact point of the bed surface, and calculate the angle between the head and the bed surface. Measure the angle of the line connecting the cervical spine and the shoulder. According to the spatial distribution map of the key points of the bones, determine the position of the cervical vertebra C7 and the acromion, and calculate the angle between the cervical spine and the shoulder. Measure the angle between the cervical spine axis and the horizontal plane. Construct the cervical spine axis with the C1-C7 vertebral key points, and calculate the angle between the axis and the horizontal plane. The multi-dimensional posture angle matrix is ​​expressed as A=[α 11 , α 12 , α 13 ; α 21 , α 22 , α 23 ;…;α n1 , α n2 , α n3 ], where αᵢ1 represents the angle between the head and the bed surface in the i-th posture, αᵢ2 represents the angle between the cervical spine and the shoulder in the i-th posture, αᵢ3 represents the angle between the cervical spine axis and the horizontal plane in the i-th posture, and n represents the total number of postures.

[0025] Step S16: performing holographic data integration and standardization on the cervical curvature parameter group and the multi-dimensional posture angle matrix to obtain cervical curvature-posture holographic data.

[0026] In an embodiment of the present invention, the holographic data integration and standardization of the cervical curvature parameter group and the multi-dimensional posture angle matrix are performed to obtain the cervical curvature-posture holographic data. The cervical curvature parameter group, the head and neck mechanical ratio data and the multi-dimensional posture angle matrix are data fused. The dimensions and data formats are unified. The data are standardized to eliminate the dimensional effect. The Z-score standardization method is used to convert each parameter into a numerical value with a mean of 0 and a standard deviation of 1. Data is classified according to the age groups of children (0-3 years old, 3-6 years old, 6-10 years old, 10-16 years old), and age-related reference standards are established. The cervical curvature-posture holographic data is expressed as H={(C, L, A)|C∈cervical curvature parameter group, L∈head and neck mechanical ratio data, A∈multi-dimensional posture angle matrix}.

[0027] Preferably, the cervical curvature parameter group extracted from the skeleton key point spatial distribution map in step S13 includes: Extract the cervical vertebra centerline trajectory based on the spatial distribution map of skeletal key points; The multi-plane curvature of the cervical vertebrae centerline trajectory is calculated to obtain a three-plane curvature distribution map; Measure the key segment angle values ​​of the three-plane curvature distribution diagram; The curve length and morphology are measured based on the cervical spine centerline trajectory and key segment angle values ​​to obtain the curve morphology feature set; Perform standard curve deviation analysis on the curve morphology feature set to obtain the curvature standard deviation value; The stress distribution is estimated based on the curvature standard deviation value to obtain the cervical vertebra stress distribution map; The dynamic change characteristics of the curvature are obtained by measuring the dynamic change characteristics based on the cervical vertebra centerline trajectory and the curve morphology feature set; The cervical curvature parameter group was constructed based on the three-plane curvature distribution diagram, key segment angle values, curve morphological feature set, curvature standard deviation value, cervical stress distribution diagram and curvature dynamic change characteristics.

[0028] In an embodiment of the present invention, the cervical vertebra centerline trajectory is extracted based on the spatial distribution map of the skeletal key points. The key points of the vertebrae C1 to C7 are connected to form a point sequence in a three-dimensional space. The point sequence is recorded as P={P1, P2, ... P7}, where Pᵢ=(xᵢ, yᵢ, zᵢ) represents the coordinates of the key point of the i-th vertebra. P is fitted with a cubic B-spline curve to obtain a smooth cervical vertebra centerline trajectory. The parametric equation of the B-spline curve is C(t)=∑ᵢNᵢ, p (t)Pᵢ, where Nᵢ, p (t) is the B-spline basis function, and p is the order of the B-spline curve, usually p = 3. The parametric equation of the cervical vertebrae centerline trajectory is calculated to be C(t) = (x(t), y(t), z(t)), where t is a parameter with a value range of [0, 1].

[0029] The multi-plane curvature calculation of the cervical spine centerline trajectory was performed to obtain a three-plane curvature distribution map. The curvature of the cervical spine centerline was calculated in the coronal, sagittal, and transverse planes.

[0030] The formula for calculating the coronal plane curvature κc(t) is: κc(t)=∣x″(t)y′(t)−x′(t)y″(t)∣ / (x′(t)²+y′(t)²)³ / ². The formula for calculating the sagittal plane curvature κs(t) is κs(t)=∣x″(t)z′(t)−x′(t)z″(t)∣ / (x′(t)²+z′(t)²)³ / ². The formula for calculating the transverse plane curvature κh(t) is κh(t)=∣y″(t)z′(t)−y′(t)z″(t)∣ / (y′(t)²+z′(t)²)³ / ². where x′(t), y′(t), z′(t) and x″(t), y″(t), z″(t) are the first-order and second-order derivatives of C(t), respectively. The curvature distribution maps of the three planes are calculated, namely κc(t), κs(t), and κh(t).

[0031] Measure the angle values ​​of the key segments of the three-plane curvature distribution map. Identify the key segments of the cervical curve, including the C1-C2, C2-C7, and C6-C7 segments. On the sagittal curvature distribution map κs(t), find the corresponding positions of the C1, C2, and C7 vertebrae. Calculate the angle α of the C1-C2 segment. 12 The calculation method is to connect points C1 and C2 and calculate the angle between the connecting line and the horizontal line. Calculate the angle α of the C2-C7 segment 27 The calculation method is to connect C2 and C7 points and calculate the angle between the connecting line and the horizontal line. Calculate the angle α of the C6-C7 segment 67 The calculation method is to connect points C6 and C7 and calculate the angle between the connecting line and the horizontal line. The key segment angle value is α 12 , α 27 , α 67 .

[0032] The curve length and morphology were measured based on the cervical centerline trajectory and key segment angle values ​​to obtain a curve morphology feature set. The total length L of the cervical centerline was calculated using the integral formula L = ∫0¹ | C′(t) | dt, where C′(t) is the first-order derivative of C(t). The lengths L of the C1-C2, C2-C7, and C6-C7 segments were calculated. 12 , L 27 , L 67 . According to the trajectory of the cervical centerline, the symmetry of the cervical curve is calculated. The calculation method is to project the cervical centerline trajectory on the sagittal plane and calculate its symmetry about the y-axis. The curve morphological feature set includes: total length L, length of each segment L 12 , L27 , L 67 , symmetry index.

[0033] Perform standard curve deviation analysis on the curve morphology feature set to obtain the curvature standard deviation. Develop a standard curve model for the pediatric cervical spine. The standard curve model is an average cervical curve model based on age and gender, derived from statistical data from a large number of children's cervical spine data. Compare the actual cervical curve with the standard curve. Select n equally spaced points along the cervical centerline trajectory. Calculate the perpendicular distance dᵢ from each point to the standard curve. Calculate the curvature standard deviation (RMSE) as √(∑ᵢdᵢ² / n).

[0034] Stress distribution is estimated based on the standard deviation of curvature, resulting in a cervical spine stress distribution map. A three-dimensional model of the child's head and neck bones and muscles is constructed using finite element analysis (FEA). The model's geometry is reconstructed from CT or MRI scan data. Material properties are set based on the child's physiological characteristics. The cervical spine centerline trajectory and the standard deviation of curvature are input as boundary conditions into the FEA model. Constraints are set to simulate the free movement of the child's head in the horizontal plane. FEA calculations are performed to obtain a cervical spine stress distribution map. The stress distribution map shows the magnitude and distribution of stress in various parts of the cervical spine.

[0035] The dynamic change characteristics of the curvature are measured based on the trajectory of the cervical spine centerline and the curve morphological feature set. The dynamic change characteristics of the curvature are obtained. Repeat the above steps in different sleeping positions to obtain the cervical curvature parameters in different sleeping positions. Measure the cervical curvature of children in different sleeping positions such as supine and side lying. For each sleeping position, repeat the above steps to obtain the cervical spine centerline trajectory, curvature distribution, key segment angle value, curve length, morphological characteristics and curvature standard deviation value. Compare the changes in these parameters in different sleeping positions to obtain the dynamic change characteristics of the curvature. For example, calculate the difference in the C2-C7 segment angle when lying on the back and lying on the side to reflect the impact of sleeping position on the cervical curvature.

[0036] The cervical curvature parameter group is constructed based on the three-plane curvature distribution map, key segment angle value, curve morphological feature set, curvature standard deviation value, cervical stress distribution map and curvature dynamic change characteristics. The cervical curvature parameter group includes: three-plane curvature distribution map κc(t), κs(t), κh(t), key segment angle value α 12 , α 27 , α 67 , curve length and morphological measurement results L, L 12 , L 27 , L 67 , symmetry index, curvature standard deviation value RMSE, cervical spine stress distribution diagram, and dynamic change characteristics under different sleeping positions.

[0037] Preferably, step S2 includes the following steps: Step S21: accurately measuring the contact area based on the cervical vertebrae curvature-posture holographic data to obtain a contact area division map; Step S22: collecting a static pressure baseline data set according to the contact area division map; Step S23: monitoring the dynamic pressure change sequence according to the static pressure baseline data set and the contact area division map; Step S24: Calculating pressure uniformity based on the dynamic pressure change sequence and the static pressure baseline data set to obtain a pressure uniformity distribution matrix; Step S25: performing posture-pressure correlation analysis on the dynamic pressure change sequence and the cervical vertebrae curvature-posture holographic data to obtain a posture-pressure mapping relationship table; Step S26: performing pressure peak distribution analysis based on the body position-pressure mapping relationship table and the pressure uniformity distribution matrix to obtain a pressure peak distribution characteristic diagram; Step S27: Calculate the pressure distribution index map according to the pressure uniformity distribution matrix, the body position-pressure mapping relationship table and the pressure peak distribution characteristic map.

[0038] In an embodiment of the present invention, the contact area is accurately measured based on the cervical curvature-posture holographic data to obtain a contact area division map. The child is placed on a test bed covered with high-density pressure sensors. The pressure sensor adopts a capacitive or piezoresistive sensor, the sensor spacing is less than 1 cm, and the measurement accuracy is greater than 0.1kPa. According to the child's sleeping posture recorded in the cervical curvature-posture holographic data, the child is guided to maintain different postures such as supine and side lying. The pressure distribution data of the contact between the head and neck and the mattress / pillow is collected in real time through the pressure sensor array. According to the pressure data, the contact area is identified. A pressure threshold is set, and the area with a pressure value greater than the threshold is defined as the contact area. The contact area division map includes the headrest contact area, the neck support area and the shoulder transition area. The image segmentation algorithm is used to process the pressure distribution data, identify and mark the boundaries of different areas. The area of ​​each contact area is calculated. The contact area partition map can be expressed as C = { (xᵢ, yᵢ, areaᵢ, zoneᵢ) | i = 1, 2, …, N}, where xᵢ, yᵢ represent the center coordinates of the contact area, areaᵢ represents the contact area, and zoneᵢ represents the zone type (headrest, neck, shoulder).

[0039] A static pressure baseline dataset was collected based on the contact area demarcation map. The static pressure baseline dataset was collected while the child maintained a stable sleeping position. The data collection lasted 30 seconds. For each area in the contact area demarcation map, the pressure value, contact area, and center of pressure position were recorded. Pressure values ​​were directly measured using a pressure sensor. The contact area was calculated based on the contact area demarcation map. The center of pressure position was calculated using the formula (x_c, y_c) = (∑ᵢpᵢxᵢ / ∑ᵢpᵢ, ∑ᵢpᵢyᵢ / ∑ᵢpᵢ), where pᵢ is the pressure value measured by each sensor and (xᵢ, yᵢ) is the sensor position. The static pressure baseline dataset is represented as S = {(zoneᵢ, pᵢ, areaᵢ, x_cᵢ, y_cᵢ)|i=1, 2, …, M}, where zoneᵢ represents the zone type, pᵢ represents the average pressure value, areaᵢ represents the contact area, (x_cᵢ, y_cᵢ) represents the pressure center coordinates, and M is the total number of contact areas.

[0040] Dynamic pressure change sequences are monitored based on a static pressure baseline dataset and a contact zone delineation map. Children are guided through natural body position changes on the test bed, such as from supine to side-lying. Simultaneously, pressure changes throughout this process are recorded using a high-frequency sampling pressure sensor array (sampling frequency greater than 20 Hz). During these body position changes, the pressure value of each sensor is continuously monitored. The pressure value of each sensor at each sampling moment is recorded. Sensor data are associated with contact zones based on the contact zone delineation map. The dynamic pressure change sequence is represented as D = {(tⱼ, zoneᵢ, pᵢⱼ) | i = 1, 2, …, M; j = 1, 2, …, T}, where tⱼ represents the sampling time, zoneᵢ represents the zone type, pᵢⱼ represents the average pressure value in zoneᵢ at time tⱼ, and T represents the total sampling time.

[0041] The pressure uniformity is calculated based on the dynamic pressure change sequence and the static pressure baseline data set to obtain the pressure uniformity distribution matrix. For each contact area, the uniformity coefficient of the pressure distribution is calculated. The variance of the pressure distribution is calculated as σ²=∑ᵢ(pᵢ−p̄)² / n, where pᵢ is the pressure value of each sensor, p̄ is the average pressure value, and n is the number of sensors. The pressure peak-to-valley ratio is calculated, that is, the ratio of the maximum pressure value to the minimum pressure value. The pressure uniformity coefficient U=1−σ² / σ_max², where σ_max² is the maximum variance. The pressure uniformity distribution matrix is ​​expressed as U=[U 11 , U 12 , U 13 ;U 21 , U 22 , U 23 ;…;U m1 , U m2 , U m3 ], where Uᵢⱼ represents the pressure uniformity of the jth region under the i-th posture.

[0042] Perform position-pressure correlation analysis on the dynamic pressure change sequence and the cervical curvature-position holographic data to generate a position-pressure mapping table. Time-synchronize the dynamic pressure change sequence D with the cervical curvature-position holographic data H. Match the different data based on the timestamp. Perform spatial coordinate mapping on the time-synchronized data to map the pressure data to the coordinates of key skeletal points. Use linear interpolation or nearest neighbor interpolation to map the pressure data to the coordinates of key skeletal points. Extract standard body positions. Define standard positions such as supine, left lateral decubitus, and right lateral decubitus. Identify the child's position based on the cervical curvature-position holographic data. Calculate the pressure transfer trajectory. During the position change process, track the changes in the center of pressure and obtain the pressure center's movement trajectory. Calculate the pressure response delay. Calculate the time required for the pressure in each region to reach a stable state after the position change. Calculate the position change influence coefficient. Calculate the degree of influence of different body positions on the mean pressure of each region. The posture-pressure mapping relationship table is expressed as M = {(poseᵢ, zoneⱼ, pᵢⱼ, trajectoryᵢⱼ, delayᵢⱼ, coefficientᵢⱼ) | i = 1, 2, …, N; j = 1, 2, …, M}, where poseᵢ represents the posture, zoneⱼ represents the zone type, pᵢⱼ represents the average pressure value of zoneⱼ under posture poseᵢ, trajectoryᵢⱼ represents the pressure transfer trajectory of zoneⱼ during posture change, delayᵢⱼ represents the pressure response delay of zoneⱼ, and coefficientᵢⱼ represents the influence coefficient of posture change.

[0043] Based on the body position-pressure mapping table and the pressure uniformity distribution matrix, peak pressure distribution analysis is performed to generate a peak pressure distribution feature map. The location of the peak pressure point is identified. Within the pressure distribution data, the point with the maximum pressure value is found. Based on the contact area demarcation map, the region containing the peak pressure point is determined. The correspondence between the peak pressure point and key anatomical structures of the cervical spine is analyzed. Based on the spatial distribution map of skeletal key points, the distance between the peak pressure point and key cervical spine points is determined. The potential impact of the peak pressure point on cervical curvature is assessed. Based on the location of the peak pressure point and cervical curvature parameters, the impact of the peak pressure point on cervical curvature is evaluated. The peak pressure distribution feature map is represented as F = {(zoneᵢ, peakᵢ, locationᵢ, distanceᵢ, influenceᵢ) | i = 1, 2, …, M}, where zoneᵢ represents the zone type, peakᵢ represents the peak pressure point, locationᵢ represents the peak point location, distanceᵢ represents the distance between the peak point and key cervical spine points, and influenceᵢ represents the degree of impact of the peak pressure point on cervical curvature.

[0044] A pressure distribution index map is calculated based on the pressure uniformity distribution matrix, the body position-pressure mapping table, and the peak pressure distribution characteristic map. A multidimensional pressure evaluation index system is established by comprehensively considering pressure uniformity, body position-pressure mapping, and peak pressure distribution. Weight coefficients are set, and weights for different factors are assigned based on the child's age group. A comprehensive pressure distribution index is calculated. The calculation formula can be a weighted average or other appropriate calculation method. The results are visualized as a color-coded map. The pressure distribution index map is represented as G = {(xᵢ, yᵢ, indexᵢ, colorᵢ) | i = 1, 2, …, N}, where xᵢ and yᵢ represent position coordinates, indexᵢ represents the pressure distribution index, and colorᵢ represents color.

[0045] It is particularly important that the static pressure baseline data set collected in step S22 is specifically: Perform posture stabilization guidance acquisition based on the contact area division map to obtain a standard posture record table; Collect the original pressure data matrix according to the standard posture record form; Extract the regional pressure characteristic table of the original pressure data matrix; Calculate the pressure uniformity distribution map based on the regional pressure characteristic table; Performing pressure gradient analysis on the pressure uniformity distribution diagram to obtain a pressure gradient distribution diagram; Perform anatomical structure pressure mapping according to the pressure gradient distribution map to obtain an anatomical pressure correspondence table; The pressure data was normalized according to the anatomical pressure correspondence table to obtain a static pressure baseline data set; In an embodiment of the present invention, posture stability guidance collection is performed based on the contact area division map to obtain a standard posture record table. According to the contact area division map C={(xᵢ, yᵢ, areaᵢ, zoneᵢ)|i=1, 2, ..., N}, the child is guided to maintain a stable posture. A visual guidance system or voice prompts are used to guide the child to maintain a standard sleeping posture such as supine or side-lying. The collection time is recorded and set to 30 seconds. The child's body position on the test bed is recorded, such as supine or left-side lying. The contact area division situation in each body position is recorded. The standard posture record table is expressed as P={(poseᵢ, timeᵢ, zoneᵢ)|i=1, 2, ..., M}, where poseᵢ represents the posture type, timeᵢ represents the collection time, and zoneᵢ represents the contact area.

[0046] The raw pressure data matrix is ​​collected according to the standard posture record table. In each standard posture, the raw pressure data is collected using a pressure sensor array. The acquisition frequency is set to be above 20Hz. The pressure sensor array consists of multiple pressure sensors, each corresponding to a data point. The pressure value of each sensor at each sampling time is recorded. The raw pressure data matrix is ​​expressed as R = [r11 r 12 …r 1k ; r 21 r 22 …r 2k ;…;r n1 r n2 …r nk ], where rᵢⱼ represents the pressure value of the jth sensor at the i-th sampling moment, n represents the total number of sampling times, and k represents the total number of sensors.

[0047] Extract the regional pressure signature table from the original pressure data matrix. Based on the contact zone partitioning map, map the pressure sensor data to the contact zones. Calculate the average, maximum, minimum, and standard deviation of pressure for each contact zone. The regional pressure signature table is represented as T = {(zoneᵢ, p̄ᵢ, p_maxᵢ, p_minᵢ, σᵢ) | i = 1, 2, …, M}, where zoneᵢ represents the zone type, p̄ᵢ represents the average pressure value, p_maxᵢ represents the maximum pressure value, p_minᵢ represents the minimum pressure value, and σᵢ represents the standard deviation of pressure.

[0048] Calculate the pressure uniformity distribution map based on the regional pressure characteristic table. Calculate the pressure uniformity for each contact area. Using the variance calculation method, calculate the variance σ² of the pressure distribution. The pressure uniformity calculation formula is U = 1 − σ² / σ_max², where σ_max² is the maximum variance. The pressure uniformity distribution map is represented as U = {(xᵢ, yᵢ, Uᵢ) | i = 1, 2, …, N}, where xᵢ and yᵢ represent the center coordinates of the contact area, and Uᵢ represents the pressure uniformity.

[0049] Perform pressure gradient analysis on the pressure uniformity distribution map to obtain a pressure gradient distribution map. Calculate the direction and magnitude of the pressure gradient. Use the Sobel operator or Prewitt operator to calculate the gradient of the pressure distribution. The gradient direction indicates the direction of the fastest pressure change, and the gradient magnitude indicates the rate of pressure change. The pressure gradient distribution map is represented as G = {(xᵢ, yᵢ, gradient_magnitudeᵢ, gradient_directionᵢ) ​​| i = 1, 2, …, N}, where xᵢ and yᵢ represent the center coordinates of the contact area, gradient_magnitudeᵢ indicates the magnitude of the pressure gradient, and gradient_directionᵢ indicates the direction of the pressure gradient.

[0050] Based on the pressure gradient distribution map, pressure mapping of anatomical structures is performed to generate an anatomical pressure correspondence table. Based on the spatial distribution map of skeletal key points, pressure gradients are mapped to anatomical structures. The direction and magnitude of the pressure gradient are mapped to anatomical structures such as the cervical spine and shoulder. For example, the direction of the pressure gradient is compared to the axis of the cervical spine. The anatomical pressure correspondence table is represented as A = {(zoneᵢ, gradient_magnitudeᵢ, gradient_directionᵢ, anatomical_structureᵢ)|i=1,2,…,M}, where zoneᵢ represents the zone type, gradient_magnitudeᵢ represents the pressure gradient magnitude, gradient_directionᵢ represents the pressure gradient direction, and anatomical_structureᵢ represents the corresponding anatomical structure.

[0051] Pressure data were normalized according to the anatomical pressure correspondence table to generate a static pressure baseline dataset. Pressure data were normalized to eliminate dimension effects. Using the Z-score normalization method, each feature was converted to a numerical value with a mean of 0 and a standard deviation of 1. The static pressure baseline dataset is represented as S = {(zoneᵢ, p̄_stdᵢ, p_max_stdᵢ, p_min_stdᵢ, σ_stdᵢ)|i=1,2,…,M}, where zoneᵢ represents the zone type, p̄_stdᵢ represents the normalized mean pressure value, p_max_stdᵢ represents the normalized maximum pressure value, p_min_stdᵢ represents the normalized minimum pressure value, and σ_stdᵢ represents the normalized standard deviation of pressure.

[0052] Preferably, the body position-pressure correlation analysis in step S25 includes: Performing time series synchronization processing on the dynamic pressure change sequence and the cervical curvature-posture holographic data to obtain a time-synchronized data pair; Perform spatial coordinate mapping transformation on the time-synchronized data pair to obtain a spatial corresponding mapping matrix; Extract the standard body posture state from the spatial corresponding mapping matrix and the time synchronization data to obtain the standard body posture pressure characteristics; Calculate the pressure transfer trajectory according to the standard body position pressure characteristics to obtain the pressure transfer trajectory set; The pressure response delay is measured on the pressure transfer trajectory set to obtain the regional pressure response delay table; Calculate the body position change influence coefficient table based on the regional pressure response time delay table and the pressure transfer trajectory set; Construct a body position-pressure mapping relationship table based on the body position change influence coefficient table.

[0053] In an embodiment of the present invention, a dynamic pressure change sequence and cervical curvature-posture holographic data are subjected to time series synchronization processing to obtain a time-synchronized data pair. The dynamic pressure change sequence D = {(tⱼ, zoneᵢ, pᵢⱼ) | i = 1, 2, ..., M; j = 1, 2, ..., T}, and the cervical curvature-posture holographic data H = {(C, L, A)}. The timestamp of the dynamic pressure change sequence is tⱼ, and the timestamp of the cervical curvature-posture holographic data is recorded by the scanner. Using a linear interpolation method, the timestamp of the cervical curvature-posture holographic data is interpolated to align it with the timestamp of the dynamic pressure change sequence. If the sampling frequency of the cervical curvature-posture holographic data is lower than that of the dynamic pressure change sequence, the cervical curvature-posture holographic data is upsampled. If the sampling frequency of the cervical curvature-posture holographic data is higher than that of the dynamic pressure change sequence, the dynamic pressure change sequence is downsampled. The time-synchronized data pair is represented as {(tⱼ, Dⱼ, Hⱼ)}, where Dⱼ represents the dynamic pressure change sequence data at time tⱼ, and Hⱼ represents the cervical curvature-posture holographic data at time tⱼ.

[0054] Perform spatial coordinate mapping transformation on the time synchronization data pair to obtain the spatial corresponding mapping matrix. The data of the dynamic pressure change sequence is compared with the spatial distribution map of the key points of the skeleton K={(x k ,y k , z k , k)|k=1, 2, …, M} for spatial mapping. For each moment tⱼ, determine the position of each pressure sensor in space. Use the coordinate transformation matrix to map the pressure sensor coordinates to the skeleton key point coordinate system. The coordinate transformation matrix is ​​determined by the calibration result of the scanner. The spatial corresponding mapping matrix is ​​expressed as M=[m 11 m 12 …m 1k ;m 21 m 22 …m 2k ;…;m n1 m n2 …m nk ], where mᵢⱼ represents the spatial coordinates of the j-th pressure sensor mapped to the skeleton key point k at time tᵢ.

[0055] Standard body position states are extracted from the spatial mapping matrix and time-synchronized data pairs to obtain standard body position pressure signatures. Standard body positions are defined, such as supine, left side, and right side. Based on the cervical vertebral curvature-position holographic data, the child's position at each moment is identified. Pressure signatures for standard body positions are extracted. For each standard body position, the mean pressure value, maximum pressure value, minimum pressure value, and pressure standard deviation are calculated. The standard body position pressure feature is expressed as P = {(poseᵢ, zoneⱼ, p̄ᵢⱼ, p_maxᵢⱼ, p_minᵢⱼ, σᵢⱼ)|i = 1, 2, …, N; j = 1, 2, …, M}, where poseᵢ represents the body position, zoneⱼ represents the zone type, p̄ᵢⱼ represents the average pressure value, p_maxᵢⱼ represents the maximum pressure value, p_minᵢⱼ represents the minimum pressure value, and σᵢⱼ represents the pressure standard deviation.

[0056] Pressure transfer trajectories are calculated based on the pressure characteristics of standard body positions, resulting in a set of pressure transfer trajectories. During body position changes, the changes in the center of pressure are tracked. The position of the center of pressure at each moment is calculated using the formula (x_c, y_c) = (∑ᵢpᵢxᵢ / ∑ᵢpᵢ, ∑ᵢpᵢyᵢ / ∑ᵢpᵢ), where pᵢ is the pressure value of each sensor and (xᵢ, yᵢ) is the sensor position. The pressure centers are connected to form pressure transfer trajectories. The set of pressure transfer trajectories is represented as T = {(poseᵢ, trajectoryᵢ) | i = 1, 2, …, N}, where poseᵢ represents the body position and trajectoryᵢ represents the trajectory of the center of pressure during the body position change.

[0057] The pressure response delay is measured for the pressure transition trajectory set to generate a regional pressure response delay table. The time required for each region's pressure to reach a stable state after a body position change is measured. A pressure stability criterion is defined, for example, a pressure change of less than 5%. For each region, the time required for the pressure to reach a stable state is calculated. The regional pressure response delay table is represented as D = {(zoneᵢ, delayᵢ) | i = 1, 2, …, M}, where zoneᵢ represents the zone type and delayᵢ ​​represents the pressure response delay.

[0058] Based on the regional pressure response delay table and the pressure transition trajectory set, a table of body position influence coefficients is calculated. The influence of different body positions on the mean pressure of each region is calculated. For each region, the average pressure difference under different body positions is calculated. The body position influence coefficient is calculated. The body position influence coefficient is expressed as C = {(poseᵢ, zoneⱼ, coefficientᵢⱼ) | i = 1, 2, …, N; j = 1, 2, …, M}, where poseᵢ represents the body position, zoneⱼ represents the zone type, and coefficientᵢⱼ represents the body position influence coefficient.

[0059] Construct a posture-pressure mapping table based on the posture change influence coefficient table. Construct the posture-pressure mapping table M. The posture-pressure mapping table includes body position, zone, average pressure, pressure transition trajectory, pressure response delay, and posture change influence coefficient. The posture-pressure mapping table is represented as M = {(poseᵢ, zoneⱼ, p̄ᵢⱼ, trajectoryᵢⱼ, delayᵢⱼ, coefficientᵢⱼ) | i = 1, 2, …, N; j = 1, 2, …, M}, where poseᵢ represents body position, zoneⱼ represents zone type, p̄ᵢⱼ represents average pressure, trajectoryᵢⱼ represents pressure transition trajectory, delayᵢⱼ represents pressure response delay, and coefficientᵢⱼ represents the posture change influence coefficient.

[0060] Preferably, the curvature-support effect quantification in step S3 includes: The age staging standard was established based on the pressure distribution index map and cervical curvature-posture holographic data to obtain the reference standard for cervical development stages; The ideal curvature model was constructed based on the cervical development stage reference standard and cervical curvature-posture holographic data to obtain the ideal curvature model for age stages. The actual support effect is measured based on the ideal curvature model of age stages and the pressure distribution index map to obtain the dynamic curvature change trajectory; The curvature deviation was quantitatively analyzed between the dynamic curvature change trajectory and the ideal curvature model of age stages, and a set of curvature deviation indices was obtained.

[0061] In an embodiment of the present invention, an age staging standard is established based on the pressure distribution index map and the cervical curvature-posture holographic data to obtain a reference standard for cervical development staging. Children are grouped according to their age and divided into four age groups: 0-3 years old, 3-6 years old, 6-10 years old, and 10-16 years old. Based on the cervical curvature-posture holographic data of a large number of children, the physiological curvature characteristics of the cervical spine in each age group are statistically analyzed. Parameters such as cervical curvature and vertebral angle are analyzed, and standard values ​​and acceptable ranges are set. The head-neck ratio of different age groups is analyzed to establish corresponding reference standards. The pressure distribution index map is analyzed to extract the pressure distribution characteristics of different age groups. The reference standard for cervical development staging includes: age group, standard value of cervical curvature parameters, reference value of head-neck ratio, and pressure distribution characteristics.

[0062] The ideal curvature model is constructed based on the cervical development stage reference standard and the cervical curvature-posture holographic data to obtain the ideal curvature model for age staging. For each age group, an ideal cervical curvature model is constructed. The ideal curvature model is based on the cervical development stage reference standard and a large number of children's cervical data. The spline curve fitting method is used to fit the cervical centerline to obtain the morphological parameters of the ideal cervical curve. Morphological parameters include: curvature, arc length, key point angle, etc. The ideal curvature model is expressed by a parametric equation, C(t)=(x(t), y(t), z(t)), where t is a parameter with a value range of [0, 1]. The ideal curvature model for age staging includes: age group, ideal cervical curve parametric equation, and cervical key point coordinates.

[0063] The ideal curvature model is constructed based on the cervical development stage reference standard and the cervical curvature-posture holographic data to obtain the ideal curvature model for age staging. For each age group, an ideal cervical curvature model is constructed. The ideal curvature model is based on the cervical development stage reference standard and a large number of children's cervical data. The spline curve fitting method is used to fit the cervical centerline to obtain the morphological parameters of the ideal cervical curve. Morphological parameters include: curvature, arc length, key point angle, etc. The ideal curvature model is expressed by a parametric equation, C(t)=(x(t), y(t), z(t)), where t is a parameter with a value range of [0, 1]. The ideal curvature model for age staging includes: age group, ideal cervical curve parametric equation, and cervical key point coordinates.

[0064] Curvature deviations were quantitatively analyzed between the dynamic curvature trajectory and the ideal curvature model for age-staged cervical spine, yielding a set of curvature deviation indices. The actual cervical curvature was compared with the ideal curvature model. The deviation between the actual and ideal cervical curvatures was calculated. The distance from a point on the actual cervical curve to the ideal cervical curve was calculated using the point-to-curve distance method. The curve integral difference, defined as the area difference between the actual and ideal curves, was calculated. The degree and distribution of curvature deviation were quantitatively assessed. A comprehensive deviation index was calculated. The curvature deviation index set is represented as D = {(deviation value, distribution, index) | (Cᵢ - Cᵢ) | i = 1, 2, …, N}, where Cᵢ represents the actual cervical curvature parameter and Cᵢ represents the ideal cervical curvature parameter.

[0065] Preferably, the analysis of the pressure-curvature response relationship in step S3 includes: Perform high-frequency synchronous data acquisition on the pressure distribution index map to obtain pressure-curvature synchronous time series data; perform time window segmentation processing on the pressure-curvature synchronous time series data to obtain a windowed change rate data set; The key pressure areas are divided according to the windowed change rate data set and the pressure distribution index map, and the pressure characteristic table of the functional areas is obtained; According to the functional area pressure characteristic table, regional sensitivity measurement is carried out to obtain the regional curvature sensitivity coefficient table; Identify critical pressure points based on the regional curvature sensitivity coefficient table and the functional area pressure characteristic table to obtain a critical pressure point location map; According to the critical pressure point position diagram, the pressure-curvature mapping function is constructed to obtain a point-curve mapping function set; The time delay characteristics are analyzed based on the point-curve mapping function set to obtain the regional response delay table; Generate a pressure-curvature response matrix based on the regional response time delay table; The recovery score is calculated based on the pressure-curvature response matrix and the curvature deviation index set to obtain the curvature recovery score table; The adaptation factor is comprehensively generated according to the curvature recovery score table and the pressure-curvature response matrix to obtain the curvature recovery adaptation factor.

[0066] In an embodiment of the present invention, high-frequency synchronous data acquisition is performed on the pressure distribution index map to obtain pressure-curvature synchronous time series data. A high-sampling rate pressure sensor array and a cervical curvature measurement device are used to synchronously acquire the pressure distribution index map and cervical curvature data. The sampling frequency of the pressure sensor array is greater than 20 Hz. The cervical curvature measurement device uses an optical or inertial measurement unit (IMU) with a sampling frequency greater than 20 Hz. The synchronously acquired data includes: the pressure value at each moment and the cervical curvature parameter at each moment. The pressure distribution index map is represented by G={(xᵢ, yᵢ, indexᵢ)|i=1, 2, …, N}, and the cervical curvature data is represented by C(t)=(x(t), y(t), z(t)). The pressure-curvature synchronization time series data is expressed as S = {(tⱼ, Gⱼ, Cⱼ)|j = 1, 2, …, T}, where tⱼ represents time, Gⱼ represents the pressure distribution index map at time tⱼ, and Cⱼ represents the cervical curvature data at time tⱼ.

[0067] The pressure-curvature synchronization time series data is segmented into time windows to obtain a windowed rate of change data set. The pressure-curvature synchronization time series data S is segmented into multiple time windows. The length of each time window is set to 1 second. For each time window, the rate of change of the pressure distribution index and the rate of change of the cervical curvature parameter are calculated. The rate of change of the pressure distribution index is obtained by calculating the difference between the pressure indices at adjacent moments. The rate of change of the cervical curvature parameter is obtained by calculating the difference between the cervical curvature parameters at adjacent moments. The windowed rate of change data set is expressed as W={(t k , ΔG k , ΔC k )|k=1, 2, …, K}, where t k represents the central moment of the time window, ΔG kIndicates the rate of change of the pressure distribution index within the time window, ΔC k represents the rate of change of the cervical curvature parameters within the time window, and K represents the total number of time windows.

[0068] Key pressure areas are divided based on the windowed rate of change dataset and the pressure distribution index map, resulting in a functional area pressure signature table. Key pressure areas are divided based on the pressure distribution index map. These areas include the headrest area, neck support area, and shoulder transition area. For each key pressure area, the average, maximum, minimum, and rate of change of the pressure distribution index are calculated. The functional area pressure signature table is represented as F = {(zoneᵢ, p̄ᵢ, p_maxᵢ, p_minᵢ, Δpᵢ)|i=1, 2, …, M}, where zoneᵢ represents the area type, p̄ᵢ represents the average pressure distribution index, p_maxᵢ represents the maximum pressure distribution index, p_minᵢ represents the minimum pressure distribution index, Δpᵢ represents the rate of change of the pressure distribution index, and M represents the total number of key pressure areas.

[0069] Based on the functional zone pressure characteristic table, regional sensitivity was measured to generate a regional curvature sensitivity coefficient table. The impact of each key pressure zone on cervical curvature was calculated. This was done by calculating the correlation coefficient between the pressure change rate and the cervical curvature change rate for each key pressure zone. The correlation coefficient formula is r = ∑ᵢ[(xᵢ−x̄)(yᵢ−ȳ)] / √(∑ᵢ(xᵢ−x̄)²∑ᵢ(yᵢ−ȳ)²), where xᵢ represents the pressure change rate, yᵢ represents the cervical curvature change rate, and x̄ and ȳ represent their average values, respectively. The regional curvature sensitivity coefficient table is represented as R = {(zoneᵢ, coefficientᵢ)|i=1, 2, …, M}, where zoneᵢ represents the zone type and coefficientᵢ represents the correlation coefficient.

[0070] Critical pressure points are identified using the regional curvature sensitivity coefficient table and the functional area pressure characteristic table, resulting in a critical pressure point location map. Based on the regional curvature sensitivity coefficient, the pressure points that have the greatest impact on cervical curvature are identified. For each critical pressure zone, the point with the greatest rate of pressure change is found. The critical pressure point location map is represented as L = {(zoneᵢ, (xᵢ, yᵢ)) | i = 1, 2, …, M}, where zoneᵢ represents the zone type and (xᵢ, yᵢ) represents the location of the critical pressure point.

[0071] Based on the critical pressure point location map, a pressure-curvature mapping function is constructed to obtain a set of point-curve mapping functions. A mapping function is constructed that relates pressure to cervical curvature. For each critical pressure point, a corresponding relationship is established between changes in pressure and changes in cervical curvature. This mapping can be performed using linear functions, polynomial functions, or neural networks. The set of point-curve mapping functions is represented as M = {(pointᵢ, fᵢ) | i = 1, 2, …, M}, where pointᵢ represents a critical pressure point and fᵢ represents a pressure-curvature mapping function.

[0072] Time delay characteristics are analyzed based on a set of point-to-curve mapping functions to generate a regional response delay table. The time delay between pressure changes and cervical curvature changes is analyzed. The time delay between pressure changes and cervical curvature changes is calculated. Time delay can be calculated using cross-correlation analysis or time series analysis. The regional response delay table is represented as D = {(zoneᵢ, delayᵢ) | i = 1, 2, …, M}, where zoneᵢ represents the zone type and delayᵢ ​​represents the time delay.

[0073] Generate a pressure-curvature response matrix based on the regional response delay table. Construct a pressure-curvature response matrix. The rows of the matrix represent the pressure regions, the columns represent the cervical curvature parameters, and the matrix elements represent the effect of pressure changes on the cervical curvature. The pressure-curvature response matrix is ​​expressed as R=[r 11 r 12 …r 1k ; r 21 r 22 …r 2k ;…;r m1 r m2 …r mk ], where rᵢⱼ represents the effect of the i-th pressure area on the j-th cervical curvature parameter.

[0074] The recovery score is calculated based on the pressure-curvature response matrix and the curvature deviation index set, resulting in a curvature recovery score table. The cervical curvature recovery score is calculated based on the pressure-curvature response matrix and the curvature deviation index set. Curvature deviations are weighted based on the degree to which different pressure zones affect cervical curvature. The recovery score ranges from 0 to 100. The curvature recovery score table is represented as S = {(scoreᵢ, weightᵢ)|i=1, 2, …, N}, where scoreᵢ represents the recovery score and weightᵢ represents the weight.

[0075] The adaptation factor is generated based on the curvature recovery score and the pressure-curvature response matrix to obtain the curvature recovery adaptation factor. The curvature recovery adaptation factor is generated by comprehensively considering the curvature recovery score and the pressure-curvature response characteristics. The adaptation factor includes the recovery coefficient, the adaptability index, and the developmental enhancement value. The curvature recovery adaptation factor is expressed as A = (recovery coefficient, adaptability index, developmental enhancement value).

[0076] Preferably, calculating the biomechanical support index table in step S4 includes: Construct a material performance parameter library based on pillow material performance data; perform material rebound response analysis based on the material performance parameter library and curvature recovery adaptation factor to obtain a material rebound characteristic evaluation table; Collect the head and neck biomechanical parameter set according to the curvature recovery adaptation factor; calculate the regional ideal support force table according to the head and neck biomechanical parameter set; Measuring material support force curve sets based on regional ideal support force tables; Calculate the regional support matching index based on the material support force curve set; Posture difference analysis is performed based on the regional support matching index to obtain the posture support correction coefficient; Monitor and analyze sleeping posture ratios based on posture support correction coefficients to obtain a sleeping posture time ratio table; Evaluate the support stability coefficient based on the sleeping posture time percentage table; The comprehensive support index is calculated based on the regional support matching index, posture support correction coefficient, sleeping posture time ratio table and support stability coefficient to obtain the biomechanical support index table.

[0077] In an embodiment of the present invention, a material performance parameter library is constructed based on pillow material performance data. Performance data of various pillow materials are collected, including memory foam, latex, fiber, hydrogel, etc. Material performance data include: hardness, rebound time, density, air permeability, thermal conductivity, durability, compressive strength, etc. Hardness is measured using a Shore hardness tester. Rebound time is measured using a pressure sensor and a timer, and is defined as the time required for the material to return to its initial state after being compressed. Density is calculated using mass and volume. Air permeability is measured using an air permeability tester. Thermal conductivity is measured using a thermal conductivity meter. Durability is tested using a fatigue tester. Compressive strength is tested using a universal testing machine. The material performance parameter library is expressed as P={(materialᵢ, hardnessᵢ, rebound_timeᵢ, densityᵢ, air_permeabilityᵢ, thermal_conductivityᵢ, durabilityᵢ, compressive_strengthᵢ)|i=1, 2, …, N}, where materialᵢ represents the material type, hardnessᵢ represents hardness, rebound_timeᵢ represents rebound time, densityᵢ represents density, air_permeabilityᵢ represents air permeability, thermal_conductivityᵢ represents thermal conductivity, durabilityᵢ represents durability, compressive_strengthᵢ represents compressive strength, and N represents the total number of materials.

[0078] Based on the material performance parameter library and the curvature recovery adaptation factor, the material rebound response is analyzed to generate a material rebound characteristic evaluation table. Key material performance indicators are determined based on the curvature recovery adaptation factor. Key material performance indicators include rebound time, hardness, and support force. Use pressure sensors and displacement sensors to measure the material's rebound response characteristics. Measure the material's rebound time under different pressures. Measure the material's deformation under different pressures. The material rebound characteristic evaluation table is represented as T = {(materialᵢ, pressureᵢ, rebound_timeᵢ, deformationᵢ) ​​| i = 1, 2, …, N}, where materialᵢ represents the material type, pressureᵢ represents the pressure, rebound_timeᵢ represents the rebound time, deformationᵢ represents the deformation, and N represents the total number of materials.

[0079] A set of head and neck biomechanical parameters was collected based on the curvature recovery adaptation factor. Based on the curvature recovery adaptation factor, the biomechanical parameters of the child's head and neck were determined. These biomechanical parameters include head weight, neck length, cervical curvature, and muscle strength. Head weight was measured using an electronic scale. Neck length was measured using a tape measure. Cervical curvature was measured using a 3D scanner. Muscle strength was measured using a dynamometer. The head and neck biomechanical parameter set is represented as B = {(weight, length, curvature, muscle_strength)}.

[0080] Calculate the ideal regional support force table based on the head and neck biomechanical parameter set. Based on the head and neck biomechanical parameters, calculate the ideal support force for each zone. The ideal support force is related to head weight, neck length, and cervical curvature. The calculation formula can use a mechanical model or an empirical formula. The ideal support force table is represented as I = {(zoneᵢ, ideal_forceᵢ) | i = 1, 2, …, M}, where zoneᵢ represents the zone type, ideal_forceᵢ represents the ideal support force, and M represents the total number of zones.

[0081] Measure the material support force curve set based on the regional ideal support force table. Use a universal testing machine to measure the support force of different materials under different pressures. Measure the support force of different materials under different deformations. The material support force curve set is represented as C = {(materialᵢ, pressureᵢ, forceᵢ)|i=1, 2, …, N}, where materialᵢ represents the material type, pressureᵢ represents the pressure, forceᵢ represents the support force, and N represents the total number of materials.

[0082] Calculate the regional support matching index based on the material support curve set. This index measures the degree of match between the material support and the ideal support. This index is calculated by taking the difference between the material support and the ideal support. For example, relative error or correlation coefficients can be used to calculate the matching index. The regional support matching index is expressed as M = {(zoneᵢ, materialᵢ, matching_indexᵢ) | i = 1, 2, …, M}, where zoneᵢ represents the zone type, materialᵢ represents the material type, matching_indexᵢ represents the support matching index, and M represents the total number of zones.

[0083] Posture differences are analyzed based on the regional support matching index to obtain the postural support correction factor. The differences in support matching between different sleeping positions are analyzed. The support matching index is calculated for different sleeping positions. The postural support correction factor is expressed as S = {(poseᵢ, zoneⱼ, correction_factorᵢⱼ)|i=1,2,…,P;j=1,2,…,M}, where poseᵢ represents the posture type, zoneⱼ represents the zone type, correction_factorᵢⱼ represents the postural support correction factor, P represents the total number of postures, and M represents the total number of zones.

[0084] Sleeping posture proportion monitoring and analysis is performed using the posture support correction factor to generate a sleeping posture time percentage table. Monitor the child's sleep time in different sleeping postures. Record the sleep time in each sleeping posture. Calculate the sleep time percentage for each sleeping posture. The sleeping posture time percentage table is represented as T = {(poseᵢ, percentageᵢ) | i = 1, 2, …, P}, where poseᵢ represents the posture type, percentageᵢ represents the sleep time percentage, and P represents the total number of postures.

[0085] Evaluate the support stability coefficient based on the sleep posture time percentage table. Evaluate the support stability of pillows / bedding. Support stability is related to the material's rebound time, hardness, and shape. Calculate the support stability coefficient. The support stability coefficient is expressed as Y = {stability_coefficient}.

[0086] A comprehensive support index is calculated based on the regional support fit index, postural support correction factor, sleep time percentage, and support stability coefficient to produce a biomechanical support index table. The biomechanical support index is calculated by comprehensively considering the regional support fit index, postural support correction factor, sleep time percentage, and support stability. The calculation formula can use a weighted average or other appropriate calculation method. The biomechanical support index table is represented as B={(zoneᵢ, indexᵢ)|i=1,2,…,M}, where zoneᵢ represents the zone type, indexᵢ represents the biomechanical support index, and M represents the total number of zones.

[0087] Preferably, the personalized pillow support recommendation in step S4 includes: According to the biomechanical support index table and the curvature recovery adaptation factor, the pillow morphological parameters are optimized to obtain a morphological parameter combination; According to the combination of morphological parameters and the biomechanical support index table, pillow materials are combined to obtain a material combination plan; Calculate the personalized adjustment parameter table based on the material combination scheme and morphological parameter combination; Comprehensive recommended parameters are generated based on the morphological parameter combination, material combination scheme and personalized adjustment parameter table to obtain the recommended support parameter set.

[0088] In an embodiment of the present invention, the pillow morphological parameters are optimized according to the biomechanical support index table and the curvature recovery adaptation factor to obtain a morphological parameter combination. The pillow morphological parameters are determined according to the biomechanical support index table B={(zoneᵢ, indexᵢ)|i=1, 2, ..., M} and the curvature recovery adaptation factor. The pillow morphological parameters include height, curvature, edge transition angle and support area width. Optimization of height: Determine the optimal height range according to the biomechanical support index. The optimal height is related to the cervical curvature, the head-neck ratio and the sleeping posture. The height range is set to: minimum value, maximum value, step value. Optimization of curvature: Determine the optimal curvature according to the cervical curvature and pressure distribution. The curvature can be expressed using a curve equation, such as a Bezier curve. Optimization of edge transition angle: Determine the optimal edge transition angle according to the shoulder structure and sleeping posture. Optimization of support area width: Determine the optimal support area width according to the cervical length and sleeping posture. The optimization algorithm can use a genetic algorithm, a particle swarm algorithm or a gradient descent algorithm. The morphological parameter combination is expressed as O={(height, curvature, edge_angle, width)}.

[0089] Based on the morphological parameter combination and the biomechanical support index table, pillow material combinations are determined to obtain a material combination plan. The pillow material combination is determined based on the morphological parameter combination O = {(height, curvature, edge_angle, width)} and the biomechanical support index table B = {(zoneᵢ, indexᵢ)|i=1, 2, …, M}. The pillow is divided into different zones, such as the head support zone, the neck support zone, and the transition buffer zone. For each zone, appropriate materials are selected.

[0090] Material selection is based on the material property parameter library P = {(materialᵢ, hardnessᵢ, rebound_timeᵢ, densityᵢ, air_permeabilityᵢ, thermal_conductivityᵢ, durabilityᵢ, compressive_strengthᵢ)|i=1,2,…,N} and the rebound property evaluation table T = {(materialᵢ, pressureᵢ, rebound_timeᵢ, deformationᵢ)|i=1,2,…,N}. The material combination scheme includes material type and combination ratio. The material combination scheme is represented as C = {(zoneᵢ, materialᵢ, ratioᵢ)|i=1,2,…,M}, where zoneᵢ represents the zone type, materialᵢ represents the material type, and ratioᵢ represents the material ratio.

[0091] A personalized adjustment parameter table is calculated based on the material combination scheme and morphological parameter combination. The personalized adjustment parameters are calculated based on the material combination scheme C = {(zoneᵢ, materialᵢ, ratioᵢ)|i=1, 2, …, M} and the morphological parameter combination O = {(height, curvature, edge_angle, width)}. Personalized adjustment parameters take into account individual differences, such as head shape, sleeping habits, and physical conditions. Head shape: Adjusts pillow height and curvature based on head shape. Sleeping habits: Adjusts pillow support and material based on sleeping posture. Physical conditions: Adjusts pillow material based on allergy history. Personalized adjustment parameters can be expressed as coefficients or offsets. The personalized adjustment parameter table is represented as A = {(parameterᵢ, adjustmentᵢ)|i=1, 2, …, K}, where parameterᵢ represents the parameter type, adjustmentᵢ represents the adjustment value, and K represents the total number of parameters.

[0092] Based on the morphological parameter combination, material combination scheme and personalized adjustment parameter table, comprehensive recommended parameters are generated to obtain the recommended support parameter set. The morphological parameter combination O, material combination scheme C and personalized adjustment parameter table A are integrated to generate comprehensive recommended parameters. Comprehensive recommended parameters include: recommended pillow height range, material type and combination ratio, hardness zoning design, morphological design parameters and usage recommendations. Recommended pillow height range: Determine the recommended height range based on the morphological parameter combination and personalized adjustment parameters. Material type and combination ratio: Determine the recommended material type and combination ratio based on the material combination scheme. Hardness zoning design: Design the hardness zoning based on the biomechanical support index and personalized adjustment parameters. Morphological design parameters: Determine the morphological design parameters of the pillow based on the morphological parameter combination and personalized adjustment parameters. Usage recommendations: Give usage recommendations based on material characteristics and individual differences, such as replacement cycle, cleaning method, etc.

[0093] The recommended support parameter set is expressed as R = {(height_range, material_composition, hardness_zone, shape_parameters, usage_advice)}.

[0094] It is particularly important that the pillow shape parameters are optimized as follows: The functional areas are divided and set according to the biomechanical support index table to obtain the functional area size table; Calculate the area height parameter set according to the functional area size table; Constructing a contour curve equation group based on the regional height parameter set; According to the contour curve equation group, the edge transition angle of the functional area size table is optimized to obtain the edge transition angle table; Posture adaptability was assessed based on the edge transition angle table to obtain a postural adaptability score sheet; Adjust the age segmentation parameters according to the posture adaptability score table to obtain the age segmentation parameter table; Calculate individual difference compensation values ​​according to the age segment parameter table; The edge transition angle table, posture adaptability score table and individual difference compensation value were integrated to obtain the morphological parameter combination; In an embodiment of the present invention, functional area division and setting are performed according to a biomechanical support index table to obtain a functional area size table. According to the biomechanical support index table B={(zoneᵢ, indexᵢ)|i=1, 2, ..., M}, the pillow is divided into functional areas. The functional areas include a head support area, a neck support area, and a shoulder transition area. According to the support index, the size of each area is determined. Head support area: its width and length are determined according to the weight and pressure distribution of the head. Neck support area: its width and height are determined according to the length and curvature of the cervical spine. Shoulder transition area: its width and angle are determined according to the shoulder width and sleeping posture. The functional area size table is expressed as S={(zoneᵢ, widthᵢ, lengthᵢ, heightᵢ)|i=1, 2, ..., M}, where zoneᵢ represents the area type, widthᵢ represents the width, lengthᵢ represents the length, and heightᵢ represents the height.

[0095] Calculate the zone height parameter set based on the functional zone size table. This table, S = {(zoneᵢ, widthᵢ, lengthᵢ, heightᵢ) | i = 1, 2, …, M}, calculates the zone height parameter set. The head support zone height is determined based on head weight and pressure distribution. The neck support zone height is determined based on cervical curvature and support requirements. The shoulder transition zone height is determined based on shoulder structure and sleeping posture. Calculate the height of each zone and set a height range. The zone height parameter set is represented as H = {(zoneᵢ, height_minᵢ, height_maxᵢ, height_stepᵢ) | i = 1, 2, …, M}, where zoneᵢ represents the zone type, height_minᵢ represents the minimum height, height_maxᵢ represents the maximum height, and height_stepᵢ represents the height step.

[0096] Construct a set of contour curve equations based on the zone height parameter set. Construct the contour curve equation based on the zone height parameter set H = {(zoneᵢ, height_minᵢ, height_maxᵢ, height_stepᵢ)|i = 1, 2, …, M}. The contour curve equation describes the shape of the side of the pillow. Use Bezier curves or spline curves to construct the contour curve equation. The control points of the Bezier curve are determined based on the zone height parameters. Construct multiple curve equations corresponding to different side shapes. The contour curve equation group is expressed as C = {C1(x), C2(x), …C n (x)}, where Cᵢ(x) represents the equation of the i-th curve.

[0097] According to the contour curve equation group, the edge transition angle of the functional area size table is optimized to obtain the edge transition angle table. According to the contour curve equation group C={C1(x), C2(x),…C n (x)} and the functional area size table S = {(zoneᵢ, widthᵢ, lengthᵢ, heightᵢ) | i = 1, 2, …, M}, optimize the edge transition angle. The edge transition angle refers to the connection angle between different areas of the pillow. The edge transition angle is changed by adjusting the control points of the curve. The optimization goal is to match the edge transition angle with the shoulder structure and sleeping position. Optimization is performed using an optimization algorithm such as a genetic algorithm or a particle swarm algorithm. The edge transition angle table is represented as A = {(zoneᵢ, angleᵢ) | i = 1, 2, …, M}, where zoneᵢ represents the area type and angleᵢ represents the edge transition angle.

[0098] Postural adaptability is assessed based on the edge transition angle table, resulting in a postural adaptability score table. The postural adaptability of a pillow is evaluated based on the edge transition angle table A = {(zoneᵢ, angleᵢ)|i=1, 2, …, M}. Postural adaptability refers to the pillow's support for different sleeping positions. Assessment methods include simulating pressure distribution and cervical curvature under different sleeping positions. Finite element analysis (FEA) or experimental measurement methods are used. FEA simulations involve building a three-dimensional model of the pillow and applying pressure and boundary conditions. Experimental measurements involve having children use the pillow and measuring pressure distribution and cervical curvature. Scoring criteria are established for different sleeping positions. The postural adaptability score table is represented as P = {(poseᵢ, scoreᵢ)|i=1, 2, …, N}, where poseᵢ represents the posture type and scoreᵢ represents the score.

[0099] Age-specific parameter adjustments are performed based on the posture adaptability score table, resulting in an age-specific parameter table. Age-specific parameter adjustments are performed based on the posture adaptability score table P = {(poseᵢ, scoreᵢ)|i=1, 2, …, N}. The pillow's morphological parameters, such as height, curvature, and firmness, are adjusted based on the child's age group. The posture adaptability score weights are adjusted based on the age group. The age-specific parameter table is represented as T = {(ageᵢ, parameterᵢ, valueᵢ)|i=1, 2, …, A}, where ageᵢ represents the age group, parameterᵢ represents the parameter type, valueᵢ represents the parameter value, and A represents the total number of age groups.

[0100] Calculate individual difference compensation values ​​based on the age-specific parameter table. Compensation values ​​are calculated based on individual differences. These factors include head shape, sleeping habits, and specific physical conditions. Head shape: Adjust pillow height and curvature based on head shape. Sleeping habits: Adjust pillow support and material based on sleeping position. Specific physical conditions: Adjust pillow material based on allergy history. Compensation values ​​can be expressed as coefficients or offsets. Calculation method: Modify age-specific parameters based on individual differences.

[0101] Integrate the edge transition angle table, the posture adaptability score table, and the individual difference compensation value to obtain a morphological parameter combination. The edge transition angle table A, the posture adaptability score table P, and the individual difference compensation value are integrated. The morphological parameter combination includes height, curvature, edge transition angle, and support zone width. The morphological parameter combination is represented as O = {(height_range, curvature, edge_angle, width)}.

[0102] Preferably, the present invention further provides a home textile recommendation system based on child health data monitoring and analysis, which is used to execute the above-mentioned home textile recommendation method based on child health data monitoring and analysis. The home textile recommendation system based on child health data monitoring and analysis includes: The physical sign three-dimensional modeling module is used to collect the original point cloud coordinate set of the child's sleeping posture; extract the cervical curvature parameter group based on the cervical curvature parameter group of the extracted skeletal key point spatial distribution map; perform cervical vertebra and body position analysis based on the cervical curvature parameter group to obtain cervical curvature-body position holographic data; The partitioned pressure mapping module is used to collect a static pressure baseline data set based on the cervical curvature-posture holographic data; perform posture-pressure correlation analysis based on the static pressure baseline data set to obtain a posture-pressure mapping relationship table; and generate a pressure distribution index map based on the posture-pressure mapping relationship table; The curvature recovery modeling module is used to quantify the curvature-support effect based on the pressure distribution index map to obtain the curvature deviation index set; the pressure-curvature response relationship between the curvature deviation index set and the pressure distribution index map is analyzed to obtain the curvature recovery adaptation factor; The support matching reasoning module is used to obtain pillow material performance data; calculate the biomechanical support index table based on the pillow material performance data and the curvature recovery adaptation factor; and make personalized pillow support recommendations based on the biomechanical support index table to obtain a recommended support parameter set.

[0103] The present invention is therefore intended to be illustrative and non-restrictive in all respects, with the scope of the invention being defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the application documents are intended to be embraced therein.

[0104] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be construed in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. A home textile recommendation method based on children's health data monitoring and analysis, characterized in that: The following steps are involved: Step S1: collecting the original point cloud coordinate set of the child's sleeping posture; extracting the cervical curvature parameter group based on the cervical curvature parameter group of the extracted skeletal key point spatial distribution map; performing cervical vertebra and body position analysis based on the cervical curvature parameter group to obtain cervical curvature-body position holographic data; Step S2: collecting a static pressure baseline data set based on the cervical vertebra curvature-posture holographic data; performing posture-pressure correlation analysis based on the static pressure baseline data set to obtain a posture-pressure mapping relationship table; generating a pressure distribution index map based on the posture-pressure mapping relationship table; Step S3: quantifying the curvature-support effect according to the pressure distribution index map to obtain a curvature deviation index set; analyzing the pressure-curvature response relationship between the curvature deviation index set and the pressure distribution index map to obtain a curvature recovery adaptation factor; Step S4: Obtaining pillow material performance data; calculating a biomechanical support index table based on the pillow material performance data and the curvature recovery adaptation factor; Personalized pillow support recommendations are made based on the biomechanical support index table to obtain a recommended support parameter set.

2. The home textile recommendation method based on children's health data monitoring and analysis according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: capturing the child's natural sleeping posture through a high-speed camera and performing multi-angle optical scanning on the child's natural sleeping posture to obtain an original point cloud coordinate set; Step S12: positioning the skeleton key points of the original point cloud coordinate set and drawing a spatial distribution map of the skeleton key points; Step S13: extracting a cervical curvature parameter group from the spatial distribution map of the skeleton key points; Step S14: performing head-neck ratio measurement and analysis based on the spatial distribution map of the key bone points and the cervical curvature parameter group to obtain head-neck mechanical ratio data; Step S15: Based on the cervical curvature parameter group and the head-neck mechanical ratio data, multi-dimensional angle parameters of the angle between the head and the bed surface, the angle between the cervical spine and the shoulder, and the angle between the cervical spine axis and the horizontal plane are measured to construct a multi-dimensional posture angle matrix; Step S16: performing holographic data integration and standardization on the cervical curvature parameter group and the multi-dimensional posture angle matrix to obtain cervical curvature-posture holographic data.

3. The home textile recommendation method based on children's health data monitoring and analysis according to claim 2, characterized in that: The cervical curvature parameter group of the skeleton key point spatial distribution map extracted in step S13 includes: Extract the cervical vertebra centerline trajectory based on the spatial distribution map of skeletal key points; The multi-plane curvature of the cervical vertebrae centerline trajectory is calculated to obtain a three-plane curvature distribution map; Measure the key segment angle values ​​of the three-plane curvature distribution diagram; The curve length and morphology are measured based on the cervical spine centerline trajectory and key segment angle values ​​to obtain the curve morphology feature set; Perform standard curve deviation analysis on the curve morphology feature set to obtain the curvature standard deviation value; The stress distribution is estimated based on the curvature standard deviation value to obtain the cervical vertebra stress distribution map; The dynamic change characteristics of the curvature are obtained by measuring the dynamic change characteristics based on the cervical vertebra centerline trajectory and the curve morphology feature set; The cervical curvature parameter group was constructed based on the three-plane curvature distribution diagram, key segment angle values, curve morphological feature set, curvature standard deviation value, cervical stress distribution diagram and curvature dynamic change characteristics.

4. The home textile recommendation method based on children's health data monitoring and analysis according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: accurately measuring the contact area based on the cervical vertebrae curvature-posture holographic data to obtain a contact area division map; Step S22: collecting a static pressure baseline data set according to the contact area division map; Step S23: monitoring the dynamic pressure change sequence according to the static pressure baseline data set and the contact area division map; Step S24: Calculating pressure uniformity based on the dynamic pressure change sequence and the static pressure baseline data set to obtain a pressure uniformity distribution matrix; Step S25: performing posture-pressure correlation analysis on the dynamic pressure change sequence and the cervical vertebrae curvature-posture holographic data to obtain a posture-pressure mapping relationship table; Step S26: performing pressure peak distribution analysis based on the body position-pressure mapping relationship table and the pressure uniformity distribution matrix to obtain a pressure peak distribution characteristic diagram; Step S27: Calculate the pressure distribution index map according to the pressure uniformity distribution matrix, the body position-pressure mapping relationship table and the pressure peak distribution characteristic map.

5. The home textile recommendation method based on children's health data monitoring and analysis according to claim 4, characterized in that: The body position-pressure correlation analysis in step S25 includes: Performing time series synchronization processing on the dynamic pressure change sequence and the cervical curvature-posture holographic data to obtain a time-synchronized data pair; Perform spatial coordinate mapping transformation on the time-synchronized data pair to obtain a spatial corresponding mapping matrix; Extract the standard body posture state from the spatial corresponding mapping matrix and the time synchronization data to obtain the standard body posture pressure characteristics; Calculate the pressure transfer trajectory according to the standard body position pressure characteristics to obtain the pressure transfer trajectory set; The pressure response delay is measured on the pressure transfer trajectory set to obtain the regional pressure response delay table; Calculate the body position change influence coefficient table based on the regional pressure response time delay table and the pressure transfer trajectory set; Construct a body position-pressure mapping relationship table based on the body position change influence coefficient table.

6. The home textile recommendation method based on children's health data monitoring and analysis according to claim 1, characterized in that: The curvature-support effect quantification in step S3 includes: The age staging standard was established based on the pressure distribution index map and cervical curvature-posture holographic data to obtain the reference standard for cervical development stages; The ideal curvature model was constructed based on the cervical development stage reference standard and cervical curvature-posture holographic data to obtain the ideal curvature model for age stages. The actual support effect is measured based on the ideal curvature model of age stages and the pressure distribution index map to obtain the dynamic curvature change trajectory; The curvature deviation was quantitatively analyzed between the dynamic curvature change trajectory and the ideal curvature model of age stages, and a set of curvature deviation indices was obtained.

7. The home textile recommendation method based on children's health data monitoring and analysis according to claim 1, characterized in that: The analysis of the pressure-curvature response relationship in step S3 includes: Perform high-frequency synchronous data acquisition on the pressure distribution index map to obtain pressure-curvature synchronous time series data; perform time window segmentation processing on the pressure-curvature synchronous time series data to obtain a windowed change rate data set; The key pressure areas are divided according to the windowed change rate data set and the pressure distribution index map, and the pressure characteristic table of the functional areas is obtained; According to the functional area pressure characteristic table, regional sensitivity measurement is carried out to obtain the regional curvature sensitivity coefficient table; Identify critical pressure points based on the regional curvature sensitivity coefficient table and the functional area pressure characteristic table to obtain a critical pressure point location map; According to the critical pressure point position diagram, the pressure-curvature mapping function is constructed to obtain a point-curve mapping function set; The time delay characteristics are analyzed based on the point-curve mapping function set to obtain the regional response delay table; Generate a pressure-curvature response matrix based on the regional response time delay table; The recovery score is calculated based on the pressure-curvature response matrix and the curvature deviation index set to obtain the curvature recovery score table; The adaptation factor is comprehensively generated according to the curvature recovery score table and the pressure-curvature response matrix to obtain the curvature recovery adaptation factor.

8. The home textile recommendation method based on children's health data monitoring and analysis according to claim 1, characterized in that: Calculating the biomechanical support index table in step S4 includes: Construct a material performance parameter library based on pillow material performance data; perform material rebound response analysis based on the material performance parameter library and curvature recovery adaptation factor to obtain a material rebound characteristic evaluation table; Collect the head and neck biomechanical parameter set according to the curvature recovery adaptation factor; calculate the regional ideal support force table according to the head and neck biomechanical parameter set; Measuring material support force curve sets based on regional ideal support force tables; Calculate the regional support matching index based on the material support force curve set; Posture difference analysis is performed based on the regional support matching index to obtain the posture support correction coefficient; Monitor and analyze sleeping posture ratios based on posture support correction coefficients to obtain a sleeping posture time ratio table; Evaluate the support stability coefficient based on the sleeping posture time percentage table; The comprehensive support index is calculated based on the regional support matching index, posture support correction coefficient, sleeping posture time ratio table and support stability coefficient to obtain the biomechanical support index table.

9. The home textile recommendation method based on children's health data monitoring and analysis according to claim 1, characterized in that: The personalized pillow support recommendation in step S4 includes: According to the biomechanical support index table and the curvature recovery adaptation factor, the pillow morphological parameters are optimized to obtain a morphological parameter combination; According to the combination of morphological parameters and the biomechanical support index table, pillow materials are combined to obtain a material combination plan; Calculate the personalized adjustment parameter table based on the material combination scheme and morphological parameter combination; Comprehensive recommended parameters are generated based on the morphological parameter combination, material combination scheme and personalized adjustment parameter table to obtain the recommended support parameter set.

10. A home textile recommendation system based on children's health data monitoring and analysis, characterized in that: For executing the home textile recommendation method based on child health data monitoring and analysis as claimed in claim 1, the home textile recommendation system based on child health data monitoring and analysis comprises: The physical sign three-dimensional modeling module is used to collect the original point cloud coordinate set of the child's sleeping posture; extract the cervical curvature parameter group based on the cervical curvature parameter group of the extracted skeletal key point spatial distribution map; perform cervical vertebra and body position analysis based on the cervical curvature parameter group to obtain cervical curvature-body position holographic data; The partitioned pressure mapping module is used to collect a static pressure baseline data set based on the cervical curvature-posture holographic data; perform posture-pressure correlation analysis based on the static pressure baseline data set to obtain a posture-pressure mapping relationship table; and generate a pressure distribution index map based on the posture-pressure mapping relationship table; The curvature recovery modeling module is used to quantify the curvature-support effect based on the pressure distribution index map to obtain the curvature deviation index set; the pressure-curvature response relationship between the curvature deviation index set and the pressure distribution index map is analyzed to obtain the curvature recovery adaptation factor; The support matching reasoning module is used to obtain pillow material performance data; calculate the biomechanical support index table based on the pillow material performance data and the curvature recovery adaptation factor; and make personalized pillow support recommendations based on the biomechanical support index table to obtain a recommended support parameter set.

Images