Method and apparatus for generating 2D panoramic images

Abstract

The present invention provides a method for creating a 2D panoramic image of an intraoral cavity (6), comprising the steps of: A) providing a plurality of projection images acquired while a recording unit comprising a radiation source (2) and an electronic image sensor (3) rotates at least partially around the intraoral cavity (6) in a given position; B) calculating an image comprising at least one anatomical feature from the plurality of projection images using a calculation rule having reconstruction parameters; C) providing a model of at least one anatomical feature describing the at least one anatomical feature at a target position in the image; D) identifying the at least one anatomical feature in the calculated image; E) applying a metric to the calculated image. step F) determining a deviation between an actual position of the at least one identified anatomical feature in the computed image and a target position of the at least one anatomical feature in the model; step F) modifying at least one of the reconstruction parameters and recalculating at least one portion of the image showing the at least one anatomical feature using the computation rule and the at least partially modified reconstruction parameters or modifying a pixel transformation and applying the pixel transformation to at least one portion of the computed image showing the at least one anatomical feature so that at least this portion of the image is recalculated; step G) deriving a 2D panoramic image from the recalculated image; and step H) outputting the 2D panoramic image.

Description

[Technical Field]

[0001] The present invention relates to a method for generating a 2D panoramic image of an intraoral cavity, the method comprising: Step A) Providing a plurality of projection images acquired while a recording unit comprising a radiation source and an electronic image sensor rotates at least partially around the oral cavity in a given position; Step B) calculating an image including at least one anatomical feature from the plurality of projection images using a calculation rule having reconstruction parameters.

[0002] The invention further relates to an apparatus for producing a 2D panoramic image of an intraoral cavity, the apparatus comprising a recording unit having a radiation source and an electronic image sensor, and a computer operatively connected to the recording unit, the computer being configured to carry out the method according to the invention. [Background technology]

[0003] In the field of dental diagnostic radiology, it is often necessary to create an X-ray image of the entire oral cavity, including the jawbone and the teeth embedded therein. For this purpose, the patient's head, including the oral cavity, is positioned so that a recording unit comprising a radiation source and an electronic image sensor can rotate around the patient's head, thereby acquiring multiple projection images of the mandibular and / or maxillary dental arches at multiple angles. If the acquisition is performed using an electronic image sensor comprising a matrix arrangement of pixels, a 2D panoramic image can be calculated from the multiple recorded projection images. This 2D panoramic image clearly depicts selected curved slices of the dental arch in a two-dimensional unfolded view.

[0004] The calculation of a 2D panoramic image from multiple projection images follows a calculation rule or algorithm that uses reconstruction parameters. On the one hand, the reconstruction parameters describe the geometric mapping of the individual projections, i.e., the recording geometry. The imaging characteristics of the 2D panoramic image to be calculated, which are selected by the user or automatically during the calculation of the 2D panoramic image, can also be defined and optionally changed via the reconstruction parameters. An example of this is the automatic selection of a curved slice in the dental arch that is clearly represented in the 2D panoramic image. Thereby, initial imaging characteristics are defined, assuming that the oral cavity is located in the target position.

[0005] However, in practice, this is often not the case: even with a well-trained person, there will continue to be more or less significant deviations between the desired target position of the oral cavity and the given position actually achieved. Since the patient may move during the recording, the difference between the given position and the target position while acquiring the projections may also vary from projection to projection.

[0006] In particular, it has been shown that the diagnostic evaluation of 2D panoramic images also depends on the position and orientation of the oral cavity relative to the radiation source or image sensor while the recording unit rotates around the oral cavity. In a 2D panoramic image of an oral cavity that is displaced relative to the ideal position and orientation, it may be impossible to distinguish pathological features of the dentition from features of the 2D panoramic image caused by the displaced position. Thus, deviations from the ideal position and orientation may lead to situations where a diagnosis based on the calculated 2D panoramic image is impossible or can only be made by very experienced diagnosticians. Japanese Patent Application Laid-Open No. 2010-148676 discloses an X-ray imaging apparatus and a panoramic image processing program. Japanese Patent Application Laid-Open No. 2016-007338 discloses a panoramic X-ray imaging apparatus and a panoramic X-ray imaging method.

[0007] In contrast to the misaligned presentation of the oral cavity in 2D panoramic images, it is an object of the present invention to provide a method that facilitates diagnosis.A further object of the present invention is to provide a method that facilitates diagnosis, which is independent of the position and orientation of the oral cavity relative to the recording unit during acquisition of the projection images. Summary of the Invention

[0008] At least one of the above-mentioned objects is achieved by a method for generating a 2D panoramic image of an intraoral cavity, the method comprising: Step A) Providing a plurality of projection images acquired while a recording unit comprising a radiation source and an electronic image sensor rotates at least partially around the oral cavity in a given position; Step B) calculating an image including at least one anatomical feature from the plurality of projection images using a calculation rule having reconstruction parameters; Step C) providing a model of at least one anatomical feature describing at least one anatomical feature at a target location in the image; Step D) identifying at least one anatomical feature in the computed image; Step E) determining a deviation between an actual position of at least one identified anatomical feature in the calculated image and a target position of at least one anatomical feature in the model by applying a metric; step F) modifying at least one of the reconstruction parameters and recalculating at least one portion of the image showing the at least one anatomical feature using the calculation rule and the at least partially modified reconstruction parameters, or applying a pixel transformation to at least one portion of the calculated image showing the at least one anatomical feature, so that at least this portion of the image is recalculated; Step G) deriving a 2D panoramic image from the recomputed images; and Step H) Outputting the 2D panoramic image. However, according to the invention, the deviation between the calculated actual position of at least one anatomical feature in the image and the target position of said anatomical feature in the model is optimized by iteration. To this end, in one embodiment, the method according to the invention comprises, between steps F) and G), the following steps: Step I) re-determining the deviation between the actual position of at least one identified anatomical feature in the calculated image and the target position of at least one anatomical feature in the model by applying a metric; Step J) repeating step F) and step I) until the determined deviation is less than a specified threshold; and Step K) if the entire image was not recalculated in step F), recalculating at least one portion of the image that was not recalculated in step F) using the at least partially changed reconstruction parameters or applying at least one pixel transformation to at least one portion of the image that was not recalculated in step F).

[0009] Providing a plurality of projection images is understood to initially only mean that the projection images are transmitted in the form of data for further processing, for example to a computer implementing the method according to the invention.

[0010] However, in one embodiment, providing a plurality of projection images comprises acquiring said plurality of projection images while the recording unit rotates at least partially around the oral cavity.

[0011] Of course, in at least one embodiment of the present invention, the radiation source produces X-ray radiation during operation of the recording unit, and the image sensor is a sensor for said X-ray radiation.

[0012] Such an electronic image sensor may comprise, for example, a CCD or CMOS detector, in which a scintillation layer in front of the CCD or CMOS detector converts the X-ray radiation into electromagnetic radiation in the wavelength range that the detector is sensitive to. Alternatively, such an electronic image sensor may be a direct conversion detector, in which the X-ray radiation is directly converted into an electrical signal. In one embodiment of the invention, the image sensor is a so-called full-frame sensor, which means that all pixels of the sensor are simultaneously exposed and read out in their entirety at a given time. In one embodiment of the invention, the individual pixels of the image sensor are arranged in a matrix.

[0013] While a recording unit comprising a radiation source and an image sensor rotates at least partially around the oral cavity, multiple projection images depicting the oral cavity in multiple radiation transmission directions are created, and each projection is preferably stored in the form of raw data.

[0014] The actual 2D panoramic image is generated from the projection images only at a later time, sometimes long after the projection images are acquired. Sharp curved slices of the dental arch, also referred to in this application as panoramic layers, are mathematically mapped onto the planes of the image.

[0015] By modifying one or more of the reconstruction parameters or applying pixel transformations in step F), it is possible to select and modify not only the position of the sharp layers, i.e. the image planes that are sharply imaged in the 2D panoramic image, but also the position of the oral cavity or its individual elements, such as teeth, maxilla and mandible, that are depicted in the computed 2D panoramic image.

[0016] In the context of the present application, a given position of the head, i.e. of the oral cavity and the elements located therein, is always understood to be the position and orientation of the head relative to the recording unit. On the other hand, the actual position of an anatomical feature refers to the true position of said anatomical feature in the calculated image. The target position refers to the desired position, i.e. the position of the anatomical feature that should be achieved in the calculated image.

[0017] In principle, it is desirable that all 2D panoramic images are generated from projection images of the oral cavity with the oral cavity in an optimal position relative to the recording unit. However, since deviations of the true given position from the ideal position of the patient's head and thus the oral cavity with the jawbone for diagnosis always occur, the invention proposes to present the entire oral cavity and its elements in the displayed 2D panoramic image by changing the position of at least one anatomical feature, with which the presentation in the displayed 2D panoramic image corresponds to the desired presentation.

[0018] To achieve a desired representation in the 2D panoramic image, an image with at least one anatomical feature is first calculated from the multiple projection images with reconstruction parameters according to a calculation rule, and a deviation between the actual position of the at least one anatomical feature identified in the image and the target position of the at least one anatomical feature described by the model is calculated by applying a metric. Needless to say, this deviation corresponds to a correlation that describes the degree of correspondence between the anatomical feature in the calculated image and the anatomical feature in the model.

[0019] In the present invention, it is initially immaterial whether the image calculated in a step prior to outputting the 2D panoramic image and used for subsequent calculations is a 2D image, e.g. a 2D panoramic image, or a 3D image, e.g. a 3D volume, or an abstract image description in the form of image data.

[0020] By applying a metric, the actual position of at least one anatomical feature in the image is modified after determining a deviation between the calculated actual position of the at least one anatomical feature in the image and a target position of said anatomical feature in the model.

[0021] According to the invention, there are two options for this modification to the actual position of at least one anatomical feature in the computed image.

[0022] First, in one embodiment, at least one of the reconstruction parameters can be modified in order to change the actual position of at least one anatomical feature in the calculated image, and at least a partial recalculation of the image is performed according to the calculation rules using the at least partially modified reconstruction parameters.

[0023] Second, in one embodiment, the pixel transformation is modified and then applied to at least a portion of the calculated image to change the position of at least one anatomical feature in the calculated image, causing at least this portion of the image to be recalculated.

[0024] In the first variant, the image is recalculated from the projection images after modifying at least one reconstruction parameter as if it had been so from the beginning, whereas in the second variant, the image calculated in step B) is used as the starting point for the calculation and is modified by applying a local or global pixel transformation.

[0025] In one embodiment, such pixel transformations are local and / or global image deformations or distortions, often referred to as image warping. In particular, such pixel transformations do not improve image quality because pixels are interpolated. In exchange, applying the pixel transformations has no effect on other imaging characteristics.

[0026] Of course, in one embodiment, the initial pixel transformation, e.g., imaging each pixel onto itself (which is used when there is no deviation between the actual position and the target position), is predefined and is modified as a function of the determined deviation.

[0027] The method according to the invention can be described as a correction of deviations of a given position of the oral cavity in a projection image determined by the recording geometry from an optimal target position defined by a model. In each case, the deviation between the actual position in the respective calculated image and the target position defined by the model is determined on the basis of at least one anatomical feature, which can be identified in the calculated image and which is also described by the model.

[0028] In one embodiment, to obtain an image and thus a 2D panoramic image having actual positions of anatomical features and actual positions of the oral cavity that substantially correspond to the target positions in the displayed 2D panoramic image, it may be sufficient to determine the deviations in step E) and modify at least one of the reconstruction parameters or modify the pixel transformation only once and apply the modified pixel transformation in step F).

[0030] In the context of this application, a model refers to an abstracted, i.e., simplified, description of one or more anatomical features of the head compared to a complete image. Determining a metric characterizing the deviation between a computed image and the description of the anatomical features in the model is significantly faster than, for example, comparing two complete images acquired at different times.

[0031] In one embodiment of the present invention, in addition to describing one or more anatomical features, the model may also describe other factors, but they are not used to determine deviations.

[0032] In one embodiment, the model may be computationally extracted from a plurality of images, in particular based on a plurality of projection images, in particular 2D panoramic images, respectively. In another embodiment, the model is based on a definition specified by a user.

[0033] In one embodiment, the amount of data of the model is smaller than the amount of data of the calculated image, preferably a 2D panoramic image. In another embodiment, the model describes at least one anatomical feature with a first number of pixels, and the calculated image, in particular a 2D panoramic image, comprises a second number of pixels, the first number of pixels being smaller than the second number of pixels.

[0034] In so doing, the model optionally describes the anatomical features in a 2D or 3D description. If the anatomical features are described in the model in the form of a 3D description but the calculated image is a 2D image, in step E) and, if applicable, in step I) of an embodiment of the method for determining a metric, first a 2D description is calculated from the 3D description, which represents a 2D projection of at least one anatomical feature, which corresponds to the presentation of the at least one anatomical feature in an ideal 2D panoramic image.

[0035] In one embodiment, after changing the at least one reconstruction parameter in step F), the entire image may be recalculated and the image thus calculated may be compared again with the model, or only the part of the image showing the at least one anatomical feature is recalculated. In the second case, the at least one part of the image not yet calculated in step F) is recalculated with the optimized reconstruction parameters only when the at least one reconstruction parameter is optimized such that a metric of the deviation between the actual position of the at least one anatomical feature in the calculated image and the target position specified by the model is optimized.

[0036] In step F), a pixel transformation can similarly be applied to the complete calculated image, changing the entire image, or the pixel transformation can be applied only to the portion of the image that shows at least one anatomical feature. The pixel transformation is not applied to the entire image calculated in step B), nor to at least one portion of the image from step B) that did not first undergo pixel transformation(s) in step F), until a metric of deviation between the actual position of the at least one anatomical feature in the calculated image and a target position specified by the model is optimized.

[0037] In one embodiment, the reconstruction parameters of the calculation rule describe both the recording geometry during acquisition of the projection images and the desired imaging characteristics of the calculated images.

[0038] In one embodiment of the present invention, the recording geometry is determined by the radiation source, the image sensor, the intraoral cavity, and the relative positioning of these elements with respect to each other. In one embodiment of the present invention, the imaging characteristics include the actual position of anatomical features and / or the intraoral cavity in the image, the position of layers that are clearly imaged in the displayed 2D panoramic image, the radiation transmission angle, and the thickness of layers that are clearly or substantially clearly imaged in the displayed 2D panoramic image. Imaging characteristics other than the position of anatomical features or the intraoral cavity are also referred to as further imaging characteristics in the context of this application.

[0039] In one embodiment of the present invention, the at least one reconstruction parameter modified in step F) describes a transformation parameter of a coordinate transformation for rotation and / or translation of the oral cavity or head relative to the recording unit.

[0040] Of course, in step D), any suitable procedure for image analysis and difference analysis can be used to determine the deviation between at least one anatomical feature identified in the calculated image and a feature in the model. Any suitable distance measure can be used as the metric, for example the sum of all Euclidean distances between reference points in the calculated image and reference points in the model.

[0041] The modification of at least one reconstruction parameter or application of a pixel transformation in step F) can be performed manually by a user or by an automatic optimization method.

[0042] In one embodiment, the deviation optimization is performed using optimization methods known from the art, for example brute force.

[0043] The calculation rules for calculating the images comprise a plurality of reconstruction parameters. In one embodiment, the calculation rules are designed such that changing at least one reconstruction parameter in step F) leaves at least one further imaging characteristic different from the position of the intraoral cavity, preferably all further imaging characteristics of the respective calculated image and the 2D panoramic image displayed in step H) different from the position of the intraoral cavity, substantially unchanged.

[0044] In one embodiment, at least one further imaging characteristic of the displayed 2D panoramic image that is not affected by a change in the at least one reconstruction parameter is selected from the position of a layer that is clearly imaged in the 2D panoramic image, the radiation transmission angle, and the thickness of a layer that is clearly or substantially clearly imaged in the 2D panoramic image.

[0045] If the definition of the recording geometry of each projection image is known, for example in the form of a system matrix, it is possible to independently change and optimize the position of the oral cavity and additional imaging characteristics by changing at least one reconstruction parameter, preferably multiple reconstruction parameters. The system matrix allows for the interpretation of data in 3D space and therefore also describes the correlations between the imaging parameters. The assumed position of the oral cavity, which is the basis for the calculation, is essentially changed until the difference between the model of the anatomical features and the presentation of the anatomical features in the 2D panoramic image to be displayed is minimized. The reconstruction parameters for correcting the position, and, if applicable, at least one additional imaging characteristic, are changed with each recalculation of the image and can be determined similarly if the system matrix is ​​known.

[0046] In one embodiment, at least one of the position of the oral cavity and the further imaging characteristic is adjusted or corrected simultaneously.

[0047] In one embodiment of the present invention, the modification of the reconstruction parameters in step F) not only leads to a modification of the assumed position of the oral cavity on which the calculation is based, but also simultaneously to a modification of at least one further imaging characteristic, for example the position of the sharply delineated layer.

[0048] Thus, for example, in one embodiment, the range of radiation transmission angles of an individual projection image available for calculation of an individual pixel of the image is a function of both the location of the sharp layer selected for calculation and the physical recording geometry.

[0049] Each radiation transmission angle, on which the calculation of the individual pixels of the image is based, can be varied as a function of the given data within a maximum radiation transmission angle locally defined by the recording geometry.

[0050] However, the local layer thickness of a pixel of the 2D panoramic image is a function of the radiation transmission angle of the underlying projection image selected for the calculation of this pixel of the 2D panoramic image: the smaller the radiation transmission angle of the projection image selected for calculating a pixel of the 2D panoramic image, the larger the resulting layer thickness at the calculated pixel.

[0051] Therefore, the selection of the sharp layer location in the 2D panoramic image indirectly affects the local layer thickness at each pixel.

[0052] Thus, in one embodiment, the calculation rules are selected such that the modification of at least one reconstruction parameter does not substantially affect any of the further imaging characteristics, as well as the correction of image distortion.

[0053] In one embodiment, a change to the at least one further imaging characteristic is substantially compensated for by additionally changing at least one further reconstruction parameter. In other words, optimization of the reconstruction parameters may result in an undesired change to the further imaging characteristic, which may be substantially offset by changing a reconstruction parameter that has no or little effect on the image of the anatomical feature.

[0054] In the context of this application, in addition to very specific anatomical structures, the term anatomical feature also includes image structures or image regions containing such image structures that can be attributed to the anatomical structure of the head with the recorded oral cavity, where the position, orientation, shape, or symmetry of one or more parts of the body are not specifically identified as producing the image structure. An example of such an image structure that forms an anatomical feature in the context of this application is the shape of a shadow that occurs reproducibly in all images of the oral cavity.

[0055] In one embodiment, the term anatomical feature in the context of the present application also includes structures such as implants or dental crowns that are artificial but permanently attached to the human body.

[0056] The anatomical feature in some embodiments may be due to elements of the oral cavity, but in one embodiment the anatomical feature may also be due to elements of the head surrounding the oral cavity or spine.

[0057] In one embodiment of the present invention, the anatomical features are selected from the position and / or orientation and / or shape and / or symmetry of the teeth, the temporomandibular joint, the palate, at least one element of the spinal column, such as a vertebra or disc, a clavicle, and / or bones.

[0058] In one example, the anatomical feature is the shape of the palate of the oral cavity, where, when the oral cavity is in an optimal position relative to the recording unit, the palate exhibits a symmetrical shape with a substantially horizontal profile in the 2D panoramic image, whereas a dorsal tilt of the head, for example, will cause the palate to appear gabled in the computed 2D panoramic image.

[0059] In one embodiment, the profile of the maxillary and / or mandibular teeth serves as an anatomical feature in the context of the present application. When the oral cavity is in an optimal position relative to the recording unit, the teeth exhibit a slight smile. In other words, the profile of the maxillary teeth has a slightly convex curved shape, and the profile of the mandibular teeth has a slightly concave curved shape. On the other hand, with a dorsal tilt of the head, the maxillary teeth have a slightly concave profile and the mandibular teeth have a slightly convex profile.

[0060] The profile of the roots of the anterior teeth may also be considered an anatomical feature: when the oral cavity is in an optimal position relative to the recording unit, the roots of the anterior teeth have a maximum inclination relative to a perpendicular line parallel to the axis of rotation of the recording unit.

[0061] In particular, bony structures are often symmetrically positioned within the head. In a 2D panoramic image in which the oral cavity is optimally positioned, the bony structures of the head are also imaged symmetrically. This knowledge can be incorporated into an anatomical model and used to identify positionally optimized reconstruction parameters or position-optimizing pixel transformations. Examples of such symmetrical bony structures when the head, and thus the oral cavity, is in an optimal position are the temporomandibular joint, jaws, palate, and eye sockets. Inferences can also be made about the position of the oral cavity from the position of the vertebrae.

[0062] If the oral cavity is in the proper position, for example, the projection of the temporomandibular joint should be horizontally symmetric in the panoramic image. From the position and size of the temporomandibular joint image, inferences can be made about the position of the head or oral cavity.

[0063] In one embodiment of the method according to the present invention, In step D), at least one anatomical feature is identified in a first section of the computed image and at least one anatomical feature is identified in a second section; In step E), a deviation between the at least one anatomical feature identified in the first section and a model of the at least one anatomical feature is determined by applying a metric, and a deviation between the at least one anatomical feature identified in the second section and a model of the at least one anatomical feature is determined by applying a metric; In step F), at least one reconstruction parameter for a first section and at least one reconstruction parameter for a second section of the calculated image are modified independently of each other, In step I), a deviation between the at least one identified anatomical feature in the first section and the model is again determined by applying a metric, and a deviation between the at least one identified anatomical feature in the second section and the model is again determined by applying a metric; In step J), steps F) and I) are repeated until each determined deviation is below a specified threshold.

[0064] The same reconstruction parameters or partially or completely different reconstruction parameters may be modified for the first and second sections.

[0065] In particular, such embodiments are suitable for correcting changes in the position of the oral cavity during rotation of the recording unit so that the displayed 2D panoramic image appears as if the underlying projection images were acquired using a single, optimized, stationary position of the oral cavity. In such embodiments, the first and second sections of the calculated image are positioned directly or indirectly adjacent to each other in the direction of the profile of the dental arch or in the direction of rotation of the recording unit, i.e., in the direction of the horizontal profile of the dental arch, so that the first and second sections correspond to different points in time during acquisition of the respective underlying projection images. Needless to say, such embodiments are not limited to two sections, but are particularly suitable for multiple sections. Such embodiments are particularly not limited to first and second sections being directly adjacent to each other.

[0066] At least one reconstruction parameter is varied for each individual section of the panoramic image until a metric of deviation between the actual position of the anatomical feature in the first section of the calculated image and the target position of the feature in the model and a metric of deviation between the actual position of the anatomical feature in the second section of the calculated image and the target position of the feature in the model are optimized.

[0067] In another embodiment of the present invention, the first section of the calculated image is an upper section including at least one portion of the maxilla, and the second section of the calculated image is a lower section including at least one portion of the mandible. Thus, different positions of the maxilla and mandible due to occlusion can be corrected independently of each other. Furthermore, movements of the mandible relative to the maxilla during acquisition of the projection images can be corrected independently of each other.

[0068] In one embodiment, the upper section includes the entire upper jaw and the lower section includes the entire lower jaw.

[0069] The aim of the method according to the invention is to obtain a 2D panoramic image that facilitates diagnosis, taking into account deviations in the patient's head position during the acquisition of projection images, by using an optimized calculation of the displayed 2D panoramic image, in particular facilitating comparison with panoramic layer images based on pre- or post-acquisition of projection images.

[0070] However, the described positional corrections during the calculation of the panoramic layer images may also have local and / or global adverse effects on image quality. In the first correction variant, this is due to the fact that the radiation source and image sensor in the acquisition of the projection images are adjusted to a target position in the oral cavity, and any deviation from the target position results in a deterioration of image quality. However, in the case of small deviations, this is barely noticeable. However, if the deviation becomes too large, the image quality will deteriorate significantly. In the second variant, the potential deterioration of local image quality is the result of pixel transformations.

[0071] Because the user relies on the correction of the 2D panoramic image presentation displayed by the method according to the present invention, there is an additional risk that the user will not pay sufficient attention to properly positioning the patient. Since image quality remains good with small deviations, the user also does not realize that the positioning was not optimal. As a result, more complex and larger corrections are required, which increases the impact on image quality. To avoid this, in one embodiment of the present invention, a scale for modifying at least one reconstruction parameter, i.e., a scale for the performed correction of the oral cavity position, preferably with instructions on the direction, amount, and / or type of modification, is displayed together with the 2D panoramic image. Thus, the user receives feedback on how well the oral cavity was positioned within the device used.

[0072] In another embodiment, image regions that have undergone correction due to positional changes during acquisition of the projection images are marked or highlighted in the displayed 2D panoramic image itself. Such highlighting can be achieved, for example, by coloring the corresponding image regions. The marking can also indicate the measure of the correction that has been performed. For example, red coloring can represent regions where the required correction is so great that the desired image quality is no longer achieved, while yellow coloring can represent regions where correction has been performed but the image quality is still acceptable.

[0073] Of course, if the image calculated in step F) or step K) is already a 2D panoramic image, then the derivation of the 2D panoramic image in step G) is a trivial operation that does not require further calculations. However, if the image calculated in step F) or step K) is a 3D image, then a 2D panoramic image must be calculated from said 3D image, for example in step G).

[0074] In one embodiment, outputting the 2D panoramic image in step H) comprises communicating the dataset describing the 2D panoramic image to, for example, a display or a further data processing device.

[0075] The above-described embodiments of the method according to the invention may be at least partly realized by using a software-controlled data processing device, i.e. a computer, and it is therefore clear that a computer program providing such software control for running on a computer, as well as a storage medium on which such a computer program is stored, should also be considered to be aspects of the invention.

[0076] At least one of the above objects is also achieved by an apparatus for producing a 2D panoramic image of an oral cavity, the apparatus comprising: a recording unit having a radiation source and an electronic image sensor, the recording unit being constructed and designed so that a plurality of projection images can be acquired while the recording unit rotates at least partially around the oral cavity in a given position; a computer; A computer is operatively connected to the image sensor such that the computer can receive a plurality of projected images from the image sensor, the computer being configured to: Step A) receiving a plurality of projection images from a recording unit; Step B) calculating an image including at least one anatomical feature from the plurality of projection images using a calculation rule having reconstruction parameters; Step C) providing a model of at least one anatomical feature describing at least one anatomical feature at a target location in the image; Step D) identifying at least one anatomical feature in the computed image; Step E) determining a deviation between an actual position of at least one identified anatomical feature in the calculated image and a target position of at least one anatomical feature in the model by applying a metric; Step F) modifying at least one of the reconstruction parameters and recalculating at least one portion of the image showing the at least one anatomical feature using the calculation rule and the at least partially modified reconstruction parameters; or modifying the pixel transformation and applying the pixel transformation to at least a portion of the calculated image showing the at least one anatomical feature such that at least this portion of the image is recalculated; Step G) deriving a 2D panoramic image from the recomputed images; and Step H) Outputting the 2D panoramic image.

[0077] Insofar as aspects of the invention have been described above with respect to a method, said aspects also apply to a corresponding apparatus for producing a 2D panoramic image of the intraoral cavity. Insofar as the method is carried out using an apparatus according to the invention, said apparatus comprises a suitable device for this purpose.

[0078] Further advantages, features and possible applications of the present invention will become apparent from the following description of embodiments and the associated figures. [Brief explanation of the drawings]

[0079] [Figure 1] FIG. 1 is a schematic cross-sectional view of an apparatus for producing 2D panoramic images according to one embodiment of the present invention. [Figure 2] FIG. 2 is a schematic front view of the device of FIG. [Figure 3] 3a) to 3c) are schematic illustrations of 2D panoramic X-ray images of a patient's oral cavity in different positions. [Figure 4] FIG. 4 is a schematic illustration of a 2D X-ray panoramic image of the oral cavity with a series of anatomical features. [Figure 5] Figure 5a) is a schematic illustration of the soft palate of Figure 4 in a 2D panoramic X-ray image when the patient's head is tilted backward, and Figure 5b) is a schematic illustration of a 2D panoramic X-ray image of the soft palate with head tilt corrected. [Figure 6]Figure 6a) is a schematic illustration of the maxillary and mandibular tooth and bone profiles in a 2D X-ray panoramic image when the patient's head is tilted backward, and Figure 6b) is a schematic illustration of a 2D X-ray panoramic image of the tooth and bone profiles of Figure 6a) with head tilt corrected. [Figure 7] Figure 7a) is a schematic diagram of a 2D panoramic X-ray image of the temporomandibular joint when the head is tilted to the side, and Figure 7b) is a schematic diagram of a 2D panoramic X-ray image of the temporomandibular joint with head tilt corrected. DETAILED DESCRIPTION OF THE INVENTION

[0080] Identical elements in the figures are labeled with identical reference numbers.

[0081] Figure 1 shows a schematic horizontal section of an apparatus 1 according to the invention for creating 2D panoramic images. Figure 2 shows the same apparatus 1 in a front view. The apparatus 1 essentially comprises a source 2 for X-ray radiation and an electronic image sensor 3 for said X-ray radiation. In the context of the present invention, the source 2 and the image sensor 3 together form a recording unit. The sensor 3 is connected via a corresponding data interface to a computer 4, which calculates the 2D panoramic image from the projection images acquired by the sensor 3.

[0082] During the recording of the patient's head 5, i.e., the patient's oral cavity 6, the radiation source 2 and the sensor 3 are rotated along a predefined curved path, whereby the radiation source 2 and the sensor 3 are respectively placed on either side of the head 5, such that the X-ray radiation from the radiation source 2 to the sensor 3 is transmitted through the head 5. The sensor 3 detects the X-ray radiation transmitted from the radiation source 2 by the head 5 from multiple angular positions along the rotation path 7, so that after the rotation multiple projection images are available to the computer 4 as raw data for the calculation of 2D X-ray panoramic layer images. In the context of the present application, the illustrated arrangement of the radiation source 2, the sensor 3 and the oral cavity 6 as well as the predefined rotation path 7 determine the recording geometry.

[0083] Figures 3a) to 3c) show schematically three 2D panoramic X-ray layer images, which show the dependence of the image of the oral cavity 6 and its elements in the panoramic image from the position of the head 5 relative to the source 2 and sensor 3 during the acquisition of the multiple projection images.

[0084] Figure 3a) shows a panoramic image where both the head tilt, i.e., orientation, and head position, i.e., head location, deviate from the optimal position. On the other hand, the panoramic image in Figure 3b) is based on the optimal head position, but the head is still tilted. In the panoramic image in Figure 3c), both the head position and head tilt are optimized.

[0085] The aim of the present invention is to obtain an optimized panoramic layer image showing the jawbone and teeth when the head position and head tilt are optimized, as in the example of Fig. 3c), however this is independent of the actual head tilt and head position, i.e. the given position of the head during the generation of the projection images using the source 2 and sensor 3, and thus the recording geometry.

[0086] As a result, the deviation between the actual position in the first calculated 2D panoramic image and the optimal or target position must be determined, and then a corrected 2D panoramic image must be calculated by changing at least one reconstruction parameter in the calculation rule. The corrected 2D panoramic image should finally look similar to the illustration in Figure 3c).

[0087] From certain anatomical features of the head 5, in their presentation in images, particularly 2D panoramic images, inferences can be drawn about the given position of the head 5 during the acquisition of multiple projection images and about the depicted actual position of the head or oral cavity in the calculated images. Figure 4 shows a schematic panoramic image of the oral cavity, with the oral cavity 6 optimally positioned relative to head tilt and head position. Possible features that can be used to determine the deviation between the optimal target position and the actual position during the acquisition of projection images are the shape and symmetry of the soft palate 8, the dental and bone profiles of the jawbone 9 and teeth 10, and the position and symmetry of the temporomandibular joints 11, 12.

[0088] The model can describe the shape and symmetry of one or more of these anatomical features 8, 9, 10, 11, 12 in an optimal image, particularly an optimal 2D panoramic image representation. This model defines the target position of the selected anatomical feature and, therefore, the target position of the oral cavity 6. Because the model need only include a simplified representation of one or more anatomical features rather than a complete 2D panoramic image of the oral cavity 6, the model comprises significantly less data than a complete 2D panoramic image. This reduces the time required to determine deviations between the calculated image and the model.

[0089] In order to identify positional misalignments, i.e. deviations between a given position of the head 5 during acquisition of multiple projection images and a target position, when calculating the 2D panoramic image, the calculated 2D panoramic image is first compared with a model, and then at least one reconstruction parameter is modified until the anatomical features in the calculated 2D panoramic image approximately correspond to the shape and position of said features described in the model.

[0090] An example of this for the soft palate 8 is shown in Figures 5a and 5b. After initially calculating a 2D panoramic image from multiple projection images, the soft palate 8' appears as a slightly sloped gable roof in the representation in Figure 5a. Comparison with the model shows that a given position during the acquisition of the multiple projection images was a dorsal tilt of the head. The reconstruction parameters responsible for the actual position in the calculated image are changed, and the 2D panoramic image is then recalculated. This iteration is performed until the deviation between the model and the representation of the soft palate 8 in the 2D panoramic image is below a predetermined threshold. Ideally, therefore, Figure 5b shows both the representation of the soft palate 8 in the model and the representation of the soft palate 8 in the corrected 2D panoramic image, in which the misalignment of the head 5 during the acquisition of the multiple projection images has been removed.

[0091] Figures 6a) and 6b) show the corresponding correction based on a model, which uses the jawbone 9 and tooth profiles as anatomical features for the correction. After calculation of the 2D panoramic image, the maxilla 9' and the teeth 10' housed therein exhibit a concave profile. On the other hand, the mandible and the teeth housed therein exhibit a convex profile. Comparison with the model again shows that the head was tilted backward during the acquisition of the multiple projection images. The reconstruction parameters responsible for the position are changed, and the 2D panoramic image is then recalculated. This iteration is performed until the deviation between the model and the representation of the jawbone and tooth profiles in the 2D panoramic image is minimized. Ideally, therefore, Figure 6b) shows both the tooth and bone profiles of the model and the tooth and bone profiles in the corrected 2D panoramic image, in which the misalignment of the head 5 during the acquisition of the multiple projection images has been eliminated.

[0092] Figures 7a and 7b show a model-based correction, which uses the position and symmetry of the temporomandibular joints 11 and 12 as anatomical features for correction. In the case of the dorsal tilt of the head in Figure 7a), the temporomandibular joints 11 and 12 appear at different heights. Therefore, the overall positioning is asymmetric. On the other hand, in the corrected 2D panoramic image in Figure 7b), which corresponds to the model in an ideal way, the positions of the temporomandibular joints 11 and 12 are symmetric. The following is a summary of the claims as originally filed: [1] A method for generating a 2D panoramic image of an oral cavity (6), comprising: Step A) Providing a plurality of projection images acquired while a recording unit comprising a radiation source (2) and an electronic image sensor (3) rotates at least partially around the oral cavity (6) in a given position; Step B) calculating an image including at least one anatomical feature from the plurality of projection images using a calculation rule having reconstruction parameters; Step C) providing a model of the at least one anatomical feature describing the at least one anatomical feature at a target location in the image; Step D) identifying said at least one anatomical feature in said computed image; Step E) determining a deviation between an actual position of the at least one identified anatomical feature in the calculated image and the target position of the at least one anatomical feature in the model by applying a metric; step F) modifying at least one of the reconstruction parameters and recalculating at least one portion of the image showing the at least one anatomical feature using the calculation rule and the at least partially modified reconstruction parameters, or modifying a pixel transformation and applying the pixel transformation to at least one portion of the calculated image showing the at least one anatomical feature, so that at least this portion of the image is recalculated; Step F-1) redetermining the deviation between the actual position of the at least one identified anatomical feature in the calculated image and the target position of the at least one anatomical feature in the model by applying a metric; Step G) deriving the 2D panoramic image from the recomputed images; and Step H) outputting the 2D panoramic image. A method for providing the above. [2] During steps F-1) to G), the method further comprises: Step F-2) repeating steps F) and F-1) until the determined deviation is less than a specified threshold; and step F-3) if the entire image was not recalculated in step F-1), recalculating at least one portion of the image that was not recalculated in step F) using the at least partially changed reconstruction parameters or applying the at least one pixel transformation to at least one portion of the image that was not recalculated in step F); The method according to [1], characterized in that it comprises: [3] The method according to [1] or [2], characterized in that in step A) providing the plurality of projection images comprises acquiring the plurality of projection images while the recording unit rotates at least partially around the oral cavity (6). [4] A method according to any one of [1] to [3], characterized in that the calculation rules are designed so that even if the at least one reconstruction parameter is changed in step F), at least one further imaging characteristic of the 2D panoramic image displayed in step H) that is different from the position of the intraoral cavity does not substantially change. [5] The method according to [4], characterized in that when changing the at least one reconstruction parameter, the influence of the at least one reconstruction parameter on the at least one further imaging characteristic is substantially compensated in step F). [6] The method according to [4] or [5], characterized in that the at least one further imaging characteristic of the 2D panoramic image displayed in step H) is selected from the position of a layer that is clearly imaged in the 2D panoramic image, the radiation transmission angle, and the thickness of the imaged layer. [7] The method according to any one of [1] to [6], characterized in that when the plurality of reconstruction parameters are changed in step F), in addition to the position of the intraoral cavity, a further imaging characteristic different from the position of the intraoral cavity also changes, and the calculation rule is designed so that the position of the intraoral cavity and the further imaging characteristic are changed independently of each other. [8] A method according to any one of [1] to [7], characterized in that the at least one anatomical feature is selected from the position and / or shape of teeth, temporomandibular joints, palate, at least one element of the spine, clavicle, and / or bone. [9] The method described in any one of [1] to [8], wherein the at least one anatomical feature includes anatomical symmetry.

[10] The method according to any one of [1] to [9], characterized in that the model describes the anatomical features in the form of a 2D or 3D description.

[11] In step D), at least one anatomical feature is identified in a first section of the computed image and at least one anatomical feature is identified in a second section; In step E), a deviation between the at least one anatomical feature identified in the first section and the model and a deviation between the at least one anatomical feature identified in the second section and the model are determined by applying a metric; In step F), at least one reconstruction parameter for the first section and at least one reconstruction parameter for the second section of the calculated image are modified independently of each other, In step F-1), the deviation between the at least one identified anatomical feature in the first section and the model and the deviation between the at least one identified anatomical feature in the second section and the model are again determined by applying a metric; In step F-2), steps F) and F-1) are repeated until each determined deviation is less than a specified threshold. The method according to any one of [1] to

[10] ,

[12] The method according to

[11] , characterized in that a method for correcting changes in the position of the oral cavity (6) is used during rotation of the recording unit, and the first section and the second section of the calculated image are positioned directly or indirectly adjacent to each other in the direction of rotation of the recording unit.

[13] The method of

[11] , wherein the first section of the calculated image is an upper section including at least one portion of the maxilla, and the second section of the calculated image is a lower section including at least one portion of the mandible.

[14] The method according to any one of [1] to

[13] , characterized in that in step H) a measure for said modification of said at least one reconstruction parameter, preferably having a specification of the direction, amount and / or type of modification, is provided together with said 2D panoramic image.

[15] An apparatus (1) for generating a 2D panoramic image of an oral cavity (6), comprising: a recording unit having a radiation source (2) and an electronic image sensor (3), the recording unit being constructed and designed so that a plurality of projection images can be acquired while the recording unit rotates at least partially around the oral cavity (6) in a given position; Computer and Equipped with the computer is operatively connected to the electronic image sensor (3) such that the computer can receive the plurality of projected images from the electronic image sensor (3); The computer (4) Step A) receiving the plurality of projection images from the recording unit; Step B) calculating an image including at least one anatomical feature from the plurality of projection images using a calculation rule having reconstruction parameters; Step C) providing a model of the at least one anatomical feature describing the at least one anatomical feature at a target location in the image; Step D) identifying said at least one anatomical feature in said computed image; Step E) determining a deviation between an actual position of the at least one identified anatomical feature in the calculated image and the target position of the at least one anatomical feature in the model by applying a metric; step F) modifying at least one of the reconstruction parameters and recalculating at least one portion of the image showing the at least one anatomical feature using the calculation rule and the at least partially modified reconstruction parameters; or modifying a pixel transformation and applying said pixel transformation to at least a portion of said calculated image showing said at least one anatomical feature, so that at least said portion of said image is recalculated; Step F-1) redetermining the deviation between the actual position of the at least one identified anatomical feature in the calculated image and the target position of the at least one anatomical feature in the model by applying a metric; Step G) deriving the 2D panoramic image from the recomputed images; and Step H) outputting the 2D panoramic image. The apparatus (1) is configured to perform the following.

[0093] [List of references] 1: Equipment 2 : Source 3: Sensor 4: Computer 5 : Head 6: Oral cavity 7: Rotation path 8 : Soft palate 9: Jawbone 10: Teeth 11,12: Temporomandibular joint

Claims

1. 1. A method for generating a 2D panoramic image of an intraoral cavity (6), comprising: Step A) Providing a plurality of projection images acquired while a recording unit comprising a radiation source (2) and an electronic image sensor (3) rotates at least partially around said oral cavity (6) in a given position; Step B) calculating an image including at least one anatomical feature from the plurality of projection images using a calculation rule having reconstruction parameters; Step C) providing a model for the at least one anatomical feature that describes the at least one anatomical feature at a target location in the image; Step D) identifying said at least one anatomical feature in said computed image; Step E) determining a deviation between an actual position of the at least one identified anatomical feature in the calculated image and the target position of the at least one anatomical feature in the model; step F) modifying at least one of the reconstruction parameters based on the determined deviation and recalculating at least one portion of the calculated image showing the at least one anatomical feature using the calculation rule and the at least partially modified reconstruction parameters, or modifying a predetermined initial pixel transformation based on the determined deviation and applying the modified pixel transformation to at least one portion of the calculated image showing the at least one anatomical feature, so that at least this portion of the calculated image is recalculated; Step G) deriving the 2D panoramic image from the recomputed images; and Step H) Outputting the 2D panoramic image. and wherein between steps F) and G), the method comprises: Step F-1) re-determining the deviation between the actual position of the at least one identified anatomical feature in the re-calculated image and the target position of the at least one anatomical feature in the model; Step F-2) repeating steps F) and F-1) until the redetermined deviation is less than a specified threshold; and Step F-3) if in step F) at least one portion of the calculated image has been recalculated but the entire image has not been recalculated, recalculating at least one portion of the entire image that has not been recalculated in step F) using the at least partially changed reconstruction parameters or applying the at least one pixel transformation to at least one portion of the entire image that has not been recalculated in step F); A method comprising:

2. 2. The method of claim 1, wherein providing the plurality of projection images in step A) comprises acquiring the plurality of projection images while the recording unit rotates at least partially around the oral cavity (6).

3. 3. The method according to claim 1, wherein the calculation rule is designed such that changing the at least one reconstruction parameter in step F) does not substantially change at least one further imaging characteristic of the 2D panoramic image displayed in step H), different from the position of the intraoral cavity.

4. 4. The method according to claim 3, characterized in that when changing the at least one reconstruction parameter, the influence of the at least one reconstruction parameter on the at least one further imaging characteristic is substantially compensated in step F).

5. 5. The method according to claim 3 or 4, characterized in that the at least one further imaging characteristic of the 2D panoramic image displayed in step H) is selected from the position of a layer that is clearly imaged in the 2D panoramic image and a radiation transmission angle.

6. 3. The method according to claim 1 or 2, characterized in that when a plurality of reconstruction parameters are changed in step F), in addition to the position of the intraoral cavity, further imaging characteristics different from the position of the intraoral cavity change, and the calculation rule is designed in such a way that the position of the intraoral cavity and the further imaging characteristics are changed independently of each other.

7. 7. The method according to any one of claims 1 to 6, characterized in that the at least one anatomical feature is selected from the position and / or shape of the teeth, the temporomandibular joint, the palate, at least one element of the spine, the clavicle, and / or bones.

8. The method of claim 1 , wherein the at least one anatomical feature comprises anatomical symmetry.

9. 9. The method according to any one of claims 1 to 8, characterized in that the model describes the anatomical features in the form of a 2D or 3D description.

10. In step D), at least one anatomical feature is identified in a first section of the calculated image and at least one anatomical feature is identified in a second section of the calculated image; In step E), a deviation between the at least one anatomical feature identified in the first section and the model and a deviation between the at least one anatomical feature identified in the second section and the model are determined; In step F), at least one reconstruction parameter for the first section and at least one reconstruction parameter for the second section of the calculated image are modified independently of each other, In step F-1), a deviation between the at least one identified anatomical feature in the first section and the model and a deviation between the at least one identified anatomical feature in the second section and the model are redetermined; In step F-2), steps F) and F-1) are repeated until each redetermined deviation is less than a specified threshold.

10. The method according to any one of claims 1 to 9, characterized in that

11. 11. The method according to claim 10, characterized in that during rotation of the recording unit, the first section and the second section of the calculated image are arranged directly or indirectly adjacent to each other in the direction of rotation of the recording unit in order to compensate for changes in the position of the oral cavity (6).

12. 11. The method of claim 10, wherein the first section of the calculated image is an upper section including at least a portion of the upper jaw, and the second section of the calculated image is a lower section including at least a portion of the lower jaw.

13. 13. The method according to claim 1, wherein in step H) a measure for the modification of the at least one reconstruction parameter is provided together with the 2D panoramic image, the measure comprising a specification of the direction, amount and / or type of modification.

14. An apparatus (1) for generating a 2D panoramic image of an oral cavity (6), comprising: a recording unit having a radiation source (2) and an electronic image sensor (3), the recording unit being constructed and designed so that a plurality of projection images can be acquired while the recording unit rotates at least partially around the oral cavity (6) in a given position; Computer and Equipped with the computer is operatively connected to the electronic image sensor (3) such that the computer can receive the plurality of projected images from the electronic image sensor (3); The device (1), wherein the computer (4) is configured to perform the steps of any one of claims 1 to 13.

Images