Solution for CT dental x-ray imaging of object

Abstract

To provide a dental X-ray imaging system for a computed tomography (CT) dental X-ray imaging of an object.SOLUTION: The system comprises: a dental X-ray imaging unit comprising an X-ray source part, an X-ray imaging detector part and a gantry part; and a control system. The control system is configured to: receive a scan request comprising an indication of a region of interest (ROI); control the dental X-ray imaging unit to acquire two X-ray scout images of the object; detect a target structure of the object from the two X-ray scout images associated with the ROI by applying a trained detection model for the X-ray scout images; define field-of-view data of the dental X-ray imaging unit at least partially based on the detected target structure; and control the parts of the dental X-ray imaging unit to implement the defined FOV data for a CT scan of the object in order to acquire dental CT X-ray image data of the object.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates generally to the technical field of dental x-ray imaging. In particular, the present invention relates to computed tomography (CT) dental x-ray imaging. [Background technology]

[0002] A dental x-ray imaging unit typically includes a gantry portion, i.e., a rotating portion, having an x-ray source portion and an x-ray imaging detector portion. The x-ray source portion provides an x-ray beam to the x-ray imaging detector portion, which passes through an object, e.g., a patient's head. In computed tomography (CT) dental x-ray imaging, the x-ray beam forms a cylindrical field of view (FOV) around a vertical axis of rotation when the gantry portion is arranged to rotate around the object. The position and dimensions of the cylindrical FOV should be defined such that the anatomical structures of the object to be included during CT scanning are within the cylindrical FOV.

[0003] Patient support accessories are typically used to support the patient's head as still as possible. The use of patient support accessories targeted to certain areas (i.e., certain anatomical structures of the patient) of the object (e.g., the anterior teeth and / or the temporomandibular joint (TMJ)) allows the certain areas to be in a known position relative to the dental X-ray imaging unit, so that the 3D position of the FOV can be known in advance. However, this only allows the certain areas to be reliably imaged, while patient support accessories for other areas do not exist. Furthermore, the assembly of patient support accessories for these certain areas is time-consuming.

[0004] One conventional FOV positioning and dimensioning approach is based on using X-ray scout images, i.e., a set of two X-ray projection images typically taken at a 90 degree angle, and a user of the dental X-ray imaging unit, e.g., an operator, manually positions and dimensions the FOV through the X-ray scout images. Manual FOV positioning and dimensioning by the operator is time consuming and requires anatomical knowledge and skill by the operator.

[0005] Another known FOV positioning and sizing technique is based on using an optical scout image, where a user of a dental x-ray imaging unit manually positions and sizes the FOV by the optical scout image. However, the optical scout image-based FOV positioning and sizing is inaccurate because the patient's dental anatomical structures are not depicted in the optical scout image. This requires that in addition to the optical scout image-based FOV positioning and sizing, an x-ray scout image-based FOV positioning and sizing is also required, which increases the imaging time. Furthermore, the optical scout image-based FOV positioning and sizing requires that the x-ray dental imaging unit is equipped with one or more optical cameras, which increases the price of the x-ray dental imaging unit. Summary of the Invention [Means for solving the problem]

[0006] The following presents a simplified summary to provide a basic understanding of some aspects of various invention embodiments. This summary is not an extensive overview of the invention. It is not intended to identify key or critical elements of the invention or to delineate the scope of the invention. The following summary merely presents some concepts of the invention in a simplified form as a prelude to a more detailed description of example embodiments of the invention.

[0007] It is an object of the present invention to provide a dental x-ray imaging system and a method for computed tomography (CT) dental x-ray imaging of an object. Another object of the present invention is to provide a dental x-ray imaging system and a CT dental x-ray imaging method for an object that enables a reduction in imaging time and an improvement in imaging accuracy.

[0008] The object of the invention is achieved by a dental X-ray imaging system, a method, a computer program and a tangible non-volatile computer readable medium as defined by the respective independent claims.

[0009] According to a first aspect, there is provided a dental x-ray imaging system for computed tomography (CT) dental x-ray imaging of an object, the system comprising: a dental x-ray imaging system comprising an x-ray source portion for emitting x-rays, an x-ray imaging detector portion for receiving x-rays from the source portion, and a gantry portion comprising the x-ray source portion and the x-ray imaging detector portion; and a control system configured to receive a scan request including an indication of at least one region of interest (ROI), control portions of the dental x-ray imaging unit to acquire two x-ray scout images of the object, and to perform a scan on the dental x-ray imaging system. and a control system configured to: detect at least one target structure of the object from two X-ray scout images associated with the at least one ROI by applying at least one trained detection model to the X-ray scout images; define field of view (FOV) data of a dental X-ray imaging unit based at least in part on the detected at least one target structure; and control portions of the dental X-ray imaging unit to implement the defined FOV data for a CT scan of the object to obtain dental CT X-ray image data of the object.

[0010] The FOV data may include an FOV position and an FOV size.

[0011] The two X-ray scout images of the object may include a lateral (LAT) viewing angle X-ray scout image and a anteroposterior (PA) viewing angle X-ray scout image.

[0012] The control system may be further configured to obtain imaging geometry data representative of an imaging geometry of the dental X-ray imaging unit for each of the two X-ray scout images, and to convert the detected at least one target structure into 3D device coordinates by using the obtained imaging geometry data.

[0013] The scan request may further include object data indicative of at least one object-related parameter and / or imaging program data indicative of a required imaging program.

[0014] The object data and / or the imaging program data may be used in defining the FOV data.

[0015] The at least one trained detection model may be a machine learning (ML) model or an artificial intelligence (AI) model.

[0016] The at least one trained detection model may include at least one common trained detection model that may be trained on both X-ray scout images.

[0017] Alternatively, the at least one trained detection model may include a different at least one trained detection model that may be trained for each X-ray scout image.

[0018] Alternatively or additionally, at least the trained detection model may depend on the region of the object imaged in the X-ray scout image.

[0019] Detecting at least one target structure of the object from the two X-ray scout images may be based on landmark-based detection, bounding box-based detection, and / or object segmentation-based detection.

[0020] In landmark-based detection, at least one landmark associated with at least one ROI may be detected from two X-ray scout images by applying at least one trained detection model to the X-ray scout image.

[0021] According to a second aspect, there is provided a method for computed tomography (CT) dental X-ray imaging, the method being implemented by the X-ray dental imaging system described above, the method including: receiving a scan request including an indication of at least one region of interest (ROI); controlling a portion of the dental X-ray imaging unit to obtain two X-ray scout images of the object; detecting at least one target structure of the object from the two scout images associated with the at least one ROI by applying at least one trained detection model to the X-ray scout images; defining field of view (FOV) data of the dental X-ray imaging unit based at least in part on the detected at least one target structure; and controlling a portion of the dental X-ray imaging unit to implement the defined FOV data for the CT scan of the object to obtain CT dental X-ray image data data of the object.

[0022] According to a third aspect there is provided a computer program comprising instructions which, when the program is executed by an X-ray dental imaging system as described above, cause the X-ray dental imaging system to perform the above method.

[0023] According to a fourth aspect, a tangible, non-volatile computer readable medium is provided, the computer readable medium including instructions which, when executed by the above-mentioned X-ray dental imaging system, cause the X-ray dental imaging system to perform the above-mentioned method.

[0024] Various exemplary and non-limiting embodiments of the present invention, both as to structure and method of operation, together with additional objects and advantages thereof, will be best understood from the following description of specific exemplary and non-limiting embodiments when read in conjunction with the accompanying drawings.

[0025] In this specification, the verbs "comprise" and "include" are used as open qualifiers that do not exclude or require the presence of unrecited features. The features recited according to the claims are freely combinable with each other unless expressly stated otherwise. Furthermore, throughout this specification, the use of "a" or "an", i.e., the singular, is to be understood as not excluding the plural.

[0026] Embodiments of the present invention are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings. [Brief description of the drawings]

[0027] [Figure 1] 1 illustrates generally an embodiment of a dental x-ray imaging system for dental x-ray imaging of an object. [Diagram 2] 1 illustrates generally one embodiment of a method for CBCT dental imaging of an object. [Figure 3A] 1 illustrates generally one embodiment of lateral axial (LAT) imaging of an object. [Figure 3B] 1 illustrates generally one embodiment for proximal-anterior (PA) imaging of an object. [Figure 3C] 1 illustrates generally one embodiment of a LAT scout image of an object. [Figure 3D] 1 illustrates generally one embodiment of a PA scout image of an object. [Figure 4A] 1 illustrates a schematic of one non-limiting example of landmarks intended for detecting maxillary teeth from a LAT scout image of an object. [Figure 4B] 1 illustrates a schematic diagram of one non-limiting example of landmarks intended for detecting maxillary teeth from a PA scout image of an object. [Figure 5A] 1 illustrates generally one embodiment of adjusting the horizontal dimension of the field of view (FOV). [Figure 5B] 13 illustrates generally one embodiment of adjusting the vertical dimension of the FOV. [Figure 6] 1 illustrates generally one embodiment of a control system for a dental x-ray imaging system. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0028] In this document, the following vocabulary is used for the different stages of the dental X-ray imaging process: The term radiation simply refers to the stage including irradiation, i.e. the stage where the X-ray source is providing an X-ray beam that passes through the object to the X-ray imaging detector. The object can be expected to remain as stationary as possible, i.e. immobile, during radiation. During radiation, one or more parts of the dental X-ray imaging unit may move. The term scanning then refers to the stage including radiation and movement of one or more parts of the dental X-ray imaging unit. Scanning does not include positioning one or more parts of the X-ray imaging unit in the correct location to provide an X-ray image. The term imaging refers to the whole process including radiation, scanning, and positioning.

[0029] FIG. 1 illustrates an embodiment of a dental x-ray imaging system 100 for dental x-ray imaging of an object 300, e.g., a patient (for clarity, the object 300 is not shown in FIG. 1 ). The imaging system 100 comprises a dental x-ray imaging unit 102 for acquiring x-ray image data from the object 300 or a calibration target in dental x-ray imaging, e.g., in extraoral dental x-ray imaging. The acquired x-ray image data is used to reconstruct a three-dimensional (3D) x-ray volume from at least a portion of the imaged object 300. The dental x-ray imaging system 100 further comprises a control system 106. The control system 106 may be electrically and / or communicatively coupled to the dental x-ray imaging unit 102. The control system 106 may be implemented as a standalone unit or as a distributed control environment among multiple standalone units providing distributed control resources. Preferably, the control system 106 may be an embedded computer. The control system 106 may comprise, for example, a control unit of the dental X-ray imaging unit 102. The control system 106 may further comprise one or more computing units external to the dental X-ray imaging unit 102. The control unit of the dental X-ray imaging unit 102 is configured to at least partially control the operation of the dental X-ray imaging unit 102. The control unit of the dental X-ray imaging unit 102 may be located in proximity to the dental X-ray imaging unit 102 or the control unit of the dental X-ray imaging unit 102 may be embedded within the dental X-ray imaging unit 102.

[0030] The dental x-ray imaging unit 102 is configured to perform a computed tomography (CT) imaging procedure (i.e., imaging mode). The CT imaging can be, for example, cone-beam CT (CBCT) imaging, where the beam is a cone-shaped beam, or other types of CT imaging, where, for example, the beam is a pyramidal beam, a half-moon shaped cone beam, or any other shaped beam. The CT imaging results (i.e., produces) x-ray image data for reconstruction of a 3D volume from at least a portion of the imaged object 300. FIG. 1 illustrates only one example of a dental x-ray imaging unit 102 for use with the concepts of the present disclosure.

[0031] The dental x-ray imaging unit 102 comprises a carriage portion 101, which may be movably supported on a support column 103. The carriage portion 101 may be moved up and down in a vertical direction (V) by a guide motor (not shown in FIG. 1 ) configured to move the carriage 101 up and down vertically along the support column 103. The upper shelf 110 is configured to support a gantry portion, i.e., a rotating portion 112, which is rotatable in a horizontal (H) plane relative to the upper shelf 110. The upper shelf 110 and / or the gantry portion 112 may comprise a rotating motor (not shown in FIG. 1 ) configured to rotate the gantry portion 112. Alternatively or additionally, the upper shelf 110 and / or the gantry portion 112 may comprise a pivot motor (not shown in FIG. 1 ) configured to pivot the upper shelf 110 about the column 103. The upper shelf 110 and / or the gantry portion 112 may alternatively or additionally comprise at least one linear motor (not shown in FIG. 1 ) configured to provide linear motion(s). Alternatively or additionally, the dental x-ray imaging unit 102 may be mounted to a support structure (not shown in FIG. 1 ), illustratively a wall supported by a column 103.

[0032] The dental X-ray imaging unit 102 further comprises an X-ray source portion 114 and an X-ray imaging detector portion 116 used in acquiring X-ray image data. The gantry portion 112 embodies and supports the source portion 114 and the imaging detector portion 116. The gantry portion 112 may have substantially the form of the letter C as presented in FIG. 1, where the source portion 114 may be mounted at one end of the gantry portion 112 and the imaging detector portion 116 may be mounted at the other end of the gantry portion 112 such that the source portion 114 and the imaging detector portion 116 face each other. The X-ray source portion 114 comprises an X-ray source that emits X-rays (i.e., generates an X-ray beam) through an object 300 to be imaged, e.g., a patient's head, to the X-ray imaging detector portion 116, and the X-ray imaging detector portion 116 comprises at least one X-ray detector that receives the X-rays emitted from the source portion 114. The X-ray imaging detector portion 116 further generates X-ray image data from the object exposed to the X-rays, ie, imaged.

[0033] The dental x-ray imaging unit 102 also includes a collimator 506 (not shown in FIG. 1 ) for the x-ray source portion 114 to limit and / or shape the x-ray beam. The x-rays pass through a portion of the object 300, e.g., the patient's anatomy, e.g., the patient's head. The anatomical structure through which the x-rays pass may absorb different amounts of x-ray energy. After passing through the object 300, the attenuated x-rays are received by the x-ray imaging detector portion 116. The x-ray imaging detector portion 116 is configured to convert the magnitude of the received x-ray energy and to produce a digitized output, i.e., x-ray image data, representative of the x-rays not absorbed in at least one x-ray detector. The collection of digitized outputs from the x-ray imaging detector portion 116, which correspond to a single emission of the beam of x-rays from the x-ray source portion 114, may be referred to as a projection image of the object 300 being imaged, e.g., the head of the patient 300.

[0034] Further, the dental x-ray imaging unit 102 may include patient support portions 124, 126 (as presented in FIG. 1, but not necessarily so) that may be used to support a patient. The patient support portions 124, 126 may include a chin support portion 124 and / or a head support portion 126. The chin support portion 124 may support the tip of the patient's chin, and the head support portion 126 may support the patient's forehead or temples. The dental x-ray imaging unit 102 may include a lower shelf 122 extending from the carriage 101. The lower shelf 122 may include the chin support portion 124 as in the exemplary dental x-ray imaging unit 102 of FIG. 1. The head support portion 126 may extend from the upper shelf 110 through the gantry portion 112 as in the exemplary dental x-ray imaging unit 102 of FIG. 1. Alternatively, the lower shelf 122 may also include the head support portion 126. The patient support portions, i.e., the chin support portion 124 and / or the head support portion 126, may be optional and patient positioning may be performed in other ways. The dental x-ray imaging unit 102 may further comprise a handle 128 for the patient to grasp.

[0035] The gantry portion 112 may be rotated, for example, by a rotation motor. The rotation of the gantry portion 112 causes the X-ray source portion 114 and the X-ray imaging detector portion 116 to rotate around the object to be imaged, for example, around a rotation axis along a motion path. The X-ray beam emitted by the X-ray source portion 114 forms a 3D cylindrical field of view (FOV) around the rotation axis as the gantry portion 112 rotates around the patient. The 3D cylindrical FOV should be dimensioned and positioned such that the anatomical structure(s) of the object 300 to be included in the scan are inside the 3D cylindrical FOV. As the X-ray source portion 114 and the X-ray imaging detector portion 116 rotate around the object 300, the dental X-ray imaging unit 102 operates to acquire multiple projection images of the object 300 captured at incrementally increasing rotation angles. The dental X-ray image may be formed from the multiple projection images by reconstructing the X-ray image data into a dental X-ray image.

[0036] At least some exemplary embodiments of a method for CT dental x-ray image imaging of an object 300, e.g., a patient, are now defined with reference to Fig. 2. Fig. 2 illustrates the method as a flow chart. The method is performed by the dental x-ray system 100 discussed above. The CT dental imaging of the object 300 may preferably be CBCT dental imaging.

[0037] At step 210, the control system 106 receives a scan request. The scan request includes an indication of at least one region of interest (ROI). The at least one ROI represents at least one region of the object 300 (e.g., at least one target anatomical structure) to be included during the CT scan. For example, the at least one ROI may include, but is not limited to, at least one of a single tooth, a range of teeth, a dental arch (either upper or lower), both dental arches (upper and lower), a temporomandibular joint (TMJ), both TMJs, the entire dentition, and a TMJ. The scan request may further include object data representing at least one object-related parameter (e.g., age, size, etc.) and / or imaging program data representing a required imaging program (e.g., dose information including an indication of the dose to be used in the scan, resolution information including an indication of the resolution to be used in the scan, etc.). The scan request may also include patient identification data (e.g., patient name, patient identification number, photograph, fingerprint, retinal scan, facial recognition, and / or other biometric data, etc.). The scan request may be received locally or remotely via the user interface portion 140a, 140b. The user interface portion 140a may be located, for example, at the same location where the dental X-ray imaging unit 102 is located, i.e., the scan request may be received locally. For example, the dental X-ray imaging unit 102 may include a user interface portion 140a, for example, a touch screen, as in the exemplary dental X-ray imaging unit 102 of FIG. 1. An operator of the dental X-ray imaging unit 102 may input the scan request via the user interface portion 140a. Alternatively or additionally, the user interface portion 140b may be located, for example, at a location other than where the dental X-ray imaging unit 102 is located, i.e., the user interface portion 140b may be located remotely from the dental X-ray imaging unit 102 and communicatively coupled to the control system 106. In that case, the scan request may be received remotely. According to an embodiment, the initial FOV data may be defined based on the received scan request. The initial FOV data may include an initial position of the FOV and an initial size of the FOV (ie, initial dimensions of the FOV).

[0038] In step 220, the control system 106 controls parts of the dental X-ray imaging unit 102 to acquire two (non-parallel) X-ray scout images 302, 304 of the object 300, e.g., a first X-ray scout image 302 of the object 300 and a second scout image 304 of the object 300. The two X-ray scout images 302, 304 are two-dimensional (2D) X-ray projection images. Preferably, the two X-ray scout images 302, 304 are acquired orthogonally, i.e., at a 90 degree angle, to each other. Most preferably, the two scout images include a lateral (LAT) viewing angle X-ray scout image 302 of the object 300, i.e., a LAT scout image, and a anteroposterior (PA) viewing angle X-ray scout image 304 of the object 300, i.e., a PA scout image. Figure 3A illustrates generally an example of lateral imaging of an object 300 to obtain a LAT scout image 302 of the object 300. Figure 3B illustrates generally an example of an anterior-posterior imaging of the object 300 to obtain a PA scout image 304 of the object 300. Figure 3C illustrates generally an example of a LAT scout image 302 of the object 300 obtained by lateral imaging. Figure 3D illustrates generally an example of a PA scout image 304 of the object 300 obtained by anterior-posterior imaging.

[0039] In step 230, the control system 106 detects at least one target structure (e.g., at least one anatomical target structure of the object 300) of the object 300 associated with at least one ROI indicated in the received scan request from the two X-ray scout images 302, 304 by using automatic detection. The automatic detection includes applying at least one trained detection model 616 to the X-ray scout images 302, 304. In other words, in step 230, the control system 106 detects at least one target structure of the object 300 associated with at least one ROI indicated in the received scan request from the two X-ray scout images 302, 304 by applying at least one trained detection model 616 to the X-ray scout images 302, 304. In further words, in step 230, the control system 106 detects the same at least one structure of the object 300 associated with at least one ROI indicated in the received scan request from both X-ray scout images 302, 304 by applying at least one trained detection model 616 on the X-ray scout images 302, 304. The two X-ray scout images 302, 304 are used as input data of the at least one trained detection model 616, and at least one target structure of the object 300 associated with at least one ROI is obtained as output data of the at least one trained detection model 616. Detecting at least one target structure of the object 300 means that the at least one target structure is localized, i.e., its spatial location in each X-ray scout image 302, 304 is defined, and its identification information (e.g., label name) is defined.The expression "at least one target structure of the object 300 associated with at least one ROI" means throughout this application that the at least one target structure of the object 300 may be within the area of ​​the at least one ROI (e.g., included in the at least one ROI) and / or that there may be at least one anatomical structure of the object 300 (outside the area of ​​the at least one ROI) that can be used to detect the at least one target structure of the object 300 within the area of ​​the at least one ROI. In other words, the at least one target structure of the object 300 within the area of ​​the ROI may be detected directly and / or indirectly.

[0040] The at least one trained detection model 616 may include at least one common trained detection model trained for both X-ray scout images 302, 304. In that case, the control system 106 may apply the at least one common trained detection model 616 to the first X-ray scout image 302 (e.g., LAT scout image) to detect at least one target structure of the object 300 associated with at least one ROI indicated in the received scan request from the first X-ray scout image 302, and apply the same at least one common trained detection model 616 to the second X-ray scout image 304 (e.g., PA scout image) to detect at least one target structure of the object 300 associated with at least one ROI indicated in the received scan request from the second X-ray scout image 304.

[0041] Alternatively, the at least one trained detection model may include at least one different detection model trained for each X-ray scout image 302, 304. As discussed above, the two X-ray scout images 302, 304 are non-parallel to each other. This allows the at least one different trained detection model 616 to be trained for different viewing angles. In other words, the at least one different detection model 616 may be trained for a first X-ray scout image (e.g., a LAT scout image) 302, and the at least one different detection model 616 may be trained for a second X-ray scout image (e.g., a PA scout image) 304. When the at least one trained detection model includes at least one different detection model trained for each X-ray scout image 302, 304, the at least one detection model 616 trained for the first scout image 302 may be referred to, for example, as at least one trained first detection model, and the at least one detection model trained for the second scout image 302 may be referred to, for example, as at least one trained second detection model 616. In this case, the control system 106 may apply at least one trained first detection model 616 to the first X-ray scout image 302 (e.g., a LAT scout image) to detect at least one target structure of the object 300 associated with at least one ROI indicated in the received scan request from the first X-ray scout image 302, and apply at least one trained second detection model 616 to the second X-ray scout image 304 (e.g., a PA scout image) to detect at least one target structure of the object 300 associated with at least one ROI indicated in the received scan request from the second X-ray scout image 304.

[0042] At least one trained detection model 616 may depend on the region of the object 300 imaged in the X-ray scout images 302, 304. In other words, different trained detection models 616 may be trained for different regions of the object 300. For example, at least one trained detection model 616 may be trained for at least one of the following regions of the object 300: maxillary teeth region, mandibular teeth region, both maxillary teeth and mandibular teeth region, and / or TMJ region, etc. At least one common trained detection model 616 for both scout images (e.g., LAT scout image and PA scout image) 302, 304 may be trained for different regions of the object 300. Similarly, at least one trained first detection model 616 for a first X-ray scout image 302 (e.g., a LAT scout image) may be trained for a different region of the object 300, and at least one trained second detection model 616 for a second X-ray scout image 304 (e.g., a PA scout image) may be trained for a different region of the object 300. Depending on the region 300 of the object 300 imaged in the X-ray scout images 302, 304, an appropriate at least one trained detection model 616 for the X-ray scout images 302, 304 may be selected.

[0043] At least one trained detection model 616 may be formed by using training data. Training the at least one detection model 616 may preferably be supervised training, i.e., supervised learning, although other training paradigms may be used, such as unsupervised learning, reinforcement learning, or a hybrid paradigm including two or more training paradigms. In supervised training, the training data includes input training data and output training data. The input training data may include, for example, a plurality of X-ray scout images. The plurality of X-ray scout images included in the input training data may include, for example, pre-collected X-ray scout images. The pre-collected X-ray scout images may preferably include X-ray scout images of a patient. Alternatively or additionally, the pre-collected X-ray scout images may include X-ray scout images of a phantom (e.g., a dried skull). Alternatively or additionally, the plurality of X-ray scout images included in the input training data may include simulated and / or artificial X-ray scout images. Data augmentation may also be used to generate further input training data. The output training data may include, for example, annotations of anatomical structures, such as annotation data representing the location and identity of anatomical structures. The output training data may be generated in a manual process, such as by a human expert, in a semi-automated process, or in a fully automated process. The at least one trained detection model 616 may be based on one or more machine learning (ML) methods or one or more artificial intelligence (AI) methods. In other words, the at least one trained detection model 616 may be an ML model or an AI model. For example, the at least one trained detection model 616 may be based on a tree-based machine learning (ML) method, such as a decision tree, in particular a regression tree. To leverage the power of a single tree, multiple trees may be combined (i.e., ensemble). Preferably, the at least one trained detection model 616 may be based on an ensemble ML method of regression trees. Alternatively, the at least one trained detection model 616 may be based on a convolutional neural network (CNN). At least one trained detection model 616 may be stored in the memory portion 608 of the control system 106.Multiple trained detection models 616 may be used as discussed above, with each trained detection model 616 based on the same ML or AI method, for example, an ensemble regression tree ML method or a CNN.

[0044] Detecting at least one target structure of the object 300 from the two X-ray scout images 302, 304 may be based on landmark-based detection, bounding box-based detection, and / or object segmentation-based detection. A combination of landmark-based detection and bounding box-based detection may include bounding box detection of a region in each X-ray scout image 302, 304, followed by landmark-based detection applied within the detected region. In landmark-based detection, at least one landmark associated with at least one ROI is detected from each X-ray scout image 302, 304 by using automatic detection, i.e., by applying at least one trained detection model 616 to the X-ray scout images 302, 304. In other words, each X-ray scout image 302, 304 may be used as input data of at least one trained detection model 616, and at least one landmark 402 may be obtained as output data of at least one trained detection model 616. The expression "at least one landmark associated with at least one ROI" means throughout this application that the at least one landmark may be within the region of the at least one ROI (e.g., included in the at least one ROI) and / or may be at least one landmark that can be used to detect the at least one landmark within the region of the at least one ROI (outside the region of the at least one ROI). In other words, the at least one landmark within the region of the ROI may be detected directly and / or indirectly. Some non-limiting examples of the at least one landmark 402 may include mandibular teeth, maxillary teeth, TMJ, anterior nasal bone (ANS), posterior nasal bone (PNS), and / or any other cranial structure. Some of the landmarks 402 may be directly related to the ROI and some of the landmarks 402 may be stable cranial structures that may be used as proxy information. For example, the ANS is located just above the maxillary tooth root, so that if the uppermost position of the maxillary tooth root needs to be estimated, the ANS may be used as proxy information for it.FIG. 4A illustrates a non-limiting example of a landmark 402 intended to detect maxillary teeth from a LAT scout image 302. FIG. 4B illustrates a non-limiting example of a landmark 402 intended to detect maxillary teeth from a PA scout image 304. In the example of FIG. 4A and FIG. 4B, a small circle in a dashed oval represents the landmark 402. In the example of FIG. 4A, the trained detection model 616 used is trained on the LAT scout image 302 in the region of maxillary teeth. In the example of FIG. 4B, the trained detection model 616 used is trained on the PA scout image 304 in the region of maxillary teeth.

[0045] In step 240, the control system 106 defines FOV data of the dental x-ray imaging unit 102 based at least in part on the at least one target structure detected in step 230. The FOV data may include an FOV location and an FOV size (i.e., a dimension of the FOV). Preferably, in step 240, the control system 106 defines the FOV data based on the at least one target structure detected in step 230. Alternatively, if the initial FOV data is defined based on the scan request as discussed above, the control system 106 may define the FOV data based in part on the at least one target structure detected in step 230 and in part on the initial FOV data. For example, the FOV size (or at least one dimension of the FOV) may be defined based on the initial FOV data, or the FOV location may be defined based on the initial FOV data, for example. According to one non-limiting example, the FOV position may be defined based on at least one target structure of the object detected in step 230, and the FOV size may be defined based on the initial FOV data, or vice versa. The FOV position may be, for example, a central vertical axis of the FOV (i.e., the FOV central axis). If the at least one target structure includes a single structure (e.g., a single tooth), the control system 106 may define the FOV central axis such that the FOV central axis passes through (i.e., passes through) the center of the single structure. Alternatively, if the at least one target structure includes multiple structures (e.g., multiple teeth), the control system 106 may define the FOV central axis such that the FOV central axis passes through the centers of the multiple structures. The centers of the multiple structures may be defined, for example, by using averaging (e.g., arithmetic mean, geometric mean, weighted mean, etc.). The dimensions of the FOV may include, for example, a horizontal dimension of the FOV (e.g., a width of the FOV, i.e., a diameter of the FOV, or a radius of the FOV) and a vertical dimension of the FOV (e.g., a vertical height of the FOV). If the at least one target structure includes a single structure, the control system 106 may estimate the horizontal dimension of the FOV based on, for example, the horizontal dimension of the structure.Alternatively, if the at least one target structure includes multiple structures, the control system 106 may estimate the horizontal dimension of the FOV, for example, by defining for each structure a distance from the defined FOV central axis, and the maximum of these distances may be used as the radius of the FOV. The width of the FOV, i.e., the diameter of the FOV, may be defined based on the radius of the FOV. The vertical dimension of the FOV may be defined based on the topmost point of the FOV and the bottommost point of the FOV. To define the topmost point of the FOV, the control system 106 may define the topmost point of the at least one target structure and assign the top of the FOV at the same level, i.e., the defined topmost point of the at least one target structure is assigned as the topmost point of the FOV. If the target, at least one target structure, includes multiple structures, the topmost point of all of these multiple structures may be assigned as the topmost point of the FOV. To define the lowest point of the FOV, the control system 106 may define the lowest point of the at least one target structure and assign the bottom of the FOV at the same level, i.e., the defined lowest point of the at least one target structure is assigned as the lowest point of the FOV. If the at least one target structure includes multiple structures, the lowest point among all of these multiple structures may be assigned as the lowest point of the FOV.

[0046] According to an embodiment, a safety margin may be added to the horizontal dimension of the FOV and to the vertical dimension of the FOV. The size of the safety margin may, for example, be fixed or may depend on the object 300, the ROI, and / or the imaging program data. The safety margin may be the same for the horizontal dimension of the FOV and for the vertical dimension of the FOV. Alternatively, different safety margins may be added to the horizontal dimension and the vertical dimension of the FOV. According to an embodiment, the object data and / or the imaging program data may be used in the definition of the FOV data, for example in the definition of the safety margin.

[0047] The FOV data may be defined in step 240 in 3D imaging device coordinates (i.e., coordinates of the dental X-ray imaging unit 102). When the at least one target structure of the object 300 detected in step 230 is presented in 2D image coordinates, the detected at least one target structure may be transformed into 3D imaging device coordinates. The transformation may be performed by using imaging geometry data representing the imaging geometry of the dental X-ray imaging unit 102 for each of the two X-ray scout images 302, 304. The control system 106 may obtain the imaging geometry data, for example, when the two X-ray scout images are acquired in step 220, i.e., in connection with acquiring the two X-ray scout images. The control system 106 may transform the detected at least one target structure into 3D imaging device coordinates by using the obtained imaging geometry data. Using the imaging geometry data, the position of the at least one target structure detected from the X-ray scout images 302, 304 may be transformed into 3D imaging device coordinates. For example, when the same target structure(s) are detected in two X-ray scout images (e.g., in the LAT scout image 302 and the PA scout image 304), the target structure(s) can be mapped to 3D imaging device coordinates by using the obtained imaging geometry data. The transformation can be performed, for example, by applying the intersection method. Next, an example of applying the intersection method to one target structure to transform the target structure to 3D imaging device coordinates is described. However, the intersection method can be applied to each target structure in the same way to transform the target structure in question to 3D imaging device coordinates. The position of the X-ray source in the imaging device coordinates can be defined from the imaging geometry data. Similarly, the position of the X-ray detector in the imaging device coordinates can be defined from the imaging geometry data. By combining the position of the X-ray detector and the detected target structure in the X-ray detector (i.e., in the X-ray scout images 302, 304), the position of the detected target structure in the X-ray scout images 302, 304 can be expressed in imaging device coordinates.Thereby, a line origin from the X-ray source to the position of the detected target structure on the X-ray detector can be formed in imaging device coordinates. Since there are two non-parallel X-ray scout images 302, 304, two non-parallel lines can be formed for the detected target structure. The position of the detected target structure in imaging device coordinates can then be solved by applying the intersection method. In addition to the intersection method, there are many similar alternative methods, such as the intersection of a 2D plane and a line, the intersection of two lines projected onto a plane (e.g., a horizontal plane), etc. Each of these alternative methods or any other known method can be used to transform at least one target structure into 3D imaging device coordinates when imaging geometry data is available.

[0048] After defining the FOV data, in optional step 250, the control system 106 may output the defined FOV data overlaid on the X-ray scout images 302, 304 to an operator of the dental X-ray imaging unit 102 via one or more user interface devices, for example, the user interface portion 140a, for review and approval. This allows the operator to verify the accuracy of the defined FOV data. If the operator is satisfied with the defined FOV data, the operator may approve the defined FOV data. Alternatively, if the operator is not satisfied with the defined FOV data, the operator may manually adjust the FOV position and / or at least one dimension of the FOV via the user interface portion 140a before approving the FOV data.

[0049] In step 260, the control system 106 controls the parts of the dental X-ray imaging unit 102 to implement the defined FOV data for the CT scan of the object 300 to obtain dental CT-ray image data of the object 300. In the case of symmetric CT imaging, the control system 106 controls the gantry part 112 of the dental X-ray imaging unit 102 to move to a start position of the CT scan, and align the rotation axis with the central axis of the FOV defined in step 240. The rotation axis can be the mechanical rotation axis of the gantry part 112 or a virtual rotation axis. The virtual rotation axis can be obtained, for example, by moving the mechanical rotation axis of the gantry part 112 along a circular path, and then the virtual rotation axis can be formed at the center of the circular path. In the case of offset CT imaging, the start position of the CT scan can be offset by a known amount in a known direction from the central axis of the FOV defined in step 240, i.e., by a known offset vector. The control system 106 may adjust the horizontal dimension 504 of the FOV 502 by the collimator 506 of the X-ray source portion 114 to correspond to the defined horizontal dimension of the FOV. The vertical dimension 510 of the FOV 502 may be adjusted by the collimator 506 to correspond to the defined vertical dimension of the FOV. For example, the vertical dimension 510 of the FOV 502 may be adjusted by controlling the collimator 506 of the X-ray source portion 114 so that the top point of the FOV 502 corresponds to the defined top point of the FOV and / or the bottom point of the FOV 502 corresponds to the defined bottom point of the FOV. The collimator 506 of the X-ray source portion 114 may include an adjustable slit. The size of the slit of the collimator 506 may be adjusted horizontally to adjust the horizontal dimension 504 of the FOV 502 and vertically to adjust the vertical dimension 510 of the FOV 502. 5A illustrates generally one embodiment of adjusting a horizontal dimension 504 of an FOV 502. In this embodiment, the horizontal dimension 504 is the width of the FOV 502, i.e., the diameter of the FOV 502. The diameter 504 of the FOV 502 can be adjusted by adjusting the size of the slits of the collimator 506 horizontally.The size of the slits of the collimator 506 in the horizontal direction, i.e., the width of the slits, is illustrated in FIG. 5A by dashed arrows 508. The larger the slits of the collimator 506 in the horizontal direction, the larger the diameter 504 of the FOV 502. In other words, by closing the slits of the collimator 506 in the horizontal direction, the diameter 504 of the FOV 502 can be adjusted to be smaller, and by opening the slits of the collimator 506 in the horizontal direction, the diameter 504 of the FOV 502 can be adjusted to be larger. FIG. 5B illustrates generally an example of adjusting the vertical dimension 510 of the FOV 502. The vertical dimension 510 in this example is the height of the FOV 502. The height 510 of the FOV 502 can be adjusted by adjusting the size of the slits of the collimator 506 in the vertical direction. The size of the slits of the collimator 506 in the vertical direction, i.e., the height of the slits, is illustrated in FIG. 5B by dashed arrows 512. The larger the slits of the collimator 506 in the vertical direction, the larger the height 510 of the FOV 502. In other words, by closing the slits of the collimator 506 in the vertical direction, the height 510 of the FOV 502 can be adjusted to be smaller, and by opening the slits of the collimator 506 in the vertical direction, the height 510 of the FOV 502 can be adjusted to be larger.

[0050] After the rotation axis is moved to a start position of the CT scan in step 270 and the horizontal and vertical dimensions of the FOV are adjusted in step 260, the control system 106 controls the dental X-ray imaging unit 102 to scan the object (300) to obtain dental CT-ray image data of the object 300. The control in step 270 may include controlling the gantry portion 112 to rotate according to a motion path of the CT scan about the rotation axis to obtain dental CT-ray image data of the object 300. The motion path may be, for example, a circular path or a non-circular path, for example, an elliptical path, about the rotation axis. The circular path or the non-circular path may be a complete rotation or a partial rotation about the rotation axis. The non-circular rotation path may be created, for example, by moving the gantry portion 112 along a motion path that deviates from a circular path, for example, an elliptical path. Other techniques or alignments for the rotation axis may also be used as recognized by those skilled in the art.

[0051] FIG. 6 illustrates one schematic embodiment of the control system 106 of the dental x-ray imaging system 100. The control system 106 may include a processor portion 602, a data transfer portion 604, a user interface portion 606, and a memory portion 608. The processor portion 602 is configured to execute instructions and process data initiated by a user and / or a computer program (software). The processor portion 602 may include at least one processor. The memory portion 608 is configured to store and maintain data. The data may be instructions, computer programs, at least one trained detection model 616, and any data files. The memory portion 608 may include at least one memory. The memory portion 608 may further include at least a data transfer application 610 for controlling the data transfer portion 604, a user interface application 612 for controlling the UI portion 606, and a computer program (code) 614 for controlling the operation of the control system 106. The memory portion 608 and the computer programs 614, together with the processor portion 602, may cause the control system 106 to at least implement one or more method steps and / or operations of the control system 106 as described above. The data transfer portion 604 may be configured to send control commands to other units, such as the dental X-ray imaging unit 102. In addition, the data transfer portion 604 may receive data from other units, such as the dental X-ray imaging unit 102, the user interface portion 140b, database(s), and / or any other external unit. The user interface (UI) portion 606 may be configured to input control commands, receive information and / or instructions, and display information. The UI portion 606 may comprise at least a display, a screen, a touch screen, at least one function key, a keyboard, a wired or wireless remote control, or any other user input and / or output device.The computer program 614 may be a computer program product, which may be contained in a tangible, non-volatile (non-transitory) computer-readable medium having computer program code 614 embodied therein for use in a computer, for example, the control system 106.

[0052] The above CT dental X-ray imaging method enables to shorten the imaging time since it does not require manual FOV positioning by the operator of the dental X-ray imaging unit 102. Furthermore, the positioning of the FOV does not depend on the anatomical knowledge and skill of the operator of the dental X-ray imaging unit 102. Therefore, the above CT dental X-ray imaging method improves the imaging accuracy. Although the positioning of the FOV does not depend on the anatomical knowledge and skill of the operator, the above CT dental X-ray imaging method provides support for the operator's skill.

[0053] The specific examples provided in the above description should not be construed as limiting the applicability and / or interpretation of the appended claims. The lists and groupings of examples provided in the above description are not exhaustive unless expressly stated otherwise.

Claims

1. A dental x-ray imaging system (100) for computed tomography (CT) dental x-ray imaging of an object (300), said system (100) comprising: A dental x-ray imaging unit (102), comprising: an X-ray source portion (114) for emitting X-rays; an x-ray imaging detector portion (116) for receiving the x-rays from the source portion (114); a gantry portion (112) comprising the X-ray source portion (114) and the X-ray imaging detector portion (116); A control system (106), Receiving a scan request including at least one region of interest (ROI) indicator; controlling the portion of the dental x-ray imaging unit (102) to acquire two x-ray scout images (302, 304) of the object (300); detecting at least one target structure of the object (300) from the two X-ray scout images (302, 304) associated with the at least one ROI by applying at least one trained detection model (616) to the X-ray scout images (302, 304); defining field of view (FOV) data of the dental x-ray imaging unit (102) based at least in part on the detected at least one target structure; and a control system (106) configured to control the portion of the dental X-ray imaging unit (102) to implement the defined FOV data for a CT scan of the object (300) to acquire dental CT X-ray image data of the object (300).

2. The dental x-ray imaging system of claim 1 , wherein the FOV data includes an FOV position and an FOV size.

3. 3. The dental X-ray imaging system (100) of claim 1 or claim 2, wherein the two X-ray scout images (302, 304) of the object include a lateral (LAT) viewing angle X-ray scout image (302) and a anteroposterior (PA) viewing angle X-ray scout image (304).

4. The control system (106) obtaining imaging geometry data representative of an imaging geometry of the dental x-ray imaging unit (102) relative to the two x-ray scout images (302, 304); 4. The dental x-ray imaging system (100) of claim 1, further configured to: transform the detected at least one target structure into 3D device coordinates by using the obtained imaging geometry data.

5. The dental X-ray imaging system (100) of any one of claims 1 to 4, wherein the scan request further comprises object data representative of at least one object related parameter and / or imaging program data representative of a required imaging program.

6. The dental x-ray imaging system (100) of claim 5, wherein the object data and / or the imaging program data are used in defining the FOV data.

7. The dental x-ray imaging system (100) of any one of claims 1 to 6, wherein the at least one trained detection model (616) is a Machine Learning (ML) model or an Artificial Intelligence (AI) model.

8. 8. The dental x-ray imaging system (100) of claim 1, wherein the at least one trained detection model (616) comprises at least one common trained detection model trained on both x-ray scout images (302, 304).

9. 8. The dental x-ray imaging system (100) of claim 1, wherein the at least one trained detection model (616) comprises at least one different trained detection model trained for each x-ray scout image (302, 304).

10. 10. The dental x-ray imaging system (100) of claim 1, wherein the at least trained detection model (616) depends on the region of the object (300) imaged in the x-ray scout images (302, 304).

11. The dental X-ray imaging system (100) of any one of claims 1 to 10, wherein detecting the at least one target structure of the object (300) from the two X-ray scout images (302, 304) is based on landmark-based detection, bounding box-based detection, and / or object segmentation-based detection.

12. 12. The dental X-ray imaging system (100) of claim 11, wherein in the landmark-based detection, at least one landmark (402) associated with the at least one ROI is detected from the two X-ray scout images (302, 304) by applying the at least one trained detection model (616) to the X-ray scout images (302, 304).

13. A method for computed tomography (CT) dental x-ray imaging, said method being implemented by a dental x-ray imaging system (100) according to any one of claims 1 to 12, said method comprising: Receiving a scan request (210) including at least one region of interest (ROI) indicator; controlling (220) the portion of the dental x-ray imaging unit (102) to acquire two x-ray scout images (302, 304) of the object (300); Detecting (230) at least one target structure of the object (300) from the two scout images (302, 304) associated with the at least one ROI by applying at least one trained detection model (616) to the X-ray scout images (302, 304); defining (240) field of view (FOV) data of the dental x-ray imaging unit (102) based at least in part on the detected at least one target structure; and controlling (260) the portion of the dental X-ray imaging unit (102) to implement the defined FOV data for a CT scan of the object (300) to acquire CT dental X-ray image data of the object (300).

14. A computer program (614) including instructions, which when executed by an X-ray dental imaging system (100) according to any one of claims 1 to 12, cause the X-ray dental imaging system (100) to perform the method according to claim 13.

15. A tangible, non-volatile computer readable medium containing instructions which, when executed by an X-ray dental imaging system (100) according to any one of claims 1 to 12, cause the X-ray dental imaging system (100) to perform the method of claim 14.

Images