GENERATION OF CONTOURTechnical FieldEmbodiments of the present disclosure relate to systems and methods for generating a contour of a patient or an organ of the patient. More specifically, embodiments of the disclosure relate to a computer-implemented method for generating a contour of a patient or an organ of the patient, and a data processing apparatus, a computer program, a non-transitory computer-readable storage medium and a computer program product configured to execute a method for generating a contour of a patient or an organ of the patient.
BackgroundA contour of a patient (the boundary between the patient and their external environment) is often required for radiation treatment planning. For example, a treatment planning system may require information on the full extent of the patient’s body to simulate the distribution of dose delivered to the patient; to calculate the total delivered dose; and/or to prevent a collision between the radiotherapy system (e.g., the rotatable gantry) and the patient. Furthermore, the contour of the patient can be used to define a bounding volume for treatment planning calculations to improve their efficiency. Such uses may be applicable to any radiotherapy delivery mode such as proton, photon, and electron radiotherapy. It can also be beneficial to obtain a contour of an organ of the patient, e.g., for calculating the total dose delivered to the organ. Information about the boundary of an organ is particularly important for organs particularly at risk from radiation damage for which radiation dose is undesirable.
Contours may be generated using contour tracing algorithms, such as the Suzuki contour tracing algorithm, square tracing algorithm, or Moore-neighbour contour tracing algorithm. However, in general, the process is time consuming. Generating a contour of the patient is particularly time consuming because the region to be contoured is large (e.g., the whole patient) . Furthermore, the complexity of the clinical anatomy and the noise associated with medical imaging can prevent the contour tracing algorithm from generating high quality contours. For example, various devices and accessories are frequently used during radiotherapy treatment, e.g., to position the patient. These can include, for example, patient beds, patient supports and replaceable top surfaces therefor, head frames in all types of radiotherapy systems, and (in the case of a Magnetic Resonance Linear Accelerator, where the patient is also imaged) Radio Frequency imaging coils and associated stands. Such devices can be included in the contour, and therefore significant user input is required to identify these devices and remove them from the contour.
Patient contours and organ contours can alternatively be generated with deep learning algorithms. However, such techniques require extensive model training and tuning and the use of clinical data. This is time consuming and can give rise to privacy and ownership issues.
Therefore, there is a need for improved methods and systems for generating a contour of a patient or an organ of the patient.
SummaryEmbodiments of the disclosure seek to provide improved methods and systems for generating a contour of a patient or an organ of the patient.
According to a first aspect, there is provided a computer-implemented method for generating a contour of a patient or an organ of the patient. The method comprises obtaining a three-dimensional binary image of the patient or the organ of the patient. The binary image comprises a plurality of voxels having one of a first value and a second value. The method further comprises processing the binary image to determine one or more connected volumes. Each connected volume consists of a plurality of voxels having the first value that are located close enough to each other to satisfy a connection criterion. The method further comprises identifying the largest connected volume of the one or more connected volumes, where the largest connected volume comprises the largest number of voxels, and removing, from the binary image, all of the one or more connected volumes except the largest connected volume to obtain an updated binary image. The method further comprises performing a contouring algorithm on the updated binary image to obtain the contour of the patient or the organ of the patient.
According to a second aspect, there is provided a data processing apparatus comprising a memory storing computer-executable instructions and a processor configured to execute the instructions to carry out the method of the first aspect.
According to a third aspect, there is provided a computer program comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method of the first aspect.
According to a fourth aspect, there is provided a non-transitory computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to carry out the method of first aspect.
According to a fifth aspect, there is provided a computer program product comprising the computer-readable storage medium of the fourth aspect.
Thus, the present disclosure provides an improved method for generating a patient contour in which connected volumes are defined in a binary image of the patient or organ, and a contouring algorithm is performed on the largest connected volume only. By removing all but the largest connected volume from the binary image, it is possible to reliably obtain a contour of the patient or organ without incorporating unwanted objects, features or noise. As such, the desired contour is reliably achieved with little to no user input required. Furthermore, there is no requirement to use clinical data to train and tune a machine learning model to perform the contouring. The present disclosure thereby provides a simple and efficient solution for generating a contour of a patient or an organ of the patient.
Brief description of the drawingsExemplary embodiments will now be described, by way of example only, with reference to the following drawings, in which:
Fig. 1 depicts a method of generating a contour of a patient;
Fig. 2 is a flow chart showing a computer-implemented method according to various embodiments of the present disclosure;
Fig. 3 is a schematic illustration of a method for generating a contour according to various embodiments of the present disclosure;
Fig. 4A is a schematic illustration of a 3D medical image;
Fig. 4B is a schematic illustration of two example 2D cuts through a 3D medical image;
Fig. 5 is a set of images showing a generated contour of a patient according to various embodiments of the present disclosure;
Fig. 6 is a block diagram illustrating a data processing apparatus according to various embodiments of the present disclosure; and
Fig. 7 is a schematic illustration of a radiotherapy apparatus.
Detailed descriptionFig. 1 depicts a method 100 of generating a contour of a patient that demonstrates some of the problems that are addressed by the present disclosure.
The method 100 comprises obtaining a three-dimensional (3D) medical image 101 of the patient. The medical image 101 could be, for example, a computed tomography (CT) scan or a magnetic resonance (MR) image of the patient lying on a bed. In Fig. 1, the 3D medical image is represented by a schematic of a 2D cut through a transverse plane of the 3D medical image 101, which shows the patient’s torso 101a and arms 101 b as well as the patient’s lungs 101c.
A user defines a volume of interest (VOI) with respect to the 3D medical image 101 of the patient, which is depicted in Fig. 1 by the rectangular box around the image 101. As illustrated in Fig. 1, the VOI typically includes components in addition to the patient, such as the bed 101d and other objects 101e that are in the vicinity of the patient. For example, these objects 101e could include devices and accessories such as positioning aids for maintaining the patient in a suitable position or imaging devices for imaging the patient. Any such devices and accessories may be referred to herein as peripheral devices.
At step 102, the VOI of the 3D medical image 101 is binarized to generate a 3D binary image 103 (an image in which each voxel has one of a first value and a second value) . The 3D binary image 103 is again represented in Fig. 1 by a 2D transverse cut through the image 103. The voxels having the first value are shaded and the voxels having the second value are unshaded. Certain features of the medical image 101 may be excluded by the binarization. For example, in the transverse cut of the binary image 103 of Fig. 1, the bed has been excluded by the binarization criteria (i.e., the voxels of the bed have the second value and therefore the bed is not visible in the binary image 103) . This will occur when the image intensity of the bed is sufficiently different to the image intensity of the patient.
At step 104, a contouring algorithm, such as the Suzuki contour tracing algorithm or Moore-neighbour contour tracing algorithm, is applied to the binary image 103 to generate a 3D contour image 105. The contour (s) of the contour image 105 show the boundary or boundaries between the voxels having the first value and the voxels having the second value.
As shown in Fig. 1, the contour image 105 obtained by this method includes contours for the features that were present in the binary image VOI 103 in addition to the patient, such as the peripheral devices 101e. This can lead to longer processing times for performing the contouring algorithm. Furthermore, a user must then determine the relevant contour, e.g., the contour showing the external boundary of the patient, and manually discard any unwanted contours.
As such, there is a need for improved methods and systems for generating a contour of a patient or an organ of the patient.
Fig. 2 is a flow chart showing a method 200 for generating a contour according to embodiments of the present disclosure. The contour may be a contour of a patient or a contour of an organ of the patient. A contour of a patient is the boundary between the patient and their external environment (also referred to as the external contour or external boundary of the patient) . A contour of an organ is the boundary of the organ, i.e., the boundary between the organ and the rest of the patient’s body.
The method 200 of Fig. 2 comprises, at step 201, obtaining a 3D binary image of the patient or the organ of the patient. The binary image may be obtained by binarizing a medical image of the patient. The medical image of the patient may be, for example, an X-ray image, CT image, MR image, positron emission tomography (PET) image, ultrasound image, or any other type of image. The medical image may be a greyscale or colour image.
Image binarization, also referred to as image thresholding, is the conversion of an image (e.g., a greyscale or colour image) to a binary image. The binary image comprises a plurality of voxels having one of a first value and a second value. For example, the binary image may be a 3D image in which each voxel of the image has either a first value, such as 0 for white, or a second value, such as 255 for black.
Binarizing the medical image of the patient may comprise allocating the first value to voxels of the medical image having an image intensity that satisfies a binarization criterion, and allocating the second value to voxels of the medical image having an image intensity that does not satisfy the binarization criterion. For example, the binarization criterion may be defined as being satisfied by an image intensity that is within a range of image intensity values. Thus, any voxels having an image intensity that is outside this range do not satisfy the binarization criterion. This may be referred to as a window/level binarization scheme.
The range of image intensity values used for binarization may be defined (e.g., inputted) by a user. For example, a display or graphical user interface (GUI) may show the binarized image for any image intensity range defined by the user, and the user may use this image to determine a suitable range, e.g., by varying the range until the patient or the organ is (comfortably) visible in the binary image. Alternatively, the step 201 of obtaining the three-dimensional binary image may comprise a computer-implemented step of determining the image intensity range. For example, the computer may select a default image intensity range for the region for which a contour is to be generated (e.g., based on the region specified by a user) . In such cases, the user may be given the opportunity to accept or alter the default image intensity range.
The step 201 of obtaining the binary image may comprise determining a VOI comprising the patient or the organ of the patient. The VOI may be determined before or after the medical image has been binarized. For example, the VOI may be determined with respect to the medical image and this VOI may be subsequently binarized. Alternatively, the medical image may be binarized first, and the VOI subsequently determined with respect to this binarized image. In either case, the obtained binary image corresponds to (e.g., consists of) the VOI.
The VOI is a volume that comprises the region for which a contour is to be generated, i.e., the patient or the organ of the patient. A size and/or position of the VOI may be determined such that the patient or the organ of the patient is the largest component of the VOI. The VOI may be cuboidal or any other shape. The VOI may be referred to as a bounding box.
The VOI may be determined based on a user input, e.g., the VOI may be inputted by the user via a display or GUI. Alternatively, the VOI may be automatically determined by the computer. For example, the determination of the VOI may be based on an algorithm and/or a machine learning model that has been trained to recognise a patient body or an organ of interest (the organ for which a contour is to be generated) . In such cases, the computer-generated VOI may be displayed to the user, and the user may be given the opportunity to accept or alter the VOI.
The method 200 of Fig. 2 further comprises processing 202 the binary image to determine one or more connected volumes. Each connected volume consists of a plurality of voxels having the first value that are located close enough to each other to satisfy a connection criterion. For example, for each voxel that has the first value, the connection criterion defines which (if any) other voxels having the first value are considered connected to that voxel. The connected voxels form a connected volume.
The connection criterion may be based on voxel connectivity of the voxels. Voxel connectivity refers to the manner in which the voxels of the image relate to their neighbours. For example, the connection criterion may be satisfied by voxels that are 6-connected. A 6-connected connection criterion in three dimensions requires that any given voxel having the first value is considered connected to other voxels having the first value that contact the given voxel via respective ones of their faces, i.e., there is face-to-face contact between the voxels. A 6-connected connection criterion in 3D can be expressed mathematically as: a voxel at (x, y, z) having the first value is connected to the voxels at (x±1, y, z) , (x, y±1, z) and (x, y, z±1) that also have the first value.
Alternatively, the connection criterion may be satisfied by voxels that are 18-connected. An 18-connected connection criterion in 3D requires that a given voxel having the first value is considered connected to other voxels having the first value that contact the given voxel either via respective ones of their faces, or via respective ones of their edges, i.e., there is face-to-face contact or edge-to-edge contact between the voxels. An 18-connected connection criterion in 3D can be expressed mathematically as: a voxel at (x, y, z) having the first value is connected to the 6-connected voxels that have the first value and the voxels at (x±1, y±1, z) , (x±1, y, z±1) , (x, y±1, z±1) , andthat have the first value.
Alternatively, the connection criterion may be satisfied by voxels that are 26-connected (also referred to as face, edge, vertex connectivity) . For a 26-connected connection criterion in 3D, a given voxel having the first value is considered connected to voxels having the first value that contact the given voxel via respective ones of their faces, via respective ones of their edges or via respective ones of their corners, i.e., there is face-to-face contact, edge-to-edge contact or corner-to-corner contact between the voxels. A 26-connected connection criterion in 3D can be expressed mathematically as: a voxel at (x, y, z) having the first value is connected to the 18-connected voxels that have the first value and the voxels at (x±1, y±1, z±1) , andthat have the first value.
The connection criterion may be defined such that peripheral devices that are present in the binary image are not considered connected to the patient (i.e., the body of the patient) . In other cases, the connection criterion may be defined such that anatomical features of the patient that neighbour the organ of interest and are visible in the binary image (e.g., anatomical features inside the body of the patient) are not considered connected to the organ.
The method 200 of Fig. 2 further comprises a step 203 of identifying the largest connected volume of the one or more connected volumes. The largest connected volume is the connected volume that comprises the largest number of voxels. The largest connected volume will correspond to the region for which a contour is to be generated, i.e., the patient or the organ of the patient. For example, a VOI may have been chosen such that the patient or the organ of the patient is the largest component of the VOI (and therefore of the binary image) .
The method 200 further comprises removing 204, from the binary image, all of the one or more connected volumes except the largest connected volume to obtain an updated binary image. Thus, the updated binary image comprises only the connected volume corresponding to the region for which a contour is to be generated, i.e., the patient or the organ of the patient. Removing all of the one or more connected volumes except the largest connected volume may comprise setting the plurality of voxels of all of the one or more connected volumes except the largest connected volume to the second value. Alternatively, all voxels of the binary image except those of the largest connected volume may be set to the same value (which may be any value, e.g., a third value) .
The steps of determining 202 the connected volumes based on a connection criterion, identifying 203 the largest connected volume, and removing 204 all but the largest connected volume may be referred to as a connectivity filter.
The method 200 of Fig. 2 further comprises performing 205 a contouring algorithm (also referred to as a contour tracing algorithm) on the updated binary image to obtain the contour of the patient or the organ of the patient. The contouring algorithm may generate one or more contours (e.g., surfaces) where the voxels change from the first value to another value (e.g., from the first value to the second value) . The contouring algorithm may be, for example, one of:the Suzuki contour tracing algorithm (also referred to as the Suzuki algorithm, the Suzuki contouring algorithm or the Suzuki contour algorithm) ; the square tracing algorithm, and the Moore-neighbour contour tracing algorithm. However, those skilled in the art will appreciate that other contouring algorithms may be used in step 205, and the present disclosure is not limited to any particular contouring algorithm.
The contouring algorithm may operate by dividing the 3D updated binary image into a plurality of 2D image slices; scanning each 2D slice of the updated binary image pixel by pixel; and identifying which pixels share the same pixel value. In this manner, the contouring algorithm may identify the contours as the boundaries at which the pixel value changes (e.g., by marking a set of boundary pixels) . Some contour tracing algorithms may be used in combination with a hole searching algorithm to identify internal contours (those bounded by the outermost contour) , while other contour tracing algorithms may identify internal contours without the use of a hole searching algorithm.
The updated binary image may have only one contour, e.g., because all of the voxels inside the patient (or the organ of the patient) have the first value, and all of the voxels outside the patient (or the organ of the patient) have a different value, e.g., the second value. In other words, the updated binary image may not show any anatomical detail inside the patient (or the organ of the patient) . For example, such detail may have been removed/lost during the binarization at step 201.
Alternatively, the updated binary image may have more than one contour, e.g., because the binary image comprises some detail inside the patient or inside the organ of the patient (e.g., anatomical features) . Thus, some of the voxels bounded by the external boundary of the patient or organ have the second value. An example of an updated binary image having more than one contour is described with reference to Fig. 3.
In some embodiments of the method 200 of Fig. 2, performing 205 the contouring algorithm on the updated binary image to obtain the contour may comprise, responsive to determining that the updated binary image has a plurality of contours, selecting an outermost contour of the plurality of contours (the outermost contour may be referred to as the external contour or parent contour) . For example, the one or more contours that are surrounded by the outermost contour (referred to as internal contours, inner contours, or child contours) may be removed and/or ignored by the contouring algorithm. If the 3D binary image is of the patient, then the outermost contour will be the boundary between the patient and the external environment. If the 3D binary image is of an organ of the patient, then the outermost contour will be the boundary of the organ.
If the updated binary image has only one contour, then this contour is selected. This contour is the contour of the patient or the organ of the patient (since the largest connected volume corresponds to the patient or the organ of the patient) .
The method 200 of Fig. 2 may further comprise determining an aspect of a radiation treatment plan based on the obtained contour. The aspect of the radiation treatment plan may comprise one or more of: a radiation dose distribution for radiotherapy treatment of the patient; a configuration of a radiotherapy system for radiotherapy treatment of the patient; a Source Surface Distance (SSD) for the patient; a construction of bolus for radiotherapy treatment of the patient; and a configuration of an imaging system for imaging the patient.
For example, the obtained contour (e.g., the size, position and/or shape of the contour) may define the scope of a dose calculation for radiotherapy treatment of the patient. If the contour represents the external boundary of the patient, then the contour can be used to calculate the total dose delivered to the patient for different dose distributions. If the contour represents the boundary of an organ, then the contour can be used to calculate the total dose delivered to that organ. This is particularly useful for monitoring and minimising the dose delivered to an organ at risk (e.g., an organ for which treatment dose is undesirable) .
The size, position and/or shape of the contour may be used to calculate the SSD, e.g., the distance from the radiotherapy source to the external surface of the patient or to the boundary of an organ.
In some radiotherapy treatments, bolus may be used to reduce or alter the dose to the patient. Bolus is a material that behaves similar to tissue when irradiated. The contour generated according to the method 200 of Fig. 2 may be used, e.g., by a physician, to determine the appropriate construction of bolus for the radiotherapy treatment of the patient. The construction of bolus may, for example, refer to any of the position, shape, size and/or composition of the bolus. In a similar manner, the contour may be used to determine the construction of other types of external structures used during radiotherapy treatment of the patient.
The contour may be used to determine the configuration of the radiotherapy system. For example, in certain radiotherapy treatments, a gantry is rotated around the patient to increase the dose delivered to unhealthy tissue and reduce the dose delivered to healthy tissue (e.g., the rotatable gantry 702 described with reference to Fig. 7) . The contour of the patient may be used to determine the positions of the gantry during treatment, e.g., to obtain the desired dose distribution or to avoid a collision between the radiotherapy apparatus and the patient.
Radiation treatment planning requires images of the patient to be treated. Therefore, in some embodiments, the contour may be used for determining a configuration of an imaging system for imaging the patient, e.g., a position of the imaging source relative to the patient.
The method 200 for generating the contour may be used for treatment planning in advance of the treatment. Alternatively, the method 200 may be used in situ as part of adaptive treatment planning, e.g., to account for movement of the patient.
Fig. 3 depicts an example implementation of the method 200 of Fig. 2. The method 300 of Fig. 3 is a computer-implemented method for generating a contour of a patient from a 3D medical image. As described above, the contour of the patient is the boundary between the patient and their external environment, which is often required for radiation treatment planning. The steps of the method 300 of Fig. 3 are depicted using 2D cuts through 3D images. However, it will be understood that the steps are performed on 3D images.
For illustration purposes, a schematic representation of the 3D medical image used in the method 300 of Fig. 3 is shown in Fig. 4A. The image 400 includes the torso 400a and arms 400b of the patient, and further shows the patient’s lungs 400c comprising a left lung and a right lung. For the sake of clarity, the bed and the peripheral devices are not shown in Fig. 4A. Fig. 4B shows two example 2D cuts through the 3D medical image 400 of Fig. 4A along the transverse planes AB and CD respectively (these planes are depicted as dashed lines in Fig. 4A) . The first 2D transverse cut 401 along plane AB includes the patient’s torso 401a, arms 401 b, and lungs 401c, as well as the bed 401d on which the patient is lying and other peripheral devices 401e. The second transverse 2D cut 402 along plane CD includes the patient’s torso 402a and arms 402b, and the bed 402d.
The method 300 of Fig. 3 comprises obtaining the 3D medical image 400 of the patient, which is represented in Fig. 3 by the 2D transverse cut 301 along the plane AB. The method 300 further comprises determining a VOI of the 3D medical image 400. The VOI is determined such that the patient is the largest component of the VOI. In Fig. 3, the VOI is represented by a rectangular box around the medical image 301. However, it will be understood that the VOI is three dimensional.
The method 300 further comprises a step 302 of obtaining a 3D binary image of the patient, which is depicted in Fig. 3 as a 2D transverse cut 303 of the 3D binary image. The shaded parts of the image 303 represent voxels that satisfy a binarization criteria and therefore have a first value, while the unshaded parts of the image 303 represent voxels that do not satisfy the binarization criteria and therefore have a second value. In the example shown in Fig. 3, the voxels defining the patient’s body and the peripheral devices all satisfy the binarization criteria and therefore have the first value (shown in the binary image 303 of Fig. 3 with dotted shading) . The voxels inside the patient’s lungs, the voxels inside the bed, and the remaining voxels outside of the patient and peripheral devices do not satisfy the binarization criteria and therefore have the second value (shown in the binary image 303 of Fig. 3 with no shading) .
At step 304, the binary image 303 is processed using a connectivity filter (as described with reference to Fig. 2) to determine the connected volumes; to identify the largest connected volume of the identified connected volumes, which in this case corresponds to the patient’s body; and to remove all of the connected volumes except the largest connected volume to obtain an updated binary image 305. The updated binary image 305 includes only the patient’s body as a single connected volume (i.e., the peripheral devices have been removed in the updated binary image 305) .
At step 306, a contouring algorithm is performed on the updated binary image 305 to obtain a contour image 307. The contouring algorithm produces a plurality of contours corresponding to the external boundary of the patient 307a and the boundaries of the patient’s left and right lung 307b, 307c. These contours are 3D surfaces showing where the voxels change from the first value to the second value.
At step 308, the method 300 comprises, responsive to determining that the updated binary image has a plurality of contours 307, selecting the outermost contour 307a of the plurality of contours 307a, 307b, 307c. Contours surrounded by the outermost contour are labelled and removed. The outermost contour 307a is the contour of the patient, i.e., the boundary between the patient and their external environment. Thus, the contour of the patient 309 is obtained.
Steps 306 and 308 of Fig. 3 may be performed by the Suzuki contour tracing algorithm, which is a hierarchical contour detection algorithm in which both outer contours (parent contours) and inner contours (child contours) are identified and labelled. Since the Suzuki contour tracing algorithm can perform both of steps 306 and 308, it is a particularly efficient option for identifying the outermost contour. In contrast, if the square tracing algorithm or Moore-neighbour contour is used to perform step 306, then an extra hole searching algorithm may be required to perform step 308 (e.g., to remove contours 307b and 307c) .
Although the method 300 of Fig. 3 is for generating a contour of a patient, those skilled in the art will understand that the method can also be used for generating a contour of an organ of the patient. For example, this can be achieved by altering the binarization criteria and/or the VOI to obtain a binary image of the organ.
Fig. 5 shows a set of images that compare results obtained using the method 100 of Fig. 1 (top row of Fig. 5) with results obtained using the method 200 of Fig. 2 (bottom row of Fig. 5) . This comparison demonstrates some of the advantages of the methods disclosed herein.
The first image 501 on the top row of Fig. 5 is a transverse cut of a 3D binary image generated by binarizing a medical image of a patient (e.g., according to step 102 of the method 100 of Fig. 1) . The second 502 and third 503 images on the top row are coronal and sagittal cuts of the same 3D binary image. The dashed rectangle in each image is a corresponding cut through the VOI. In this example, the VOI includes the patient and excludes the patient bed. Although difficult to make out in the images 502 and 503, the VOI also includes a head support device, the position of which is shown by the dashed circle 503a. The fourth image 504 is the contour obtained by the method 100 of Fig. 1. The contour 504 includes the head support device 504a because the contouring method is unable to distinguish between the device and the patient’s body.
The bottom row of Fig. 5 shows corresponding 2D cuts 511, 512, 513 for a 3D binary image generated according to step 201 of the method 200 of Fig. 2. The dashed rectangle in each image shows the VOI which, in this example, includes the bed. Although difficult to make out in the images 512 and 513, the VOI also includes a head support device, the position of which is shown by the dashed circle 513a. The fourth image 514 on the bottom row of Fig. 5 shows the contour generated by the method 200 of Fig. 2 (i.e., the method of the present disclosure) . The contour 514 includes only the patient’s body; the head support device is not included in the contour 514. Thus, the present disclosure provides a simple and efficient method of generating a contour that is effective even when there are peripheral devices and/or other features in the VOI of the binary image. The method of the present disclosure can identify and remove features that are not part of the patient’s body or the organ of interest. Furthermore, the use of the connectivity filter to remove these components prior to performing the contouring algorithm means that the contouring process is simpler and more efficient than if the contouring algorithm were to be performed on the entire binary image without the use of the connectivity filter.
Fig. 6 is a block diagram illustrating a data processing apparatus 600 according to some embodiments of the present disclosure. The data processing apparatus 600 may be, for example, a computer.
The data processing apparatus 600 comprises a memory 601 storing computer-executable instructions. The data processing apparatus 600 further comprises a processor 602 configured to execute the instructions to carry out the various methods described above (e.g., the method 200 of Fig. 2) .
For example, the steps of the methods described in relation to Fig. 2 may be performed by computer code stored on the data processing apparatus 600. The steps of the methods described herein may be performed in any suitable order unless a step is explicitly described as following or preceding another step and/or where it is implicit that a step must follow or precede another step.
The computer program and/or the code for performing such methods may be provided to the apparatus 600 on one or more computer readable media or, more generally, a computer program product. The computer readable media may be transitory or non-transitory. The one or more computer readable media could be, for example, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, or a propagation medium for data transmission, for example for downloading the code over the Internet. Alternatively, the one or more computer readable media could take the form of one or more physical computer readable media such as semiconductor or solid state memory, magnetic tape, a removable computer diskette, a random access memory (RAM) , a read-only memory (ROM) , a rigid magnetic disc, and an optical disk, such as a CD-ROM, CD-R/W or DVD. The instructions may also reside, completely or at least partially, within the memory 601 and/or within the controller circuitry during execution thereof by the computing system, the memory 601 and the controller circuitry also constituting computer-readable storage media.
In an implementation, the modules, components and other features described herein can be implemented as discrete components or integrated in the functionality of hardware components such as ASICS, FPGAs, DSPs or similar devices.
A “hardware component” is a tangible (e.g., non-transitory) physical component (e.g., a set of one or more processors) capable of performing certain operations and may be configured or arranged in a certain physical manner. A hardware component may include dedicated circuitry or logic that is permanently configured to perform certain operations. A hardware component may comprise a special-purpose processor, such as an FPGA or an ASIC. A hardware component may also include programmable logic or circuitry that is temporarily configured by software to perform certain operations.
In addition, the modules and components can be implemented as firmware or functional circuitry within hardware devices. Further, the modules and components can be implemented in any combination of hardware devices and software components, or only in software (e.g., code stored or otherwise embodied in a machine-readable medium or in a transmission medium) .
Fig. 7 depicts a radiotherapy apparatus according to some embodiments of the present disclosure. Those skilled in the art will appreciate that the techniques of the present disclosure are also applicable to other types of radiotherapy and radiotherapy systems.
Fig. 7 shows a cross-section through a radiotherapy apparatus 700 comprising a radiation head 704 and a beam receiving apparatus 706, both of which are attached to a gantry 702. The radiation head 704 includes a radiation source 707 which emits a beam of radiation 722.
The radiation head 704 also includes a beam shaping apparatus 750 which controls the size and shape of the radiation field associated with the beam.
The beam receiving apparatus 706 is configured to receive radiation emitted from the radiation head 704, for the purpose of absorbing and/or measuring the beam of radiation. In the view shown in Fig. 7, the radiation head 704 and the beam receiving apparatus 706 are positioned diametrically opposed to one another.
The gantry 702 is rotatable and supports the radiation head 704 and the beam receiving apparatus 706 such that they are rotatable around an axis of rotation 705, which may coincide with the patient longitudinal axis. As shown in Fig. 7, the gantry provides rotation of the radiation head 704 and the beam receiving apparatus 706 in a plane which is perpendicular to the patient longitudinal axis (e.g. a sagittal plane) . Three gantry directions XG, YG, ZG can be defined, where the YG direction is perpendicular with gantry axis of rotation. The ZG direction extends from a point on the gantry corresponding to the radiation head, towards the axis of rotation of the gantry. Therefore, from the patient frame of reference, the ZG direction rotates around as the gantry rotates.
Fig. 7 also shows a support surface 710 on which a subject (or patient) is supported during radiotherapy treatment. The radiation head 704 is configured to rotate around the axis of rotation 705 such that the radiation head 704 directs radiation towards the subject from various angles around the subject in order to spread out the radiation dose received by healthy tissue to a larger region of healthy tissue while building up a prescribed dose of radiation at a target region.
The radiotherapy apparatus 700 is configured to deliver a radiation beam towards a radiation isocentre which is substantially located on the axis of rotation 705 at the centre of the gantry 702 regardless of the angle at which the radiation head 704 is placed.
The rotatable gantry 702 and radiation head 704 are dimensioned so as to allow a central bore 780 to exist. The central bore 780 provides opening sufficient to allow a subject to be positioned therethrough without the possibility of being incidentally contacted by the radiation head 704 or other mechanical components as the gantry rotates the radiation head 704 about the subject.
As shown in Fig. 7, the radiation head 704 emits the radiation beam 722 along a beam axis 790 (or radiation axis or beam path) , where the beam axis 790 is used to define the direction in which the radiation is emitted by the radiation head. The radiation beam 722 is incident on the beam receiving apparatus 706 which can include at least one of a beam stopper and a radiation detector. The beam receiving apparatus 706 is attached to the gantry 702 on a diametrically opposite side to the radiation head 704 to attenuate and/or detect a beam of radiation after the beam has passed through the subject.
The radiation beam axis 790 may be defined as, for example, a centre of the radiation beam 722 or a point of maximum intensity.
The beam shaping apparatus 750 delimits the spread of the radiation beam 722. The beam shaping apparatus 750 is configured to adjust the shape and/or size of a field of radiation produced by the radiation source. The beam shaping apparatus 750 does this by defining an aperture (also referred to as a window or an opening) of variable shape to collimate the radiation beam 722 to a chosen cross-sectional shape. In this example, the beam shaping apparatus 750 may be provided by a combination of a diaphragm and a multi-leaf collimator (MLC) . Beam shaping apparatus 750 may also be referred to as a bean modifier.
The radiotherapy apparatus 700 may be configured to deliver both coplanar and non-coplanar (also referred to as tilted) modes of radiotherapy treatment. In coplanar treatment, radiation is emitted in a plane which is perpendicular to the axis of rotation of the radiation head 704. In non-coplanar treatment, radiation is emitted at an angle which is not perpendicular to the axis of rotation. In order to deliver coplanar and non-coplanar treatment, the radiation head 704 can move between at least two positions, one in which the radiation is emitted in a plane which is perpendicular to the axis of rotation (coplanar configuration) and one in which radiation is emitted in a plane which is not perpendicular to the axis of rotation (non-coplanar configuration) .
In the coplanar configuration, the radiation head is positioned to rotate about a rotation axis and in a first plane. In the non-coplanar configuration, the radiation head is tilted with respect to the first plane such that a field of radiation produced by the radiation head is directed at an oblique angle relative to the first plane and the rotation axis. In the non-coplanar configuration, the radiation head is positioned to rotate in a respective second plane parallel to and displaced from the first plane. The radiation beam is emitted at an oblique angle with respect to the second plane, and therefore as the radiation head rotates the beam sweeps out a cone shape. The beam receiving apparatus 706 remains in the same place relative to the rotatable gantry when the radiotherapy apparatus is in both the coplanar and non-coplanar modes. Therefore, the beam receiving apparatus 706 is configured to rotate about the rotation axis in the same plane in both coplanar and non-coplanar modes. This may be the same plane as the plane in which the radiation head rotates.
The beam shaping apparatus 750 is configured to reduce the spread of the field of radiation in the non-coplanar configuration in comparison to the coplanar configuration.
The radiotherapy apparatus 700 includes a controller 740 which is programmed to control the radiation source 707, beam receiving apparatus 706 and the gantry 702. Controller 740 may perform functions or operations such as treatment planning, treatment execution, image acquisition, image processing, motion tracking, motion management, and/or other tasks involved in a radiotherapy process.
Controller 740 is programmed to control features of apparatus 700 according to a radiotherapy treatment plan for irradiating a target region, also referred to as a target tissue, of a patient. The treatment plan includes information about a particular dose to be applied to a target tissue, as well as other parameters such as beam angles, dose-histogram-volume information, the number of radiation beams to be used during therapy, the dose per beam, and the like. Controller 740 is programmed to control various components of apparatus 700, such as gantry 702, radiation head 704, beam receiving apparatus 706, and support surface 710, according to the treatment plan.
Hardware components of controller 740 may include one or more computers (e.g., general purpose computers, workstations, servers, terminals, portable/mobile devices, etc. ) ; processors (e.g., central processing units (CPUs) , graphics processing units (GPUs) , microprocessors, digital signal processors (DSPs) , field programmable gate arrays (FPGAs) , special-purpose or specially-designed processors, etc. ) ; memory/storage devices such as a memory (e.g., read-only memories (ROMs) , random access memories (RAMs) , flash memories, hard drives, optical disks, solid-state drives (SSDs) , etc. ) ; input devices (e.g., keyboards, mice, touch screens, mics, buttons, knobs, trackballs, levers, handles, joysticks, etc. ) ; output devices (e.g., displays, printers, speakers, vibration devices, etc. ) ; circuitries; printed circuit boards (PCBs) ; or other suitable hardware. Software components of controller 140 may include operation device software, application software, etc.
The radiation head 704 may be connected to a head actuator 730 which is configured to actuate the radiation head 704, for example between a coplanar configuration and one or more non-coplanar configurations. This may involve translation and rotation of the radiation head 704 relative to the gantry. In some implementations, the head actuator may include a curved rail along which the radiation head 704 may be moved to adjust the position and angle of the radiation head 704. The controller 740 may control the configuration of the radiation head 704 via the head actuator 730.
The beam shaping apparatus 750 includes a shaping actuator 732. The shaping actuator is configured to control the position of one or more elements in the beam shaping apparatus 750 in order to shape the radiation beam 722. In some implementations, the beam shaping apparatus 750 includes an MLC, and the shaping actuator 732 includes means for actuating leaves of the MLC. The beam shaping apparatus 750 may further comprise a diaphragm, and the shaping actuator 732 may include means for actuating blocks of the diaphragm. The controller 740 may control the beam shaping apparatus 750 via the shaping actuator 732.
A treatment plan may comprise positioning information of beam shaping apparatus 750. The positioning information of beam shaping apparatus 750 may comprise information indicating a configuration of one or more elements of beam shaping apparatus 750, such as leaf configuration of an MLC of beam shaping apparatus 750, a configuration of a diaphragm of beam shaping apparatus 750, a configuration of an opening (e.g., window or aperture) of the MLC, and/or the like.
Unless specifically stated otherwise, as apparent from the following discussion, it is appreciated that throughout the description, discussions utilizing terms such as "receiving” , “determining” , “comparing ” , “enabling” , “maintaining, ” “identifying, ” , “obtaining” , “accessing” or the like, refer to the actions and processes of a computer system, or similar electronic computing device, that manipulates and transforms data represented as physical (electronic) quantities within the computer system's registers and memories into other data similarly represented as physical quantities within the computer system memories or registers or other such information storage, transmission or display devices.
While certain embodiments have been described, these embodiments have been presented by way of example only and are not intended to limit the scope of the claims appended hereto. It will be appreciated by those skilled in the art that other examples may be employed apart from these specific details. Indeed, the novel methods and apparatus described herein may be embodied in a variety of other forms; furthermore, various omissions, substitutions and changes in the form of methods and apparatus described herein may be made.
In some instances, detailed descriptions of well-known methods and devices have been omitted so as not to obscure the description with unnecessary detail. Where appropriate, the technology may be considered to be embodied within any form of computer-readable memory, such as solid-state memory, magnetic disk, or optical disk containing an appropriate set of computer instructions that would cause a processor to carry out the techniques described herein.