System and method for providing an extended image of a patient

Abstract

The present invention discloses a computer-based method for obtaining a 3D image of a part of a patient's body, the method being based on a segmented image with a limited field of view and extending the field of view using information from a contour image of the patient obtained from a scan of the patient's surface. Anatomical data from the planning image is preferably used to fill the contour image by means of a contour-guided deformable registration between the planning image and the contour.

Description

Technical Field

[0001] The present invention relates to a method, a computer program product and a computer system for providing an image of a patient, which can be used, for example, for radiation therapy planning. Background Art

[0002] Radiation therapy is typically distributed to a patient in multiple courses or multiple fractions. Before treatment begins, a planning image of the patient is obtained. The planning image provides input data to the treatment plan, which defines the treatment to be given to the patient in terms of treatment mode, dose, beam angle and other variables. Before each course, a segmented image is obtained to help position the patient relative to the treatment unit before delivery. The segmented image can also be used to evaluate the dose actually delivered to the patient during the course of treatment, and to study changes in the patient's geometry since the planning image was acquired. These changes are important for treatment evaluation, and the results of the evaluation may lead to a decision to modify the treatment plan. In the context of this specification, both planning images and segmented images are typically 3D images constructed from multiple 2D images that constitute a projection of the patient's body.

[0003] Dose planning requires information about the location of various organs as well as their material properties, such as density and / or atomic composition. Density information is used for dose planning. If photon radiotherapy is used, density and atomic composition determine the attenuation of the radiation. If ion radiotherapy, such as proton radiotherapy, is used, density and atomic composition determine the stopping power, which will affect how far the ions travel in the patient. For initial planning, this information is taken from the planning images. Segmented images are often used to determine new boundaries of the region of interest in order to properly aim the radiation beam.

[0004] Therefore, the planning images should include not only information about the contours, but also information about the material properties of each region of interest. As the geometry of the tumor and other tissues changes during treatment, segmented images are used to obtain the most up-to-date contour information. However, segmented images may have much less information than planning images, especially information about material properties. For example, fan-beam CT scans (referred to as CT in this document) can be used for planning images, while cone-beam CT (CBCT) scans are used for segmentation images. CT images include all the information required for dose planning, but are relatively expensive, and CT imaging equipment is more difficult to incorporate into a gantry than CBCT imaging equipment. On the other hand, CBCT generally gives images with a more limited field of view than CT images.

[0005] Typically, the field of view of a CBCT image does not cover the entire patient outline. This means that when calculating dose based on a CBCT image such as a segmented image, the density of parts outside the field of view of the CBCT image needs to be estimated for the dose calculation. One solution is to superimpose the patient outline from the planning image onto the segmented image and assume that it has the same properties as water. Water is a reasonable compromise as it is a good approximation for most parts of the body, but this solution still leads to inaccuracies in the dose calculation.

[0006] Ruchala, Olivera, Kapatoes, Reckwerdt, and Mackie: Methods for improving limited field-of-view radiotherapy reconstruction using imperfect a prioriimages, Medical Physics 29, 2590 (2002); doi: 10.1118 / 1.1513163 discloses a method for dealing with this problem directly in 2D images, which are then used to construct 3D images. The method proposed by Ruchala et al. is not applicable when dealing directly with 3D images. Typically, treatment planning systems do not provide access to 2D images, so any changes made by these systems must be made in the 3D images.

[0007] European patent publication EP3181049 proposes using data from a planning image outside the field of view of the segmented image to create a model outline of a part of the patient's body and including material information from the planning image outside the field of view of the segmented image to obtain a composite image with information from the segmented image (if available) and information from the planning image outside the area. The planning image is not always available, or the patient may have changed so much, for example due to weight loss, that the planning image becomes less relevant. Summary of the invention

[0008] It is an object of the present invention to provide an extended image of a patient, comprising, based on a first image with a limited field of view, the extended image having a larger field of view.

[0009] According to the invention, this object is achieved by a computer-based method for creating a composite image of a patient, the method comprising the following steps

[0010] a. providing a first image of a portion of a person, wherein the first image has a first field of view;

[0011] b. Provide a surface scan corresponding to the exterior of a person;

[0012] c. Providing a second image of an outer contour of at least a portion of the person based on the surface scan, the outer contour including a region of the patient corresponding to the first field of view and extending outside the first field of view.

[0013] d. Providing approximate image data of the interior of the outer contour to generate an approximate image of the interior of the person;

[0014] e. Applying data from the first image in the area of ​​the approximate image corresponding to the first field of view to create a composite image.

[0015] In this way, a composite image is obtained that includes the latest information from the area where the segmentation image can be obtained, and anatomical information based on the surface scan and suitable for the segmentation image in the area outside the second field of view. Therefore, the composite image includes image data from the first image where the image data of the first image is obtained and approximate data in the second part outside the first part. The first image is generally referred to as the segmentation image in this document, and is usually, but not necessarily, a CBCT image. It should be understood that it can be any image for which it is desired to expand the field of view of the image.

[0016] The surface scan is used to generate an outer contour of the entire patient or a portion of the patient, filled with an approximation of the patient's anatomy.

[0017] The use of surface scanning devices is becoming more and more widespread in radiotherapy units, where they are used to ensure the correct positioning of the patient in order to deliver treatment fractions. According to the present invention, such devices can be used for the purpose of providing an approximate image of the patient, which can be used to extend the field of view of another image of the patient, for example, a CT or CBCT image. Providing an approximate whole-body image of the body outside the field of view is useful, for example, in dose planning discussed in the background technology section. Whole-body images can also be used for other purposes, such as, for example, positioning the patient for radiotherapy to ensure correct treatment and to ensure that the patient is not in the way when the treatment bed and / or gantry are moved. Such images are also useful for other applications, for example, for planning medical procedures such as surgery.

[0018] Approximate image data for the interior of the contour can be estimated in a variety of ways, depending on the type of information and the level of detail and accuracy required for a particular purpose. For example, the density value of the entire interior can be set to an appropriate density value, such as the density of water. The single density value can also be combined with data about the location and extent of the bone structure in the patient, such as obtained from X-ray imaging such as fluoroscopy.

[0019] In some embodiments, the step of providing approximating image data of the interior of the external contour comprises: including anatomical image data from an image such as a CT image. The CT image may be an earlier image of the same patient, or a representative atlas image taking into account the patient's characteristics. The image data from the CT image is adapted into the contour by means of deformable registration of the CT image to the external contour and setting the approximating image data based on the results of the deformable registration.

[0020] If bone structure data is available, the CT image can be deformably registered to both the external contours and data about the bone location and density within the patient.

[0021] The invention further relates to a computer program product for controlling a computer, preferably on a carrier such as a non-transitory storage medium, said computer program product comprising a computer readable code arranged to cause a computer to perform a method according to any of the preceding claims.

[0022] The method is mainly used in conjunction with a radiotherapy apparatus, but it may also have other uses as briefly discussed above. Therefore, the present invention also relates to a computer system comprising a processor, a data memory and a program memory. The data memory is arranged to store at least one set of surface data or contour information, and / or at least one segmented image, and the program memory is arranged to store a computer program product as defined above, so that the computer program product can be used to control the processor. It will be appreciated that the method of the present invention itself can be executed in any computer with the necessary software installed.

[0023] The approximation image and the segmentation image preferably comprise the same type of data. This means that if the segmentation image is a CT or CBCT image, the same type of data should be used within the contour. If the segmentation image is an MRI image, the data provided within the contour should be the MRI image. It is also possible to create a synthetic CT image from the MRI image, i.e. an image comprising the same type of data as the CT image. Such a synthetic CT image can be used as a planning image or a segmentation image, together with other CT or CBCT images as segmentation images.

[0024] Obtaining anatomical data for approximation images from a high quality CT scan such as a planning image is particularly advantageous as it provides high quality approximation data without the need for a full CT scan each time. This saves time and resources and also avoids unnecessary doses to the patient. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The present invention will be described in more detail below by way of examples and with reference to the accompanying drawings, in which

[0026] Figure 1a and Figure 1bA 3D image of the patient's head and neck is shown. Figure 1c A complete outline of the same area is shown.

[0027] Figure 2a , Figure 2b and Figure 2c A contour obtained from a surface scan of a patient is shown, with approximate anatomical information inside.

[0028] Figure 3a and Figure 3b An example of a composite image is shown where a 3D patient image has values ​​within a limited field of view acquired from the 3D image and values ​​outside the limited field of view based on a surface scan of the patient.

[0029] Figure 4 is a flow chart of a method according to the present invention.

[0030] Figure 5 is a schematic diagram of a system that can be used to implement the present invention. DETAILED DESCRIPTION

[0031] Figure 1a A coronal slice of a 3D image is shown, including anatomical data about the patient, but with a limited field of view, wherein the outer contour of the patient is not included in at least some directions. In particular, in this example, the top of the patient's head and the area from the shoulders down are not included. Figure 1b A patient outline from a 3D image is shown. The method according to the invention is applicable to any type of anatomical image, but may be particularly useful for segmented images used in radiation therapy planning. Therefore, in this specification, the term segmented image is often used, it being understood that the invention can be used to extend any patient image with a limited field of view. The patient outline is obtained using known methods and is referred to as a segmented outline in this document.

[0032] The goal is to obtain an image with all the data from the segmented image, but with a larger field of view than the limited field of view. In areas outside the limited field of view, contour data obtained from a surface scan of the patient and possibly approximate anatomical data are used to supplement the segmented image. Figure 1c This outline of the patient's head and neck area is shown and can be used to supplement Figure 1a The resulting segmented image supplemented with the approximate contours of the patient's body parts outside the limited field of view of the segmented image is referred to herein as an enhanced segmented image. An enhanced segmented image is referred to as a composite image in which contour information and approximate anatomical data are added outside the field of view of the segmented image based on information from the surface scan.

[0033] Figure 2a , Figure 2b and Figure 2cSchematically shown are a first, a second and a third simplified approximation image of a patient based on a patient contour 21 obtained by surface scanning. Figure 2a In , the patient’s density is assumed to be uniform, so an approximate image is created by setting the density inside the contour to a suitable density, such as the density of water. Figure 2b A more detailed approximation of the patient's interior is included, indicated by the presence of a first organ 23 and a second organ 25 inside the patient, which may correspond to, for example, the liver and the heart. Figure 2c Another embodiment is shown in which image data about the patient's bone structure, which is shown slightly simplified, has been included. Such data can be obtained, for example, by suitable X-ray methods. In this example, the rest of the interior is assumed to have the density of water. Preferably, the patient's interior in the approximated image is a true image of the patient, based on a specific image that can be an earlier image or an atlas image of the patient. The co-pending application (EP19180004.4) discusses in detail how the interior of a patient can be approximated with various levels of accuracy to produce an approximated image based on a surface scan.

[0034] In a preferred embodiment, the approximation image is based on the patient's contour data obtained from a surface scan and internal data about the anatomical structure obtained from an earlier taken planning image. This approximation image with a level of detail and accuracy selected based on the requirements of the situation can be used together with the segmented image to obtain a composite image. The surface scan is preferably performed later to provide contour data that is as accurate as possible. In this embodiment, the current anatomical data within the contour is approximated by a deformable registration between the planning image and the contour data, and the approximated anatomical data within the contour is set based on the deformation results of the planning image. In addition to this, information obtained from X-ray imaging, such as fluoroscopy obtained at the same time as the surface scan, can also be added to the patient's contour. Information about the bone structure obtained from the X-ray image can be used to improve the deformable registration of the planning image to the contour to improve the image outside the field of view of the segmented image. Figure 2c A simplified approximation image including the patient's skeletal structure is schematically shown.

[0035] The planning image usually has a larger field of view than the segmented image, but does not always cover the entire part of the patient that should be imaged. In many cases, the planning image can be used to extend the segmented image using the planning image as discussed in European patent publication EP3181049 and using contour data outside the field of view of the planning image. Thus, in this case, the approximation image can include data from the deformed planning image in the field of view of the planning image and approximation data obtained in another way in the rest of the contour, such as for example for Figure 2a , Figure 2b or Figure 2c obtained as discussed.

[0036] Figure 3a A composite image is shown, where Figure 1a The images shown have been Figure 2b The outlines 21 shown are combined to create a composite image having accurate data in the head and neck area and approximate data for the rest of the body.

[0037] Figure 3b A composite image is shown in which data in the field of view of a segmented image 29 is acquired from the segmented image, and data outside the field of view is acquired from an approximation image 21 based on a surface scan and internal data obtained, for example, from a planning image, as described above. In this example, as is sometimes the case, the segmented image does not include any portion of the patient's contour. As described above, the approximation image preferably also includes anatomical data obtained from an image, such as a planning image of the patient. Of course, the data acquired from the segmented image may include only data from a limited portion of the segmented image, for example if it is known that certain portions of the image are of low quality.

[0038] Figure 4 is a flow chart of the method according to the invention. The input data is a 3D image (I41) with a limited field of view, typically a CBCT image used as a segmented image, and contour data (I42) covering the entire patient or at least a portion of the patient that is larger than the limited field of view of the 3D image, which includes at least a portion of the patient's external contour based on a surface scan. An example of how to calculate contour data based on a surface scan is given in (EP19180004.4).

[0039] In the first step S41, an appropriate density is set inside the contour. Figure 2a to Figure 2c As discussed, this can be done with an appropriate level of precise detail, depending on what the situation requires. In a preferred embodiment, the density is set based on a deformation of a high-quality CT image taken at an earlier time, such as a planning image. The density can also be obtained from an atlas image representing a similar patient. In this case, the CT image or atlas image is registered to the outer contour to obtain anatomical data that can be applied to the inside of the outer contour.

[0040] In a second step S42, data from the segmented image is included in the approximation image. This can be achieved by setting the data values ​​in the portion of the approximation image corresponding to the field of view that should be included in the segmented image to 0, and then inserting the data values ​​from the segmented image into the portion. Typically, data from the entire segmented image is used so that the composite image will have data from the segmented image in the entire field of view of the segmented image. However, it will be appreciated that data from only a portion of the segmented image may be included. The output from this process is a composite image O41 that includes data from the segmented image (if available) and provides a larger field of view with the aid of an approximation image based on the surface contour of all or part of the patient. Typically, the surface scan on which the approximation image is based and the segmented image are taken at approximately the same time, in which case the images can be combined without registration. In other cases, rigid or deformable registration may help to adapt the images to each other.

[0041] Before adapting the model contour to the segmented image, an initial overlap between the contour and the segmented image is preferably determined. This can be achieved by using an automatic rigid registration algorithm between the contour and the segmented image. The images can also be superimposed based on a visual assessment of the initial overlap or in any other suitable manner.

[0042] Instead of using contour data associated with the same patient as the 3D image, in some cases it may be sufficient to provide contour data that is of a similar stature and shape to the patient but obtained from another source such as an atlas.

[0043] Figure 5 1 is a schematic diagram of a system 10 for radiation therapy and / or treatment planning. It will be appreciated that such a system may be designed in any suitable manner. Figure 5 The design shown is only an example. The patient 11 is located on the treatment couch 13. The system includes an imaging / treatment unit having a radiation source 15 mounted in a gantry 17 for emitting radiation toward the patient positioned on the treatment couch 13. Typically, the treatment couch 13 and the gantry 17 can be moved relative to each other in several dimensions to provide radiation to the patient as flexibly and correctly as possible. These components and their functions are well known to the technician. Typically, there are many passive devices provided for lateral and deep shaping of the beam, which will not be discussed in more detail here. In this example, the system also includes an optical scanner for providing data related to the patient's external contours and capable of generating contour images of the whole or part of the patient's body. The system also includes a computer 21, which can be used for radiation therapy planning and / or for controlling radiation therapy. It can be understood that the computer 21 can be an independent unit not connected to the imaging / treatment unit.

[0044] The computer 21 comprises a processor 23, a data memory 24 and a program memory 25. Preferably, there are also one or more user input devices 28, 29 in the form of a keyboard, a mouse, a joystick, a voice recognition device or any other available user input device. The user input device may also be arranged to receive data from an external storage unit.

[0045] The data store 24 includes clinical data and / or other information used to obtain a treatment plan, including contour data provided by a contour scanner. According to an embodiment of the present invention, the data store 24 also includes one or more patient images used in treatment planning. The nature of these patient images and how they are obtained have been discussed above. The program store 25 stores at least one computer program that is arranged to cause the processor to execute a program according to the present invention. Figure 4 The program memory 25 also stores a program arranged to cause the computer to execute the method in conjunction with Figure 4 A computer program for the method steps discussed and / or a computer program arranged to cause a computer to control radiation treatment of a patient.

[0046] It will be appreciated that the data memory 24 and program memory 25 are only schematically shown and discussed. There may be several data memory units, each storing one or more different types of data, or there may be one data memory that stores all data in a suitably structured manner, as well as the program memory. One or more memories may also be stored on other computers. For example, a computer may be arranged to perform only one of the methods, with another computer being used to perform the optimization.

Claims

1. A computer-based method for creating a composite image of a patient, comprising the steps of: a. providing a first image of a portion of a person, wherein the first image has a first field of view; b. providing a surface scan corresponding to the exterior of the person; c. providing a second image of an outer contour of at least a portion of the person based on the surface scan, the outer contour including a region of the patient corresponding to the first field of view and extending outside the first field of view; d. providing approximate image data of the interior of the outer contour to generate an approximate image of the interior of the person; e. Applying data from the first image in a region of the approximation image corresponding to the first field of view to create a composite image.

2. The method according to claim 1, wherein: The first image is a CBCT image.

3. The method according to claim 1, wherein: The first image is a segmented image.

4. The method according to any one of claims 1 to 3, wherein: The step of providing approximated image data of the interior of the outer contour comprises setting a suitable uniform density value within at least a portion of the interior of the outer contour.

5. The method according to any one of claims 1 to 3, wherein: The step of providing approximated image data of the interior of the external contour comprises including within the interior of the contour data relating to the location and density of bones within the patient.

6. The method according to claim 5, wherein: The step of providing approximating image data of the interior of the outer contour comprises a deformable registration of the CT image of the person to the outer contour, and setting the approximating image data based on a result of the deformable registration.

7. The method according to claim 6, wherein: The CT image may be deformably registered to the external contour and to data regarding the position and density of bones within the patient.

8. The method according to any one of claims 1 to 3, wherein: The step of providing approximating image data of the interior of the outer contour comprises deformable registration of an atlas image to the outer contour, and setting the approximating image data based on a result of the deformable registration.

9. A computer program product stored in a non-transitory storage device (25) which, when executed in a computer system (21), causes the computer system to perform a method according to any one of claims 1 to 8.

10. A computer system (21) for providing an approximate image of a patient, the computer system comprising a processing device (23), the computer system having a non-transitory storage device (25) in which a computer program product according to claim 9 is stored, so that when the computer program product is executed, it will control the processing device (23) to perform dose planning.

11. The computer system according to claim 10 further comprises a contour scanning device, which is arranged to provide contour data representing a contour or a body of the patient, and the processing device is arranged to calculate a model of the contour based on the contour data.

Images