Method and equipment for generating a digital tomosynthetic 3D X-ray image of an examination object

Abstract

Method for generating a digital tomosynthetic 3D X-ray image (T) of an object under investigation (4) using an X-ray tube (2) and a digital X-ray detector (10), in which a plurality of individual images (E) of the object under investigation (4) are produced k ) with different projection angles (α k ) relative to the normal (13) of the patient positioning table (8), and in which a starting position (k=0) is used with a dose that is smaller than that of the individual images (E k ) dose used, at least one model (V0) is recorded, which is used to determine the dose required to record the individual images (E k ) the required recording parameters are evaluated, wherein the dose is used as one recording parameter, characterized in that for each projection angle (α k ) with a weight factor dependent on this (c k ) is weighted and used to adjust the different projection angles (α)k ) the X-ray tube (2) is swung on a circular path around the object under investigation (4) and the weight factor (c) k ) at least for the projection angles (α) different from the starting position (k=0). k ) through the relationship c k = 1 / cosα k is given, where c k the weight factor (c k ) and α k the projection angle (α k ) is.

Description

[0001] The invention relates to a method and a device for generating a digital tomosynthetic 3D X-ray image of an object under investigation.

[0002] When taking an X-ray image of an object, for example in mammography, it is essential to ensure that the exposure parameters, especially the exposure time, are correctly set to guarantee image quality suitable for diagnostic evaluation. To limit the radiation exposure of the irradiated area of ​​the object to the diagnostically necessary minimum, the aim is to set the exposure parameters correctly from the very first exposure. For this purpose, a so-called automatic exposure control (AEC) system is used in the prior art. In analog mammography, a number of solid-state detectors are arranged behind the X-ray film, viewed in the direction of X-ray propagation. These detectors measure the intensity of the X-rays transmitted through the film, and their output signal is used to control the exposure parameters (e.g., exposure time).exposure time, operating voltage of the X-ray tube, tube current, anode-filter combination) is used.

[0003] Continuous improvement of imaging techniques aims to produce highly informative X-ray images with the lowest possible radiation dose. This is crucial, for example, in mammography, to differentiate between benign and malignant lesions and to reduce the number of false positives—that is, the number of suspicious findings caused by benign lesions and the number of undetected malignant tumors. Conventional analog X-ray mammography produces a single two-dimensional image of the compressed breast in a single projection direction. Because tissue layers lying one behind the other in the direction of the X-ray beam are superimposed in such a projection, highly absorbing benign structures can obscure a malignant tumor and make it difficult to detect.

[0004] To avoid this, a mammography technique called tomosynthesis is known, for example, from T. Wu et al., Tomographic mammography using a limited number of low-dose cone-beam projection images, Med. Phys. 30, 365 (2003). In this technique, individual images of the female breast are acquired from multiple projection angles using a digital X-ray detector. From these digital images acquired at different projection angles—that is, from the image data associated with these images—multiple layered images can be reconstructed, each depicting layers of the breast oriented parallel to the receiving surface of the X-ray detector. Such a reconstruction-generated image dataset is referred to below as a tomosynthetic 3D X-ray image. This method allows for better visualization of tissue structures than conventional projection radiography.

[0005] However, the exposure control described at the beginning is not possible due to the higher absorption of the digital X-ray detectors used in tomosynthesis.

[0006] To enable automatic exposure control even when using digital X-ray detectors, it is known, for example, from DE 101 64 170 A1 or DE 103 09 268 A1, to take a so-called preliminary image with a low dose before taking an X-ray image intended for diagnostic use. The required optimal X-ray dose for the final X-ray image is then determined based on this preliminary image. However, taking such a preliminary image leads to a significant, undesirable additional radiation exposure for both the patient and the operating personnel during tomosynthesis, where numerous X-ray images are taken from different directions.

[0007] The invention is based on the objective of providing a method for taking a digital X-ray image of an object under examination, thereby avoiding the aforementioned disadvantage. Furthermore, the invention is based on the objective of providing a device for carrying out the method.

[0008] With regard to the method, the aforementioned problem is solved by the features of claim 1. In the method for generating a digital tomosynthetic 3D X-ray image of an examination object, a plurality of individual images with different projection angles relative to the normal of the patient positioning table are recorded of an examination object using an X-ray tube and a digital X-ray detector and assembled to form the tomosynthetic 3D X-ray image, wherein at least one model is recorded in an initial position with a dose that is smaller than the dose used for the individual images, which is evaluated to determine the recording parameters required for recording the individual images, and wherein the dose is used as one recording parameter and is weighted for each projection angle with a weighting factor that depends on this.This procedure ensures correct exposure of all individual images with a single model when producing a digital tomosynthetic X-ray image.

[0009] In an advantageous embodiment of the method, particularly in the case where the X-ray tube is pivoted on a circular path around the object under investigation, this weight factor can be determined by the relationship c k = 1 / cosα k given, where α k which specifies the angle relative to the normal. With such a weighting factor c k Correct exposure is ensured even when X-rays strike the receiving surface of the X-ray detector at an angle.

[0010] In a particularly preferred embodiment of the method, the model is recorded with a spatial resolution corresponding to the spatial resolution of the individual images. This makes it possible to use the model recorded to determine the recording parameters for tomosynthetic reconstruction, either directly or by combining it with a subsequent image recorded in the initial position, taking into account the recording parameters determined for that initial position, to create a complete individual image belonging to that initial position.

[0011] Alternatively, the model can be captured with a lower spatial resolution than the spatial resolution used to capture the individual images. This allows for faster readout and evaluation of the model.

[0012] With regard to the equipment, the problem according to the invention is solved with an equipment having the features of claim 9, the advantages of which result analogously from the advantages of the associated method claim.

[0013] For further explanation of the invention, please refer to the drawing. It shows: Fig. 1. A device according to the invention in a schematic diagram, Fig. 2 a flowchart that explains the method according to the invention.

[0014] According to Fig. 1 The mammography unit comprises an X-ray tube 2 for generating X-rays 3 that pass through an examination object 4. The examination object 4 is a female breast, which is embedded between a compression plate 6 and a patient positioning table 8. The X-rays 3 passing through the examination object 4, the compression plate 6, and the patient positioning table 8 are received by a large-area digital X-ray detector 10, which is composed of a multitude of individual detectors 12 arranged in a matrix-like array. Starting from an initial position k=0, the X-ray source 2 can be moved to different angular positions k up to a maximum projection angle α. max (largest deflection) are swivelled so that the object under investigation 4 is viewed with different projection angles α k relative to the normal 13 of the patient positioning table 8 individual images E kcan be generated. In the initial position k=0 (projection angle α0), a model V0 is generated with a reduced dose, which is used in a control and evaluation unit 14 containing an image computer to determine the parameters for correct exposure of the individual images E. k The required recording parameters are evaluated. The subsequently obtained, correctly exposed individual images E k In the control and evaluation unit 14, the images are reconstructed to form a tomosynthetic 3D X-ray image T, which consists of a plurality of layer images that represent different object planes 15 parallel to the patient positioning table 8.

[0015] The angular position k of the X-ray tube 2 and its operating parameters are controlled by control signals S generated by the control and evaluation unit 14. Using input and display elements, symbolically illustrated in this example by a keyboard 16 and a monitor 18, the user can select and execute various procedure variants, which are explained below. As an alternative to the embodiment shown in the figure with a stationary X-ray detector 10, the X-ray tube 2 and the X-ray detector 10 can also be moved together around the stationary object 4 under investigation.

[0016] According to Fig. 2. In a first step, a reference image V0 with a reduced dose is acquired in an initial position k=0 at a projection angle α0. Based on this reference image V0, the acquisition parameters for the individual images E0 - E required to generate a tomosynthetic 3D X-ray image T are then determined.n The dose required to produce a correctly exposed single image E0 in the initial position k=0 is determined. In the initial position k=0, either the single image E0 is then produced, or the X-ray tube is swiveled to an angular position corresponding to the projection angle α1, and a single image E1 is acquired in this position. The dose correct for this position, for example, the exposure time (irradiation duration) with other constant acquisition parameters (tube voltage, anode-filter combination), is weighted by a weighting factor c1 corresponding to this projection angle α1. This weighting factor is determined by the relationship c k = 1 / cosα k This is given when the starting position corresponds to a projection angle α0 = 0° and the X-ray tube is rotated along a circular path. In this way, n individual images E1 to E are acquired. nThese are generated. As shown in the left half of the figure, they are combined with the model V0 to form a tomosynthetic 3-D X-ray image T.

[0017] Alternatively, the projection angle α0 in the initial position k=0 of the largest displacement α can be max to correspond to one side. Then, immediately after the model V0, the corresponding single image E0 is taken according to the exposure time calculated from the model V0. In the next steps, the X-ray tube is moved to the projection angles α1 to α n panned and the individual images E1 to E are displayed n recorded.

[0018] If, in the initial position k=0, the model V0 was acquired with a spatial resolution corresponding to that of the individual image E0, a subsequent image F0 can be generated in this initial position k=0, which is then combined with the model V0 to form the individual image E0. However, it is also fundamentally possible to use a high-resolution model V0 directly for tomosynthetic reconstruction.

Claims

[1] Method for generating a digital tomosynthetic 3D X-ray image (T) of an object of investigation (4) using an X-ray tube (2) and a digital X-ray detector (10), wherein a plurality of individual images (E) of the object of investigation (4) k ) with different projection angles (α k ) relative to the normal (13) of the patient positioning table (8), and in which a starting position (k=0) is used with a dose that is smaller than that of the individual images (E k ) dose used, at least one model (V0) is recorded, which is used to determine the dose required to record the individual images (E k ) the required intake parameters are evaluated, whereby the dose is used as one intake parameter, characterized by , that for every projection angle (α k ) with a weight factor dependent on this (c k ) is weighted and used to adjust the different projection angles (α) k) the X-ray tube (2) is swung on a circular path around the object under investigation (4) and the weight factor (c) k ) at least for the projection angles (α) different from the starting position (k=0). k ) through the relationship c k = 1 / cosα k is given, where c k the weight factor (c k ) and α k the projection angle (α k ) is. [2] A method according to any of the preceding claims, wherein the model (V0) is recorded with a spatial resolution that corresponds to the spatial resolution of the individual images (E k ) corresponds. [3] Method according to one of claims 1, wherein the model (V0) is recorded with a spatial resolution that is smaller than the spatial resolution of the individual images (E k ) is. [4] Method according to claim 2, wherein the model (V0) is combined with a subsequent image (F0) taken in the initial position (k=0) taking into account the recording parameters determined for this initial position (k=0) to form the single image (E0) belonging to this initial position (k=0). [5] Method according to claim 2, wherein in the initial position (k=0) only a model (V0) is recorded, and wherein this model (V0) is used for the composition of the tomosynthetic 3D x-ray image (T) taking into account the recording parameters available for this model (V0). [6] Method according to one of the preceding claims, wherein the model (V0) is recorded at the projection angle α0=0°. [7] Method according to any one of claims 1 to 6, wherein the model (V0) at one of the largest deflections corresponds to a projection angle α0=α max is recorded. [8] Device for generating a digital tomosynthetic 3D X-ray image (T) of an examination object (4) with a digital X-ray detector (10) and an X-ray tube (2) for taking a plurality of individual images (E k ) from the object under investigation (4) with different projection angles (α k ) is arranged to pivot relative to the normal (13) of the patient positioning table (8), and is equipped with a control and evaluation unit (14) for capturing at least one model (V0) in a starting position (k=0) with a dose that is smaller than that of the individual images (E k ) dose used, and to determine the dose required to take the individual images (E k ) each required recording parameter by evaluation of the model (V0), where a recording parameter is the dose that is required for each projection angle (α) k ) with a weight factor dependent on this (c k ) is weighted, characterized by, that in the control and evaluation unit (14) software communicating with operating and display elements (16,18) for carrying out the methods according to one of claims 1 to 7 is implemented.

Images