Dental data analysis system based on medical big data

Abstract

The present invention discloses a dental data analysis system based on medical big data. The present invention relates to the field of dental analysis technology and solves the problem of determining whether the dental arrangement is normal based on the arrangement status of all teeth. The present invention focuses on the tooth arrangement in the dental bed through a tooth feature analysis processing end and monitors the midpoint distance between adjacent teeth in the upper and lower dental beds. It can detect abnormal tooth arrangement in time, which is of great significance to orthodontic patients. Doctors can quickly grasp the tooth movement deviation in the orthodontic process according to the arrangement signal generated by the system, fine-tune the appliance parameters, and ensure that the orthodontic effect is achieved on time. For patients with oral rehabilitation, it can avoid the new restoration affecting the surrounding tooth arrangement and maintain the mechanical balance of the oral chewing system.

Description

Technical Field

[0001] The present invention relates to the technical field of dental analysis, and in particular to a dental data analysis system based on medical big data. Background Art

[0002] Traditional periapical radiographs clearly show the condition of a single tooth and its root within the gums, revealing root length, shape, and the presence of periapical lesions. Panoramic radiographs, on the other hand, offer a panoramic view of the entire dentition and bilateral gums, useful for examining the eruption of wisdom teeth, jaw fractures, and the extent of alveolar bone resorption across multiple teeth. For example, in patients with periodontitis, panoramic radiographs can reveal the extent of alveolar bone resorption horizontally and vertically along the dentition, assisting doctors in determining the stage of the disease.

[0003] Application Publication No. CN111083922A discloses a dental image analysis method for orthodontic diagnosis and a device utilizing the same. The method includes the following steps: acquiring a dental image of a patient; and detecting at least a portion of a plurality of measurement points (landmarks) used for orthodontic diagnosis from the dental image using a measurement point detection module. The measurement points are anatomical reference points that indicate the relative positions of at least one of the facial skeleton, teeth, and facial contours required for orthodontic diagnosis. The measurement point detection module includes a mechanical training module based on an artificial neural network.

[0004] For the confirmed dental images, the relevant features of the teeth in the dental images are analyzed, and the corresponding teeth are verified and checked with the corresponding tooth templates. Based on the verification results, it is assessed whether the teeth are arranged normally. However, in the actual assessment process, whether the arrangement of the gums is normal should be determined based on the arrangement status of all teeth, and the parameter representation between the center points of the teeth should be determined based on the center point distance between the teeth, so as to determine the specific arrangement result. Summary of the Invention

[0005] In response to the deficiencies of the prior art, the present invention provides a gum data analysis system based on medical big data, which solves the problem of determining whether the gum arrangement is normal based on the arrangement status of all teeth.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a dental data analysis system based on medical big data, comprising:

[0007] The CT image acquisition terminal acquires and confirms the dental CT images involved in the dental detection, and transmits the confirmed dental CT images to the tooth area determination terminal;

[0008] The tooth region determination end, based on the CT values associated with different points in the dental CT image, screens the tooth region from the dental CT image, then identifies the individual tooth from the features of the screened tooth region, and then identifies the dental pulp region from the edge contour of the individual tooth. The specific method is as follows:

[0009] Based on the confirmed dental CT image, the CT values associated with different points in the dental CT image are determined and calibrated as H i , where i represents different points in the dental CT image;

[0010] Will CT meet: H i Points ∈[1500Hu, 3000Hu] are calibrated as tooth points, where Hu is a numerical unit. Based on the calibrated groups of tooth points, the image areas covered by the groups of tooth points are confirmed. Based on the overall edge contours of such image areas, the areas within the overall edge contours in the dental CT images are calibrated as tooth areas.

[0011] Locking H from the tooth area i ∈[30Hu, 100Hu], the locked points are marked as pulp points, and the area covered by several groups of pulp points is marked as the pulp area;

[0012] The pulp areas associated with different tooth areas were calibrated in sequence using the same processing method;

[0013] The tooth calibration processing end, based on the pulp area confirmed in the corresponding tooth area, confirms the center points of the tooth area and the internal pulp area, confirms the distance characteristics between the two sets of center points, and based on the numerical expression of the distance characteristics, calibrates the single tooth associated with such tooth area as a normal tooth or a missing tooth. The specific method is as follows:

[0014] Based on the determined tooth region, the edge contour of the region is locked, and an edge contour line belonging to the tooth region is generated. The edge contour line is placed in a set of two-dimensional coordinate systems. The two-dimensional coordinates of different contour points within the edge contour line are determined from the two-dimensional coordinate systems. Then, several sets of two-dimensional coordinates associated with the different contour points are averaged to determine the average coordinates. Then, based on the location of the average coordinates, point calibration is performed within the tooth region to determine the center point A belonging to the tooth region.

[0015] Then, from the pulp area confirmed within the tooth area, the edge contour of the pulp area is preliminarily confirmed, and the edge contour line of the pulp area is generated. Then, the edge contour line is placed in a two-dimensional coordinate system in the same manner, and the center point B of the pulp area is simultaneously confirmed in the two-dimensional coordinate system.

[0016] Based on the specific positions of center point A and center point B, determine the straight-line distance L between the two sets of center points A and B, and identify whether this straight-line distance L satisfies: L≤Y1, where Y1 is a preset value. If so, the teeth associated with this tooth area are marked as normal teeth. If not, the teeth associated with this tooth area are marked as missing teeth.

[0017] The feature midpoint confirmation end reanalyzes the single tooth calibrated as missing tooth, and locks the symmetry midline based on the overall edge contour of the single tooth and the center point of the internal pulp area. Then, based on the determined symmetry midline, the feature center point of the missing tooth is locked based on this symmetry midline. The specific method is as follows:

[0018] The center point of the pulp area inside the missing tooth is marked as the main point, and the feature point closest to the main point is located in the outer area of the missing tooth, and this feature point is used as the secondary point of the main point;

[0019] Connect the primary point and the secondary point, and extend the line. The extended line is used as the symmetrical midline of the missing tooth.

[0020] Based on the determined symmetry midline, different tooth regions located on both sides of the symmetry midline are identified from the tooth image belonging to the missing tooth; based on the external contours of the tooth regions on both sides, a tooth region on the side with a larger area expression value is determined, and this tooth region is used as the area to be mirrored; based on the determined symmetry midline, the area to be mirrored is mirrored to confirm the mirrored area; the area formed by merging the area to be mirrored and the mirrored area is used as the main feature area; based on the overall edge contour of the main feature area, a feature center point of the main feature area is locked, and the locked feature center point is used as the feature center point of the missing tooth;

[0021] The missing teeth after the feature center point calibration is completed are transmitted to the tooth feature analysis and processing end;

[0022] The tooth feature analysis and processing end performs arrangement analysis on missing teeth after feature center point calibration and normal teeth after center point calibration. It prioritizes identifying the upper or lower gums from dental CT images, and numerically analyzes the center point spacing of corresponding teeth in the upper or lower gums. The analysis results are determined and displayed in the following ways:

[0023] The upper or lower gum is locked from the dental CT image. The upper or lower gum has been calibrated in advance. The upper gum is the area where the teeth are arranged above, and the lower gum is the area where the teeth are arranged below.

[0024] Confirm the characteristics of the internal teeth located in the upper gum. The internal teeth include normal teeth or missing teeth, and mark the characteristic center points of the missing teeth and the center points of the normal teeth as the midpoints of the internal teeth: identify the straight-line distances between the inner midpoints of the adjacent internal teeth in the corresponding gums, and then mark the confirmed straight-line distances as L q , where q represents different straight-line distances, and the confirmed L q Perform mean processing to confirm the characteristic parameter L1 of the upper gum;

[0025] Treat the inner teeth of the lower gum in the same way as the upper gum, and confirm the characteristic parameter L2 of the lower gum;

[0026] Identify whether L1 and L2 satisfy: |L1-L2|>Y2, where Y2 is a preset value. If so, generate and display an abnormal arrangement signal; if not, generate a normal arrangement signal.

[0027] The present invention provides a dental data analysis system based on medical big data. Compared with the existing technology, it has the following advantages:

[0028] The tooth calibration processing terminal uses a unique algorithm to calibrate the tooth condition based on the distance characteristics between the center point of the tooth area and the pulp area. Compared with traditional manual judgment based on experience, this system can quickly process large amounts of dental CT image data and quickly identify tooth loss. If tooth loss causes a deviation in the center point, the system will immediately and accurately identify it and provide key guidance for the subsequent oral restoration plan, accelerating the diagnosis and treatment process and allowing patients to receive appropriate restorative treatment more quickly.

[0029] The tooth feature analysis and processing end focuses on the tooth arrangement in the gums and monitors the midpoint distance between adjacent teeth in the upper and lower gums. It can detect abnormal tooth arrangement in a timely manner, which is of great significance for patients returning for orthodontic treatment. Doctors can use the arrangement signals generated by the system to quickly grasp the deviation of tooth movement during the orthodontic process, fine-tune the parameters of the orthodontic appliance, and ensure that the orthodontic effect is achieved on time. For patients with oral rehabilitation, it can prevent the new restoration from affecting the arrangement of surrounding teeth and maintain the mechanical balance of the oral chewing system. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a schematic diagram of the principle framework of the present invention. DETAILED DESCRIPTION

[0031] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0032] First embodiment

[0033] See also Figure 1 The present application provides a dental data analysis system based on medical big data, including a CT image acquisition end, a tooth region determination end, a tooth calibration processing end, a feature midpoint confirmation end, and a tooth feature analysis processing end, wherein the CT image acquisition end is electrically connected to an input node of the tooth region determination end, the tooth region determination end is electrically connected to an input node of the tooth calibration processing end, and the tooth calibration processing end is electrically connected to an input node of the feature midpoint confirmation end or a tooth feature analysis processing end, respectively, and the feature midpoint confirmation end is electrically connected to an input node of the tooth feature analysis processing end;

[0034] The CT image acquisition end acquires and confirms the dental CT images involved in the dental detection, and transmits the confirmed dental CT images to the tooth area determination end, wherein the dental CT images are detected by the corresponding dental CT detection equipment. The specific detection process and specific detection method are instructed by the relevant operator to perform image detection processing on the person to be detected, and the dental CT images are determined based on the corresponding detection and processing process.

[0035] The tooth region determination end selects the tooth region from the dental CT image based on the CT values associated with different points in the dental CT image, then identifies the individual tooth from the features of the selected tooth region, and then identifies the dental pulp region from the edge contour of the individual tooth. The specific method for identification is as follows:

[0036] Based on the confirmed dental CT image, the CT values associated with different points in the dental CT image are determined and calibrated as H i , where i represents different points in the dental CT image;

[0037] Will CT meet: H i∈[1500Hu, 3000Hu] is calibrated as a tooth point, where Hu is a numerical unit. Based on the calibrated groups of tooth points, the image areas covered by the groups of tooth points are confirmed. Based on the overall edge contour of such image areas, the area inside the overall edge contour in the dental CT image is calibrated as the tooth area. Specifically, since each tooth is separated, there are several determined tooth areas. Each different tooth area corresponds to a group of teeth, and its tooth area also has a corresponding edge contour. The specific position of the corresponding tooth can be locked based on the corresponding edge contour. The relevant points of the middle points inside such tooth areas do not belong to the CT value range of [1500Hu, 3000Hu], because the central area belongs to the pulp area, and the outer areas of the pulp area all belong to the tooth area. The corresponding points will generate an outer area surrounding the pulp area. Therefore, based on the overall edge contour of the corresponding outer area, the tooth area belonging to this image can be locked;

[0038] Locking H from the tooth area i ∈[30Hu, 100Hu], the locked points are marked as pulp points, and the area covered by several groups of pulp points is marked as the pulp area;

[0039] The dental pulp regions associated with different tooth regions are calibrated in sequence using the same processing method. Specifically, within the determined tooth region, the dental pulp is located at the center of the tooth region.

[0040] Among them, the tooth calibration processing end, based on the pulp area confirmed in the corresponding tooth area, confirms the center points of the tooth area and the internal pulp area, confirms the distance characteristics between the two groups of center points, and based on the numerical expression of the distance characteristics, calibrates the single tooth associated with such tooth area as a normal tooth or a missing tooth. Specifically, if the distance between the center points associated with the corresponding teeth is relatively close, then there is no problem with the characteristics between their center points. However, if the corresponding teeth are missing, the corresponding center point deviation will be large, so the corresponding tooth is a missing tooth;

[0041] The specific method for calibrating a single tooth is as follows:

[0042] Based on the determined tooth region, the edge contour of the region is locked, and an edge contour line belonging to the tooth region is generated. The edge contour line is placed in a set of two-dimensional coordinate systems. The two-dimensional coordinates of different contour points within the edge contour line are determined from the two-dimensional coordinate systems. Then, several sets of two-dimensional coordinates associated with the different contour points are averaged to determine the average coordinates. Then, based on the location of the average coordinates, point calibration is performed within the tooth region to determine the center point A belonging to the tooth region.

[0043] Then, from the pulp area confirmed within the tooth area, the edge contour of the pulp area is preliminarily confirmed, and the edge contour line of the pulp area is generated. Then, the edge contour line is placed in a two-dimensional coordinate system in the same manner, and the center point B of the pulp area is simultaneously confirmed in the two-dimensional coordinate system.

[0044] Based on the specific positions of center point A and center point B, determine the straight-line distance L between the two sets of center points A and B, and identify whether this straight-line distance L satisfies: L≤Y1, where Y1 is a preset value, and its specific value is determined by the operator based on experience. If it satisfies, the teeth associated with this tooth area are marked as normal teeth, otherwise, the teeth associated with this tooth area are marked as missing teeth;

[0045] Specifically, in the actual confirmation processing process, when the center points of the tooth area and the pulp area are greatly different, it means that the corresponding tooth is missing, which is a missing tooth. Its pulp area is located at the center of the tooth area and generally will not be reduced due to damage or loss of the tooth. When the corresponding tooth is missing, it will cause the center point of the tooth to shift, resulting in a large difference between the center point of the tooth area and the center point of the pulp area. Under normal circumstances, the pulp area is located at the center of the tooth area, and the center point of the tooth area and the center point of the pulp area are basically close, so the straight-line distance between the two is also close.

[0046] Among them, the feature midpoint confirmation end re-analyzes the single tooth calibrated as missing tooth, locks the symmetry midline based on the overall edge contour of the single tooth and the center point of the internal pulp area, and then locks the feature center point of the missing tooth based on the determined symmetry midline. The specific locking method is:

[0047] The center point of the pulp area of the missing tooth is calibrated as the main point. The feature point closest to the main point in the outer area of the missing tooth (that is, the enamel area, the CT value of the enamel area is between 2500-3000 Hu) is located and used as the secondary point of the main point.

[0048] Connect the primary point and the secondary point, and extend the line. The extended line is used as the symmetrical midline of the missing tooth.

[0049] Based on the determined symmetry midline, different tooth areas on both sides of the symmetry midline are identified from the tooth image belonging to the missing tooth, and based on the external contours of the tooth areas on both sides, the tooth area on the side with the larger area expression value is determined, and this tooth area is used as the area to be mirrored, and based on the confirmed symmetry midline, the area to be mirrored is mirrored to confirm the mirrored area, and the area after the area to be mirrored and the mirrored area are merged is used as the main feature area, and based on the overall edge contour of the main feature area, the feature center point of the main feature area is locked, and the locked feature center point is used as the feature center point of the missing tooth. The feature center point of the main feature area is confirmed in the same way as the center point of the dental pulp area, and based on the corresponding overall edge contour and the associated two-dimensional coordinate system, the two-dimensional coordinates of the contour points on the contour are confirmed, and after locking the average coordinates of several two-dimensional coordinates, the center point is specifically confirmed;

[0050] The missing teeth after the feature center point calibration is completed are transmitted to the tooth feature analysis and processing end.

[0051] Second embodiment

[0052] In the specific implementation process of this embodiment, the arrangement of internal teeth in the dental CT image is calibrated and analyzed to assess whether the corresponding upper and lower teeth are arranged as per the standard;

[0053] The tooth feature analysis and processing end performs arrangement analysis on the missing teeth after feature center point calibration and the normal teeth after center point calibration (i.e., center point A). It prioritizes identifying the upper or lower gums from the dental CT images, and performs numerical analysis on the center point spacing of the corresponding teeth in the upper or lower gums. The analysis results are determined and displayed. The specific method for performing numerical analysis is as follows:

[0054] The upper or lower gums are locked in dental CT images. When the images were taken, they were calibrated in advance. The upper gums are the upper area where the teeth are arranged, and the lower gums are the lower area where the teeth are arranged.

[0055] Confirm the characteristics of the internal teeth located in the upper gum. The internal teeth include normal teeth or missing teeth, and mark the characteristic center points of the missing teeth and the center points of the normal teeth as the midpoints of the internal teeth: identify the straight-line distances between the inner midpoints of the adjacent internal teeth in the corresponding gums, and then mark the confirmed straight-line distances as L q , where q represents different straight-line distances, and the confirmed L q Perform mean processing to confirm the characteristic parameter L1 of the upper gum;

[0056] Treat the inner teeth of the lower gum in the same way as the upper gum, and confirm the characteristic parameter L2 of the lower gum;

[0057] Identify whether L1 and L2 satisfy: |L1-L2|>Y2, where Y2 is a preset value, and its specific value is determined by the operator based on experience. If it satisfies, an abnormal arrangement signal is generated and displayed; if not, a normal arrangement signal is generated;

[0058] Specifically, in the actual treatment process, the relative arrangement between adjacent teeth in the upper and lower gums should be in a relatively standard state. If it is not in a standard state, it means that the corresponding upper and lower arrangements of the teeth are abnormal. For such abnormal situations, timely confirmation and signal display are required to facilitate medical staff to take corresponding targeted measures.

[0059] Third embodiment

[0060] The specific implementation process of this embodiment includes the entire implementation process of the above two groups of embodiments.

[0061] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0062] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. The dental data analysis system based on medical big data is characterized by: include: The CT image acquisition terminal acquires and confirms the dental CT images involved in the dental detection, and transmits the confirmed dental CT images to the tooth area determination terminal; The tooth region determination end selects the tooth region from the dental CT image based on the CT values associated with different points in the dental CT image, then identifies the individual tooth from the features of the selected tooth region, and then identifies the dental pulp region from the edge contour of the individual tooth; The tooth calibration processing end, based on the pulp area confirmed in the corresponding tooth area, determines the center points of the tooth area and the internal pulp area, determines the distance characteristics between the two sets of center points, and calibrates the single tooth associated with such tooth area as a normal tooth or a missing tooth based on the numerical expression of the distance characteristics; The feature midpoint confirmation end reanalyzes the single tooth calibrated as missing tooth, and locks the symmetry midline based on the overall edge contour of the single tooth and the center point of the internal pulp area. Then, based on the determined symmetry midline, the feature center point of the missing tooth is locked based on this symmetry midline. The specific method is as follows: The center point of the pulp area inside the missing tooth is marked as the main point, and the feature point closest to the main point is located in the outer area of the missing tooth, and this feature point is used as the secondary point of the main point; Connect the primary point and the secondary point, and extend the line. The extended line is used as the symmetrical midline of the missing tooth. Based on the determined symmetry midline, different tooth regions located on both sides of the symmetry midline are identified from the tooth image belonging to the missing tooth; based on the external contours of the tooth regions on both sides, a tooth region on the side with a larger area expression value is determined, and this tooth region is used as the area to be mirrored; based on the determined symmetry midline, the area to be mirrored is mirrored to confirm the mirrored area; the area formed by merging the area to be mirrored and the mirrored area is used as the main feature area; based on the overall edge contour of the main feature area, a feature center point of the main feature area is locked, and the locked feature center point is used as the feature center point of the missing tooth; The missing teeth after the feature center point calibration is completed are transmitted to the tooth feature analysis and processing end; The tooth feature analysis and processing end performs arrangement analysis on missing teeth after feature center point calibration and normal teeth after center point calibration. It prioritizes identifying the upper or lower gums from dental CT images, and numerically analyzes the center point distances of corresponding teeth in the upper or lower gums to determine and display the analysis results.

2. The dental data analysis system based on medical big data according to claim 1, characterized in that: The tooth region determination end screens the tooth region in the following manner: Based on the confirmed dental CT image, the CT values associated with different points in the dental CT image are determined and calibrated as H i , where i represents different points in the dental CT image; Will CT meet: H i Points ∈[1500Hu, 3000Hu] are calibrated as tooth points, where Hu is a numerical unit. Based on the calibrated groups of tooth points, the image areas covered by the groups of tooth points are confirmed. Based on the overall edge contours of such image areas, the areas inside the overall edge contours in the dental CT images are calibrated as tooth areas.

3. The dental data analysis system based on medical big data according to claim 2, characterized in that: The tooth area determination end determines the dental pulp area in the following specific ways: Locking H from the tooth area i ∈[30Hu, 100Hu], the locked points are marked as pulp points, and the area covered by several groups of pulp points is marked as the pulp area; The pulp areas associated with different tooth areas were calibrated in turn using the same processing method.

4. The dental data analysis system based on medical big data according to claim 2, characterized in that: The tooth calibration processing end calibrates a single tooth in the following specific manner: Based on the determined tooth region, the edge contour of the region is locked, and an edge contour line belonging to the tooth region is generated. The edge contour line is placed in a set of two-dimensional coordinate systems. The two-dimensional coordinates of different contour points within the edge contour line are determined from the two-dimensional coordinate systems. Then, several sets of two-dimensional coordinates associated with the different contour points are averaged to determine the average coordinates. Then, based on the location of the average coordinates, point calibration is performed within the tooth region to determine the center point A belonging to the tooth region. Then, from the pulp area confirmed within the tooth area, the edge contour of the pulp area is preliminarily confirmed, and the edge contour line of the pulp area is generated. Then, the edge contour line is placed in a two-dimensional coordinate system in the same manner, and the center point B of the pulp area is simultaneously confirmed in the two-dimensional coordinate system. Based on the specific positions of center point A and center point B, confirm the straight-line distance L between the two sets of center points A and B, and identify whether this straight-line distance L satisfies: L≤Y1, where Y1 is a preset value. If it satisfies, the teeth associated with this tooth area are calibrated as normal teeth.

5. The dental data analysis system based on medical big data according to claim 4, characterized in that: If the straight-line distance L does not satisfy: L≤Y1, the teeth associated with this tooth area are marked as missing teeth.

6. The dental data analysis system based on medical big data according to claim 1, characterized in that: The tooth feature analysis processing end performs numerical analysis on the center point distance of corresponding teeth located in the upper or lower gums in a specific manner as follows: The upper or lower gum is locked from the dental CT image. The upper or lower gum has been calibrated in advance. The upper gum is the area where the teeth are arranged above, and the lower gum is the area where the teeth are arranged below. Confirm the characteristics of the internal teeth located in the upper gum. The internal teeth include normal teeth or missing teeth, and mark the characteristic center points of the missing teeth and the center points of the normal teeth as the midpoints of the internal teeth: identify the straight-line distances between the inner midpoints of the adjacent internal teeth in the corresponding gums, and then mark the confirmed straight-line distances as L q , where q represents different straight-line distances, and the confirmed L q Perform mean processing to confirm the characteristic parameter L1 of the upper gum; Treat the inner teeth of the lower gum in the same way as the upper gum, and confirm the characteristic parameter L2 of the lower gum; Identify whether L1 and L2 satisfy: |L1-L2|>Y2, where Y2 is a preset value. If so, generate and display an arrangement abnormality signal.

7. The dental data analysis system based on medical big data according to claim 6, characterized in that: If L1 and L2 do not satisfy: |L1-L2|>Y2, a normal arrangement signal is generated.

Images