Early warning system for risk of peri-implantitis
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HAIKOU PEOPLES HOSPITAL
- Filing Date
- 2026-05-21
- Publication Date
- 2026-06-19
Smart Images
Figure CN122245791A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of personal health risk assessment technology, specifically to an early warning system for the risk of peri-implantitis. Background Technology
[0002] Dental implant restoration is the preferred treatment for patients with missing teeth, but peri-implantitis leading to bone resorption is a major cause of long-term failure. Currently, clinical monitoring of peri-implantitis risk mainly relies on regular X-ray imaging. Existing monitoring standards are usually based on measuring the bone integration height (i.e., geometric height) on both sides of the implant neck in images. When the average annual bone resorption is less than a certain threshold (e.g., 0.2 mm), it is considered stable. However, this monitoring method based solely on macroscopic geometric height has significant technical blind spots. Pathologically, peri-implant bone tissue destruction often begins with changes in microstructure, namely a decrease in trabecular bone density (osteoporosis). This decrease in micro-density weakens the bone tissue's mechanical support for the implant, causing the effective mechanical support center of gravity to shift downwards towards the root. At this point, although there may not be a visible decrease in macroscopic bone height on imaging, the actual mechanical load borne by the implant neck, as the stress fulcrum, has already increased significantly due to insufficient support stiffness. Existing methods for early warning of peri-implantitis risk, such as relying solely on grayscale mean calculation or edge detection, cannot analyze microscopic density changes and their spatial distribution characteristics, which affects the accuracy of early warning results before macroscopic bone destruction occurs. Summary of the Invention
[0003] To address the technical problem of low accuracy in early warning methods for peri-implantitis risk, the present invention aims to provide an early warning system for peri-implantitis risk, the specific technical solution of which is as follows: This invention provides an early warning system for the risk of peri-implantitis, comprising: The data acquisition module is used to acquire slices of bone tissue around the implant along the implant axis. The early warning module, which is electrically connected to the data acquisition module, is configured as follows: Based on the grayscale values of each slice region, the bone mineral density intensity characteristics of each slice region are obtained; The location of the bone interface of the implant is obtained based on the growth of bone density intensity characteristics in adjacent slice regions. The deviation between the cumulative value of bone mineral density characteristics up to each slice region and the corresponding ideal cumulative value of bone mineral density characteristics is used to obtain the amount of implant center of gravity shift. By integrating the center of gravity shift, geometric cantilever features, and absolute bone mineral density features, a composite risk index for the implant is obtained; the geometric cantilever features are obtained from the bone interface position, and the absolute bone mineral density features are obtained from the total gray level of all slice regions; Early warning of peri-implantitis risk using a composite risk index.
[0004] In an exemplary embodiment, the process of obtaining each slice region includes: The length of each slice region is obtained based on the axial length of the implant and the preset number of slice regions. Based on the preset width and length of each slice region, the bone tissue around the implant is sliced to obtain each slice region.
[0005] In an exemplary embodiment, the process of obtaining the bone mineral density strength feature includes: Obtain the average grayscale value of each slice region; The ratio of the mean gray level of each slice region to the total gray level of all slice regions is calculated to obtain the bone mineral density characteristics of each slice region; the total gray level of all slice regions is the sum of the mean gray levels of all slice regions.
[0006] In an exemplary embodiment, the process of obtaining the bone interface location includes: A first-order difference sequence of bone mineral density intensity features is obtained, wherein the bone mineral density intensity features of each slice region are sorted in the direction from the neck of the implant to the root. The maximum first-order difference value is determined from the target slice region range in the first-order difference sequence; the target slice region range includes the first preset percentage of slice regions selected in the direction from the neck to the root of the implant; The location of the slice region corresponding to the maximum first-order difference value is determined as the location of the bone interface.
[0007] In an exemplary embodiment, the process of obtaining the ideal cumulative value of the bone mineral density strength feature includes: An initial bone mineral density intensity characteristic reference value is set for each slice region; wherein, the initial bone mineral density intensity characteristic reference value of each slice region before the bone interface position is set as a first value, and the initial bone mineral density intensity characteristic reference value of each slice region at and after the bone interface position is set as a second value; the first value is less than the second value. The ideal value of bone mineral density (BMD) for each slice region is obtained by calculating the ratio of the initial reference value of BMD for each slice region to the sum of the initial reference values of BMD for all slice regions. Based on the ideal values of bone mineral density intensity characteristics in each slice region, the cumulative ideal values of bone mineral density intensity characteristics up to each slice region are obtained.
[0008] In an exemplary embodiment, the process of obtaining the downward shift of the center of gravity includes: The Manhattan distance between the cumulative bone mineral density feature sequence and the ideal cumulative bone mineral density feature sequence is determined to obtain the centroid shift amount; the cumulative bone mineral density feature sequence is obtained by sorting the cumulative bone mineral density feature values up to each slice region; the ideal cumulative bone mineral density feature sequence is obtained by sorting the ideal cumulative bone mineral density feature values up to each slice region.
[0009] In an exemplary embodiment, the process of obtaining the absolute bone density feature includes: calculating the ratio of the total gray level of all slice regions to the preset standard total gray level as the absolute bone density feature; The composite risk index is positively correlated with the center of gravity shift and geometric cantilever characteristics, and negatively correlated with the absolute bone mineral density characteristics.
[0010] In one exemplary embodiment, the early warning of peri-implantitis risk using a composite risk index includes: The relative severity of the composite risk index is obtained by calculating the ratio of the composite risk index to the preset stable value of the composite risk index. Determine the preset risk range in which the relative severity falls, and output an early warning signal for the risk of peri-implantitis.
[0011] In an exemplary embodiment, the process of obtaining the preset risk range includes: The range below the first preset risk threshold is defined as the low-risk range; The range that is greater than or equal to the first preset risk threshold and less than the second preset risk threshold is defined as the medium-risk range. The intervals that are greater than or equal to the second preset risk threshold are defined as high-risk intervals; the first preset risk threshold is less than the second preset risk threshold.
[0012] In an exemplary embodiment, the process of obtaining the preset composite risk index stable value includes: calculating the average value of multiple composite risk indices obtained within a preset historical stable period, and using it as the preset composite risk index stable value.
[0013] This invention offers the following advantages: By acquiring image information of various sliced regions of key bone tissue surrounding the implant, it lays the foundation for analyzing the spatial distribution of micro-density. Through axial slicing, the continuous bone tissue is discretized into a series of tiny analytical units, providing a data basis for subsequently capturing changes in bone density intensity from the neck to the root. Bone density intensity features quantify the bone density of each micro-slice unit, converting grayscale values into bone density intensity features. This transforms micro-density changes, which are previously difficult to distinguish with the naked eye or ignored by traditional algorithms, into calculable and comparable mathematical variables. Utilizing the axial growth of density features to determine the interface allows for a more stable and accurate identification of the transition point from non-bone tissue (low density) to bone tissue (high density), reducing noise interference. By accumulating the actual bone density intensity feature values and... By comparing the ideal cumulative values, the deviation between the two is determined, thus obtaining the implant's center of gravity shift. This shift is closely related to bone density uniformity, mapping microscopic density distribution changes to macroscopic mechanical support performance indicators. The center of gravity shift represents the deterioration of mechanical support caused by microstructural degradation, the geometric cantilever characteristic represents the current macroscopic bone resorption level (static geometry), and the absolute bone density characteristic represents the overall bone density level. By integrating data from these three aspects, the composite risk index comprehensively considers the macroscopic resorption that has already occurred, the mechanical failure caused by impending microstructural degradation, and the overall bone quality. This allows the composite risk index to reflect the true health status of the implant more comprehensively and accurately than any single indicator. Early warning of peri-implantitis risk using the composite risk index can significantly improve the accuracy and reliability of early warning of peri-implantitis risk in oral implants. Attached Figure Description
[0014] Figure 1 This is a schematic diagram of the module composition of an early warning system for the risk of peri-implantitis provided by the present invention; Figure 2 This is a data processing flowchart of the early warning module provided by the present invention. Detailed Implementation
[0015] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the specific implementation methods, structures, features, and effects of the present invention are described in detail below with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.
[0016] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. All data and information collected in this application have been obtained with full consent.
[0017] This embodiment provides an early warning system for the risk of peri-implantitis in the oral cavity, which will be referred to as the system for ease of explanation. This system is used to provide early warning of the risk of peri-implantitis in the oral cavity of a monitored individual.
[0018] like Figure 1 As shown, the system includes a data acquisition module and an early warning module. The data acquisition module and the early warning module are electrically connected. In terms of hardware configuration, the connection between the data acquisition module and the early warning module can be wired, such as through a data transmission line, or wireless (such as Bluetooth, WiFi, etc.). Moreover, the data acquisition module and the early warning module can also be integrated to form an integrated system.
[0019] This embodiment requires acquiring implant images, specifically X-ray images of the implant (the X-ray images are grayscale images), which are acquired using conventional X-ray image acquisition equipment. In addition to the image area of the implant itself, the implant image also includes a certain range of bone tissue surrounding the implant, facilitating the sectioning of the surrounding bone tissue.
[0020] The data acquisition module acquires implant images and slices the surrounding bone tissue in the implant images to obtain various slice regions of the surrounding bone tissue. Since there may be uncertainties in the shooting angle and distance when capturing implant images of different frames, this embodiment needs to clearly define the division method of the slice regions and perform slicing along the implant's axial direction. The axial direction refers to the distance from the implant's neck (i.e., the stress initiation end) to the root (i.e., the implant's root tip). In an exemplary embodiment, this embodiment performs edge detection processing on the implant image (e.g., using the Canny operator) to identify the implant's metal contour with high radiation-blocking properties. Then, based on the contour's geometric features, it fits the implant's central long axis and defines the longitudinal coordinate axis (i.e., the y-axis) along this long axis direction as the two-dimensional coordinate system. The y-axis direction is the axial direction, pointing from the implant's neck to the implant's root. A direction perpendicular to the long axis direction is defined as the transverse coordinate axis (i.e., the x-axis), thus obtaining the implant's two-dimensional coordinate system.
[0021] To standardize the number of slice regions obtained from implant images across different frames, this embodiment pre-obtains the actual physical length and diameter of the implant (unit: mm), for example, by consulting relevant product descriptions for the same implant model. Then, this embodiment sets a globally uniform standardized slice thickness, which defines the minimum physical resolution for system analysis. The specific value is set according to actual needs; in this embodiment, it is set to 0.1 mm. Based on the actual physical length of the implant and the standardized slice thickness, the number of slice regions is obtained. Specifically, the ratio of the actual physical length of the implant to the standardized slice thickness is calculated, and the result is the number of slice regions. For example, if the actual physical length of a certain implant model is 10 mm and the standardized slice thickness is 0.1 mm, then the number of slice regions is 10 / 0.1 = 100, meaning there are 100 slice regions. This calculation method ensures that regardless of the actual resolution of the acquired implant image, the entire length of the implant is ultimately divided into 100 slice regions representing the same physical scale.
[0022] To cover the critical bone tissue surrounding the implant, this embodiment defines a region of interest (ROI). The lateral width of the ROI is a preset multiple of the implant diameter, which is the actual physical diameter of the implant obtained beforehand. The primary purpose of this preset multiple is to cover the effective bone tissue providing mechanical support around the implant. If the preset multiple is set too large, the lateral width of the ROI will be too large, introducing density interference from adjacent natural teeth or unrelated jawbones. If the preset multiple is set too small, the lateral width of the ROI will be too small, failing to completely encompass the osseointegration area. In this embodiment, the preset multiple ranges from 0.5 to 1.0, for example, a value of 0.6, used to precisely cover the effective trabecular bone region on the side of the implant.
[0023] The axial length of the contour obtained by edge detection processing of the implant image is acquired. This axial length represents the total length of the implant from the top of the neck to the bottom of the root tip. The ratio of this axial length to the number of slice regions obtained above is calculated, and the result is the axial length of each slice region. Therefore, all slice regions have the same length.
[0024] Following the direction from the neck of the implant towards the root, the indices of each slice region are determined sequentially. If the number of slice regions is set to L, then the index range of the slice regions is... The index of the slice area gradually increases according to the direction from the neck of the implant to the root of the implant. In terms of the vertical direction, the neck of the implant represents the upper end of the implant, and the root of the implant represents the lower end of the implant. The direction from the neck of the implant to the root of the implant is from top to bottom.
[0025] For any slice region, with the first Taking a slice region as an example ( ), No. Each slice region comprises several rows of horizontal pixels, the number of rows being equal to the length of the slice region. For any row, based on the implant image, starting from the edge pixels of the metal edge contour lines on the left and right sides of the implant, the horizontal width of the region of interest is extended outwards to obtain the bilateral extended pixels of that row, which serve as the horizontal pixels of the slice region for that row. This process yields the 1st slice region. The horizontal pixels of each row of the slice region in the slice region will be the first... The horizontal pixels of all rows in the slice region are integrated as the first slice region. This allows for the creation of individual slices of the bone tissue surrounding the implant.
[0026] The early warning module receives slices of peri-implant bone tissue from the data acquisition module and performs subsequent data processing. The early warning module can be a standard data processing chip, such as a CPU (Central Processing Unit), which can be configured within an oral healthcare monitoring platform.
[0027] Figure 2 The implementation steps of the data processing process for the early warning module are as follows: Figure 2 As shown, the data processing steps of the early warning module include: Step S1: Obtain the bone mineral density intensity characteristics of each slice region based on the grayscale values of each slice region; Step S2: Based on the growth of bone density intensity characteristics in adjacent slice regions, the location of the bone interface of the implant is obtained; Step S3: Combine the deviation between the cumulative value of bone mineral density characteristics up to each slice region and the corresponding ideal cumulative value of bone mineral density characteristics to obtain the amount of implant center of gravity shift; Step S4: Integrate the center of gravity shift, geometric cantilever characteristics, and absolute bone mineral density characteristics to obtain the composite risk index of the implant; Step S5: Early warning of peri-implantitis risk using the composite risk index.
[0028] The following is a detailed explanation of each step.
[0029] Step S1: Obtain the bone density intensity characteristics of each slice region based on the grayscale values of each slice region.
[0030] As is common knowledge, in X-ray images, higher gray values indicate more bone tissue and higher bone density, while lower gray values indicate more soft tissue. Therefore, the bone density characteristics of each slice region can be obtained based on its gray values.
[0031] With the first Taking the nth slice region as an example, calculate the nth slice region. The average grayscale value of pixels within the nth slice region is used as the nth slice region. The higher the mean gray level of each slice region, the higher the bone mineral density (BMD) characteristic. However, due to variations in overall X-ray exposure (such as voltage fluctuations or different development times), the absolute magnitudes of gray levels may lack direct comparability. Therefore, the mean gray level of each slice region must be normalized. In an exemplary embodiment, the sum of the mean gray levels of all slice regions is calculated as the total gray level of all slice regions. Then, the ratio of the mean gray level of each slice region to the total gray level of all slice regions is calculated. The result is the BMD characteristic of each slice region. ; ; in, Indicates the first The average gray level of each slice region This represents the total gray level of all sliced regions. Indicates the first The bone mineral density characteristics of each slice region are calculated as follows: After the above calculations, the sum of the bone mineral density characteristics of all slice regions is equal to 1.
[0032] It should be understood that since soft tissue background pixels may be retained within the sliced area, the grayscale mean of the sliced area may be mixed with the grayscale values of some soft tissue background pixels during calculation. Therefore, the normalized bone mineral density intensity feature will contain values corresponding to certain soft tissue background pixels, which may be miscalculated as bone loss to some extent, causing processing errors. In other embodiments, this example can also preset a soft tissue background grayscale threshold. This soft tissue background grayscale threshold is used as a criterion for judging whether the sliced area is a soft tissue background. It can be obtained in the following way: obtain the soft tissue background area in several implant images within a historical time period (e.g., two months), for example, by manually marking the soft tissue background area, and then calculate the average grayscale value of the soft tissue background area in the several implant images, using this average value as the soft tissue background grayscale threshold.
[0033] The grayscale mean of each slice region is compared with the grayscale threshold of the soft tissue background. The grayscale mean of slice regions that are less than or equal to the soft tissue background grayscale threshold is forcibly set to 0, thus obtaining the grayscale mean of each slice region after this processing. The total grayscale level of all slice regions is then calculated based on this, thereby obtaining the bone mineral density characteristic of each slice region. It should be understood that for slice regions whose grayscale mean is forcibly set to 0, their bone mineral density characteristic is also 0.
[0034] In addition, to prevent calculation errors caused by image acquisition errors (such as completely black images or extremely low contrast), this embodiment can also introduce a preset minimum positive threshold (the minimum positive threshold represents the gray value, for example, gray value 2). The minimum positive threshold is used to represent that the image is completely black. The total gray level of all slice areas is compared with the value of the minimum positive threshold. If the total gray level of all slice areas is less than or equal to the minimum positive threshold, the currently acquired implant image is determined to be invalid, the subsequent process is terminated and a prompt indicating that the image quality is unqualified is output; otherwise, the subsequent steps are continued.
[0035] In this embodiment, the absolute bone mineral density (BMD) feature is obtained based on the total gray level of all slice regions. The absolute BMD feature reflects the absolute quantitative information of the overall degree of osteoporosis. In an exemplary embodiment, a standard total gray level is preset. This standard total gray level is a reference value for the standard total gray level of all slice regions. This standard total gray level can be a preset constant with known values determined through experience. Alternatively, in the initial state, reference X-ray images of the same type of implant under a standard bone mineral density phantom are acquired, and multiple slice regions are obtained using the same slice region division method. The total gray level of all slice regions is then calculated as the standard total gray level. It should be understood that the standard total gray level is a positive number with a certain value and cannot be 0.
[0036] The ratio of the total gray level of all slice regions to the standard total gray level is calculated as a feature of absolute bone mineral density. ; in, Indicates absolute bone mineral density characteristics, This indicates the standard total gray level.
[0037] Absolute bone mineral density characteristics The physical meaning lies in: representing the ratio of the current actual overall density level of the peri-implant bone tissue to the standard state, when When the actual bone density is low, it indicates that the overall bone density is actually low. The smaller the value, the more severe the overall osteoporosis; when When, it indicates that the actual bone density is consistent with the standard condition; when When the value is high, it indicates that the actual bone mineral density is at a relatively high level overall. Absolute bone mineral density characteristics The smaller the implant size, the higher the risk; that is, the higher the composite risk index of the implant. The composite risk index is related to absolute bone mineral density characteristics. Inverse correlation.
[0038] The bone mineral density intensity characteristics of each slice region represent the relative distribution pattern of bone mineral density along the axial direction (e.g., whether it is sparse at the top and dense at the bottom, or uniformly distributed), but do not contain absolute intensity information; absolute intensity information is represented by absolute bone mineral density characteristics.
[0039] Step S2: Based on the growth of bone density intensity characteristics in adjacent slice regions, the location of the bone interface of the implant is obtained.
[0040] Based on the bone mineral density (BMD) characteristics of each slice region, the growth of BMD characteristics in adjacent slice regions is determined. This growth in BMD characteristics yields the geometric boundary between bone and soft tissue, thus identifying the bone interface location of the implant. For ease of explanation, the BMD characteristics of each slice region are ordered from the implant neck to the root, forming a BMD characteristic sequence.
[0041] To assess the degree of bone resorption at the implant neck, it is necessary to accurately locate the point where bone tissue begins to appear axially (i.e., the bone interface of the implant). Since there is a significant difference in grayscale values between bone tissue (high bone density) and soft tissue (low bone density), this difference manifests as abrupt changes in the bone mineral density (BMD) characteristic value sequence. Therefore, this embodiment identifies abrupt change points by calculating the first-order difference value of the BMD characteristic. The first-order difference value represents the degree of change in the BMD characteristic value between two adjacent slice regions. Accordingly, the first-order difference sequence of the BMD characteristic sequence is obtained. For any first-order difference value in the first-order difference sequence: ; in, Indicates the first First-order difference values of bone mineral density intensity characteristics of each slice region Indicates the first The bone mineral density (BMD) characteristics of each slice region. It should be understood that this embodiment does not calculate the first-order difference value of the last BMD characteristic in the BMD characteristic sequence, and it is not included in subsequent processing.
[0042] For the The first-order difference value of the bone mineral density intensity characteristic of each slice region, if positive, indicates that the first-order difference value is positive. The bone mineral density intensity characteristic of the first slice region is greater than that of the second slice region. The bone mineral density (BMD) value of the slice region indicates an increasing trend (i.e., bone density is increasing, suggesting a possible transition from soft tissue to bone tissue). A higher value indicates a greater increase, or a faster rate of increase in BMD. A value of 0 indicates the [missing value]. The bone mineral density characteristic of the slice region is equal to that of the slice region. The bone mineral density (BMD) characteristics of the slice region are stable and unchanged; a negative value indicates that the first slice region is stable and unchanged. The bone mineral density characteristics of the first slice region are less than those of the second slice region. The bone mineral density (BMD) characteristics of the slice region are in a decreased state.
[0043] To prevent misidentification caused by image noise (such as metal artifacts or isolated bone spurs) and to accurately identify the true bone interface location rather than other local interference points, this embodiment introduces a search range constraint mechanism. Specifically, this embodiment presets a target slice area range, which is the first portion of the slice area from the implant neck to the root. In an exemplary embodiment, a preset percentage of the slice area is selected, forming the target slice area range. Considering that peri-implantitis mainly leads to neck bone resorption, the true bone interface is usually located in the upper middle part of the implant. Therefore, the preset percentage range can be [50%, 70%]. This embodiment uses 60% as an example. Therefore, the index range of the slice area corresponding to the target slice area range is... , This represents a preset percentage, for example, 60%. If the number of slice regions is 100, then the index range of the slice regions corresponding to the target slice region range is... .
[0044] This embodiment searches for the bone interface location of the implant within the target slice area. Specifically: First-order difference values of bone density intensity characteristics of each slice region within the target slice area are obtained and sorted in ascending order of index to form a first-order difference target sequence. The largest first-order difference value in the first-order difference target sequence is obtained, and the slice region corresponding to this largest first-order difference value is determined, thereby obtaining the location of this slice region. This location of the slice region is taken as the bone interface location, that is, the index corresponding to the slice region is taken as the bone interface location, represented as... ,So, For interval An integer within.
[0045] It should be understood that the maximum first-order difference value is a positive number. To ensure the accuracy and reliability of the bone interface location, this embodiment can set a first-order difference effective threshold (also a positive number). This first-order difference effective threshold is used as a judgment benchmark to compare with the maximum first-order difference value to determine whether the maximum first-order difference value is effective. The first-order difference effective threshold can be directly set by experience, or it can be obtained in the following way: Select several standard samples with good bone integration from historical clinical data (the standard samples with good bone integration can be judged by doctors from a large number of samples in historical clinical data and then manually calibrated), calculate the average value of the bone density intensity characteristics at the bone interface location of each standard sample, multiply the average value by a preset safety reduction factor (the value ranges from 0.3 to 0.5, for example, 0.3), and the product is used as the first-order difference effective threshold. If the maximum first-order difference value is less than the effective threshold of the first-order difference, the maximum first-order difference value is determined to be invalid, and the currently acquired implant image is determined to be invalid. The subsequent process is terminated and a prompt indicating that the image quality is unqualified is output; otherwise, the subsequent steps are continued.
[0046] In addition, to further improve the robustness of recognition, this embodiment can also obtain the location of the bone interface after obtaining the location of the bone interface. Afterwards and with The process involves taking several adjacent, consecutive slice regions (e.g., 5), calculating the average bone mineral density (BMD) of these regions, and determining whether this average value remains at a relatively high level. Specifically, it checks whether the average value exceeds a preset BMD threshold. This preset BMD threshold can be obtained empirically or set as the average BMD of the bone interface locations of the standard samples mentioned above. If it exceeds the threshold, the bone interface location selection is deemed valid; otherwise, the bone interface location selection is deemed invalid, the currently acquired implant image is deemed invalid, the subsequent process is terminated, and an image quality failure message is output.
[0047] Step S3: Combine the cumulative value of bone mineral density characteristics up to each slice region with the corresponding ideal cumulative value of bone mineral density characteristics to obtain the amount of implant center of gravity shift.
[0048] To quantify the positional deviation of the actual bone mineral density distribution relative to the ideal state, this step obtains the deviation between the cumulative value of the bone mineral density intensity characteristic and the ideal value.
[0049] Because minute differences in the initial stage (i.e., the neck) of the bone mineral density (BMD) characteristic sequence are propagated and superimposed on all subsequent locations during the accumulation process, osteoporosis occurring in the neck will contribute a larger numerical value to the calculation results than osteoporosis at the root. This perfectly aligns with the biomechanical characteristics of the implant neck, where the stress is greatest. Therefore, this embodiment obtains the cumulative BMD characteristic values up to each slice region, and by quantifying the deviation from the ideal value, obtains the amount of implant center of gravity displacement.
[0050] First, obtain the cumulative value of bone mineral density characteristics up to each slice region: ; in, This represents the cumulative value of bone mineral density (BMD) up to the k-th slice region, i.e., the cumulative value of BMD in the k-th slice region. The cumulative bone mineral density (BMD) value of the k-th slice region represents the cumulative proportion of actual bone mass from the tip of the implant neck to the k-th slice region. The cumulative BMD values of each slice region are obtained from this, and then sorted in ascending order of index to obtain a sequence of cumulative BMD values.
[0051] This embodiment establishes a reference baseline, specifically the ideal cumulative value of bone mineral density (BMD) characteristics up to each slice region, to characterize the distribution of BMD characteristics in each slice region assuming completely healthy and dense bone. In an exemplary embodiment, the location of the bone interface is used as a reference. As a dividing point, all slice regions are divided into two parts: the slice region before the bone interface (i.e., the index is less than 100%). The slice region), and the slice region located at or after the bone interface (i.e., index greater than or equal to). (slice area).
[0052] index less than The initial bone mineral density intensity feature reference value for each slice region is set to the first value, and the index is greater than or equal to The initial bone mineral density intensity characteristic reference value for each slice region is set to the second value. (Index less than...) Each slice area is the region above the implant neck. In this embodiment, the index is set to be less than... Each slice region represents a soft tissue or bone deficiency area, therefore the ideal bone volume is relatively small, meaning the first value is smaller; the index is greater than or equal to Each slice region is a bone integration zone. In this embodiment, the index is set to be greater than or equal to... The trabecular bone structure in each slice region is completely intact, dense, and uniformly distributed, resulting in a larger ideal bone mass, i.e., a larger second value. Therefore, the first value is smaller than the second value. In an exemplary embodiment, for ease of data processing, this embodiment sets the first value to 0 and the second value to 1. Then: ; in, Indicates the first Initial bone mineral density intensity characteristic reference values for each slice region.
[0053] To ensure that the initial bone mineral density (BMD) intensity reference value can be compared with the aforementioned BMD intensity feature on the same mathematical dimension, this embodiment performs the same normalization process on the initial BMD intensity reference value as on the BMD intensity feature: the sum of the initial BMD intensity reference values for all slice regions is calculated, and then the ratio of the initial BMD intensity reference value for each slice region to this sum is calculated to obtain the normalized initial BMD intensity reference value, which is defined as the ideal value of the BMD intensity feature. ; in, Indicates the first Ideal values of bone mineral density intensity characteristics for each slice region Indicates the first Initial bone mineral density intensity characteristic reference values for each slice region. After normalization. Represents the first The theoretically optimal bone density distribution pattern for each slice region.
[0054] Then, the ideal cumulative value of bone mineral density characteristics up to each slice region is obtained: ; in, This represents the ideal cumulative value of bone mineral density (BMD) up to the k-th slice region, i.e., the ideal cumulative value of BMD in the k-th slice region. From this, the ideal cumulative values of BMD for each slice region are obtained. These ideal cumulative values are then sorted in ascending order of their indices to obtain a sequence of ideal cumulative BMD values.
[0055] The downward shift of the center of gravity represents the distance by which the actual bone distribution center of gravity moves downward (i.e., towards the root) relative to the ideal state. In an exemplary embodiment, the downward shift of the center of gravity is obtained by determining the Manhattan distance (i.e., L1 distance) between the cumulative bone mineral density feature sequence and the ideal cumulative bone mineral density feature sequence. ; in, This indicates the amount of downward shift in the center of gravity. (Amount of downward shift in the center of gravity) Essentially, it represents the area enclosed by the two cumulative curves corresponding to the cumulative value sequence of bone mineral density (BMD) characteristics and the ideal cumulative value sequence of BMD characteristics, and the downward shift of the center of gravity. It is a dimensionless scalar, with a range of values within 1000. Between. The amount of downward shift in the center of gravity. The larger the value, the farther the actual bone distribution center of gravity has shifted downwards relative to the ideal state. This means that the bone in the neck area is relatively hollow, the mechanical support is poor, and the composite risk index of the implant is higher. The composite risk index is positively correlated with the amount of downward shift of the center of gravity.
[0056] Step S4: Integrate the center of gravity shift, geometric cantilever characteristics, and absolute bone mineral density characteristics to obtain the composite risk index of the implant.
[0057] The center of gravity shift represents the change in the center of gravity of microscopic bone density. Besides the change in the center of gravity of microscopic density, the macroscopic bone integration height (i.e., cantilever length) directly determines the lever arm. Therefore, the geometric risk component, defined as the geometric cantilever characteristic, is obtained from the bone interface location. The larger the value of the bone interface location (i.e., the larger the index of the bone interface location), the closer the bone interface location is to the root of the implant, the larger the cantilever length, the higher the geometric cantilever characteristic, and the higher the composite risk index of the implant. The composite risk index is positively correlated with the geometric cantilever characteristic. The specific process of obtaining the geometric cantilever characteristic is as follows: First, based on the index of the bone interface location... and the number of slice regions Calculate the proportion of cantilever sections : ; This embodiment can directly calculate the proportion of the cantilever section. As a geometric cantilever feature Due to the index of the bone interface location Less than the number of slice regions For example, the index of the bone interface location mentioned above. Maximum ,for example Then the geometric cantilever features Less than 1 (maximum value is) ), and geometric cantilever features There is a certain gap between it and 1.
[0058] To characterize the proportion of cantilever segments When the cantilever increases in size (i.e., the bone interface moves towards the root, and the cantilever becomes longer), the geometric characteristics of the cantilever change. The range of variation is greater, achieving geometric cantilever features. The nonlinear change reflects the amplification effect of the mechanical load when the bone interface moves towards the root. Therefore, as an optimized implementation method, the geometric cantilever characteristics can also be given below. Another way to obtain it: .
[0059] Therefore, absolute bone mineral density characteristics Reflects overall bone mass, the amount of downward shift of the center of gravity Reflects microscopic bone density distribution, geometric cantilever characteristics Reflecting the macroscopic lever arm, and to comprehensively assess the mechanical risk of the implant, this embodiment integrates the above three characteristics into a composite index, defined as the composite risk index. Based on the logical relationship between the above characteristics and the composite risk index, a specific calculation method for the composite risk index is given below: ; in, This represents the composite risk index. In this embodiment, we take... The reciprocal of the value is used to characterize the inverse correlation, when overall osteoporosis (i.e., When the value decreases (e.g., from 1.0 to 0.5), An increase (e.g., from 1.0 to 2.0) directly leads to a higher composite risk index. Doubling this figure ensures that the overall risk of osteoporosis is not missed due to normalization. This represents the sensitivity adjustment coefficient for the amount of center of gravity shift. This coefficient is used to adjust the degree of response to microscopic changes in the center of gravity. In practical applications, this sensitivity adjustment coefficient can be calibrated through retrospective analysis of clinical failure case data, or a recommended empirical value can be directly used (as set in this embodiment). ). It represents an exponential function with the natural constant as the base. The amount of the center of gravity shift is amplified by an exponential function, which means that even a small linear increase in the amount of the center of gravity shift will lead to a non-linear and sharp increase in the composite risk index value. It simulates the mutation process of biological structures from compensation to instability and can effectively amplify the weak early warning signals.
[0060] Composite Risk Index It comprehensively reflects multiple risk factors, both macroscopic and microscopic, absolute and relative, providing an accurate data foundation for subsequent dynamic judgment.
[0061] Step S5: Early warning of peri-implantitis risk using the composite risk index.
[0062] According to the composite risk index This embodiment allows for early warning of peri-implantitis risk by directly setting a warning threshold (which serves as a warning benchmark to determine the composite risk index). Whether the warning requirements are met depends on the specific value of the warning threshold, which is set according to the warning needs. For example, if a more secure warning logic is required, the warning threshold can be set to a smaller value (in this embodiment, it is set to 2.0). When the composite risk index... If the value is greater than or equal to the threshold, an early warning signal for peri-implantitis risk will be output; otherwise, no early warning signal for peri-implantitis risk will be output.
[0063] In other embodiments, to address the issue of false alarms or missed alarms caused by differences in congenital bone quality among patients, this embodiment may not employ a single fixed threshold. Instead, it establishes a dynamic judgment mechanism based on individual historical data and generates a targeted early warning strategy for peri-implantitis risk. Different patients have inherent differences in bone quality (such as cortical bone thickness and trabecular bone density) at the initial stage of implantation, leading to potentially different initial composite risk indices. Using a fixed warning threshold might result in false alarms for patients with poor congenital bone quality immediately after implantation, or failure to trigger alarms even after significant degeneration in patients with excellent congenital bone quality. Therefore, this embodiment introduces the relative severity of the composite risk index as a core judgment indicator.
[0064] This embodiment requires a preset stable composite risk index value, which characterizes the physiological risk level of the monitored subject. In an exemplary embodiment, the historical stable period of the monitored subject is obtained. To prevent baseline deviation caused by the initial implantation, this embodiment uses the third month after implantation as the historical stable period. Multiple implant images (e.g., 5 times) are acquired within this historical stable period, and the composite risk index of each implant image is obtained according to the above process. The average value of the multiple composite risk indices acquired within the historical stable period is calculated as the preset stable composite risk index value.
[0065] The relative severity of the composite risk index is calculated as follows: ; in, This indicates the relative severity of the composite risk index. This indicates the stable value of the composite risk index. Relative severity. This intuitively reflects the current compound risk index as a multiple of deterioration relative to an individual's normal physiological state. When the relative severity... A value greater than 1 indicates that the composite risk index is higher than the stable value of the composite risk index, and the relative severity is higher. The larger the implant, the higher the risk of peri-implantitis; when the relative severity When the value equals 1, it indicates that the composite risk index equals the stable value of the composite risk index; when the relative severity... A value less than 1 indicates that the composite risk index is below the stable value, and the current risk of peri-implantitis is low. As an example: when the relative severity... A value less than or equal to 1 indicates that the patient's current bone condition is stable compared to the initial period after implantation, and the risk is low.
[0066] This embodiment presupposes several risk ranges, based on their relative severity. The risk zone it is in provides an early warning signal of the risk of peri-implantitis.
[0067] In one exemplary embodiment, two preset thresholds are defined as a first preset risk threshold and a second preset risk threshold. The first preset risk threshold is lower than the second preset risk threshold. The specific values of these two thresholds are set according to the judgment requirements. For example, the first preset risk threshold is set to 1.5, and the second preset risk threshold is set to 3.0.
[0068] The intervals below the first preset risk threshold are defined as low-risk intervals; the intervals above or equal to the first preset risk threshold and below the second preset risk threshold are defined as medium-risk intervals; and the intervals above or equal to the second preset risk threshold are defined as high-risk intervals.
[0069] If relative severity If the risk level is low, it is determined that there is no risk of peri-implantitis, the implant is in a mechanically stable state, and a risk-free signal can be output.
[0070] If relative severity If the risk level is in the medium-risk range, it indicates that there is a moderate risk of peri-implantitis, meaning that the implant is in a compensatory period of microstructural changes. At this time, although there may be no obvious abnormalities in the macroscopic imaging, the microscopic support capacity has been significantly reduced (i.e., the risk value has increased to more than 1.5 times the baseline and less than 3.0 times). The output early warning signal of peri-implantitis risk is a moderate risk signal.
[0071] If relative severity If the implant is in the high-risk zone, it is determined that there is a high degree of risk of peri-implantitis, indicating that the implant is in a high-risk period of mechanical instability, and the output early warning signal for peri-implantitis risk is a high-risk signal.
[0072] The invention ultimately outputs warning signals of different risk levels, which can be visually displayed through different display content or different alarm light colors.
[0073] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0074] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.
Claims
1. An early warning system for the risk of peri-implantitis, characterized in that, include: The data acquisition module is used to acquire slices of bone tissue around the implant along the implant axis. The early warning module, which is electrically connected to the data acquisition module, is configured as follows: Based on the grayscale values of each slice region, the bone mineral density intensity characteristics of each slice region are obtained; The location of the bone interface of the implant is obtained based on the growth of bone density intensity characteristics in adjacent slice regions. The deviation between the cumulative value of bone mineral density characteristics up to each slice region and the corresponding ideal cumulative value of bone mineral density characteristics is used to obtain the amount of implant center of gravity shift. By integrating the center of gravity shift, geometric cantilever features, and absolute bone mineral density features, a composite risk index for the implant is obtained; the geometric cantilever features are obtained from the bone interface position, and the absolute bone mineral density features are obtained from the total gray level of all slice regions; Early warning of peri-implantitis risk using a composite risk index.
2. The early warning system for the risk of peri-implantitis as described in claim 1, characterized in that, The process of obtaining each slice region includes: The length of each slice region is obtained based on the axial length of the implant and the preset number of slice regions. Based on the preset width and length of each slice region, the bone tissue around the implant is sliced to obtain each slice region.
3. The early warning system for the risk of peri-implantitis as described in claim 1, characterized in that, The process of obtaining the bone mineral density strength characteristics includes: Obtain the average grayscale value of each slice region; The ratio of the mean gray level of each slice region to the total gray level of all slice regions is calculated to obtain the bone mineral density characteristics of each slice region; the total gray level of all slice regions is the sum of the mean gray levels of all slice regions.
4. The early warning system for the risk of peri-implantitis as described in claim 1, characterized in that, The process of obtaining the bone interface location includes: A first-order difference sequence of bone mineral density intensity features is obtained, wherein the bone mineral density intensity features of each slice region are sorted in the direction from the neck of the implant to the root. The maximum first-order difference value is determined from the target slice region range in the first-order difference sequence; the target slice region range includes the first preset percentage of slice regions selected in the direction from the neck to the root of the implant; The location of the slice region corresponding to the maximum first-order difference value is determined as the location of the bone interface.
5. The early warning system for the risk of peri-implantitis as described in claim 1, characterized in that, The process of obtaining the ideal cumulative value of the bone mineral density strength characteristic includes: An initial bone mineral density intensity characteristic reference value is set for each slice region; wherein, the initial bone mineral density intensity characteristic reference value of each slice region before the bone interface position is set as a first value, and the initial bone mineral density intensity characteristic reference value of each slice region at and after the bone interface position is set as a second value; the first value is less than the second value. The ideal value of bone mineral density (BMD) for each slice region is obtained by calculating the ratio of the initial reference value of BMD for each slice region to the sum of the initial reference values of BMD for all slice regions. Based on the ideal values of bone mineral density intensity characteristics in each slice region, the cumulative ideal values of bone mineral density intensity characteristics up to each slice region are obtained.
6. The early warning system for the risk of peri-implantitis as described in claim 1, characterized in that, The process of obtaining the downward shift of the center of gravity includes: The Manhattan distance between the cumulative bone mineral density feature sequence and the ideal cumulative bone mineral density feature sequence is determined to obtain the centroid shift amount; the cumulative bone mineral density feature sequence is obtained by sorting the cumulative bone mineral density feature values up to each slice region; the ideal cumulative bone mineral density feature sequence is obtained by sorting the ideal cumulative bone mineral density feature values up to each slice region.
7. The early warning system for peri-implantitis risk as described in claim 1, characterized in that, The process of obtaining the absolute bone density feature includes: calculating the ratio of the total gray level of all slice regions to the preset standard total gray level, which is used as the absolute bone density feature; The composite risk index is positively correlated with the center of gravity shift and geometric cantilever characteristics, and negatively correlated with the absolute bone mineral density characteristics.
8. The early warning system for the risk of peri-implantitis as described in claim 1, characterized in that, The early warning of peri-implantitis risk using a composite risk index includes: The relative severity of the composite risk index is obtained by calculating the ratio of the composite risk index to the preset stable value of the composite risk index. Determine the preset risk range in which the relative severity falls, and output an early warning signal for the risk of peri-implantitis.
9. The early warning system for peri-implantitis risk as described in claim 8, characterized in that, The process of obtaining the preset risk range includes: The range below the first preset risk threshold is defined as the low-risk range; The range that is greater than or equal to the first preset risk threshold and less than the second preset risk threshold is defined as the medium-risk range. The intervals that are greater than or equal to the second preset risk threshold are defined as high-risk intervals; the first preset risk threshold is less than the second preset risk threshold.
10. The early warning system for peri-implantitis risk as described in claim 8, characterized in that, The process of obtaining the stable value of the preset composite risk index includes: calculating the average value of multiple composite risk indices obtained within a preset historical stable period, and using it as the stable value of the preset composite risk index.