Search

EP-4234010-B1 - REMOVING ARTEFACTS IN RADIOTHERAPY IMAGING

EP4234010B1EP 4234010 B1EP4234010 B1EP 4234010B1EP-4234010-B1

Inventors

  • SAYEED, Abdul

Dates

Publication Date
20260506
Application Date
20230223

Claims (13)

  1. A radiotherapy device comprising: a source of kV or MV radiation; a detector configured to detect the kV or MV radiation to generate a plurality of images of a subject located between the source and the detector; and a controller configured to: detect an erroneous pixel in an image of the plurality of images; and generate an averaged image, comprising generating each pixel of the averaged image by taking an average of respective pixels in corresponding locations of one or more of the plurality of images, wherein generating a pixel of the averaged image which is in a corresponding location to the erroneous pixel comprises excluding the erroneous pixel from the taking of the average, wherein the image comprises a row of erroneous pixels including the erroneous pixel, and wherein the controller is configured to detect the row of erroneous pixels in the image of the plurality of images.
  2. A radiotherapy device according to claim 1, wherein generating the averaged image comprises generating a pixel of the averaged image which is not in a corresponding location to the erroneous pixel by taking an average of n pixels, n being an integer, and generating the pixel in the averaged image which is in the corresponding location to the erroneous pixel comprises taking an average of fewer than n pixels.
  3. A radiotherapy device according to claim 1 or claim 2, wherein the averaged image is generated based on pixels of the image other than the erroneous pixel or wherein the averaged image is generated based on one or more pixels of the image adjacent to the erroneous pixel.
  4. A radiotherapy device according to any preceding claim, wherein the controller is configured to detect the row of erroneous pixels by summing pixel values of the image along each of its rows to determine a plurality of summed pixel values and identifying the row of erroneous pixels as a row corresponding to a summed pixel value that exceeds a threshold or that exceeds an average of the plurality of summed pixel values by a predetermined factor, and to exclude each erroneous pixel of the row of erroneous pixels from the generating of respective pixels of the averaged image.
  5. A radiotherapy device according to any preceding claim, wherein the controller being configured to generate the averaged image comprises: the controller being configured to temporally average the plurality of images to generate the averaged image, wherein each pixel in the averaged image is generated by taking an average of respective pixels in corresponding locations of multiple of the plurality of images; or the controller being configured to spatially average an image of the plurality of images to generate the averaged image, wherein each pixel in the averaged image is generated by taking an average of a respective sub-grid of pixels in a corresponding location of the image.
  6. A radiotherapy device according to any preceding claim, wherein the erroneous pixel corresponds to an artefact caused by application of a radiotherapy beam at the same time as reading out of data associated with the erroneous pixel, or wherein the controller is configured to detect the erroneous pixel and generate the averaged image in real-time during a radiotherapy treatment.
  7. A computer-implemented method comprising: obtaining a plurality of images of a subject located between a source and a detector of a radiotherapy device, the source being a source of kV or MV radiation and the detector being for detecting the kV or MV radiation; detecting an erroneous pixel in an image of the plurality of images; and generating an averaged image, comprising generating each pixel of the averaged image by taking an average of respective pixels in corresponding locations of one or more of the plurality of images, wherein generating a pixel of the averaged image which is in a corresponding location to the erroneous pixel comprises excluding the erroneous pixel from the taking of the average, wherein the image comprises a row of erroneous pixels including the erroneous pixel, and wherein the method comprises detecting the row of erroneous pixels in the image of the plurality of images.
  8. A computer-implemented method according to claim 7, wherein generating the averaged image comprises generating a pixel of the averaged image which is not in a corresponding location to the erroneous pixel by taking an average of n pixels, n being an integer, and generating the pixel in the averaged image which is in the corresponding location to the erroneous pixel comprises taking an average of fewer than n pixels.
  9. A computer-implemented method according to claim 7 or claim 8, comprising generating the averaged image based on pixels of the image other than the erroneous pixel or comprising generating the averaged image based on one or more pixels of the image adjacent to the erroneous pixel.
  10. A computer-implemented method according to any of claims 7-9, comprising detecting the row of erroneous pixels by summing pixel values of the image along each of its rows to determine a plurality of summed pixel values and identifying the row of erroneous pixels as a row corresponding to a summed pixel value that exceeds a threshold or that exceeds an average of the plurality of summed pixel values by a predetermined factor, and further comprising excluding each erroneous pixel of the row of erroneous pixels from the generating of respective pixels of the averaged image.
  11. A computer-implemented method according to any of claims 7-10, wherein the generating of the averaged image comprises: temporally averaging the plurality of images to generate the averaged image, wherein each pixel in the averaged image is generated by taking an average of respective pixels in corresponding locations of multiple of the plurality of images; or spatially averaging an image of the plurality of images to generate the averaged image, wherein each pixel in the averaged image is generated by taking an average of a respective sub-grid of pixels in a corresponding location of the image.
  12. A computer-implemented method according to any of claims 7-11, wherein the erroneous pixel corresponds to an artefact caused by application of a radiotherapy beam at the same time as reading out of data associated with the erroneous pixel.
  13. A computer-readable medium comprising computer-executable instructions which, when executed by a processor, cause the processor to perform the method of any of claims 7-12.

Description

This disclosure relates to artefacts in radiotherapy imaging, and in particular to removing artefacts in MV/kV images. Background Radiotherapy can be described as the use of ionising radiation, such as X-rays, to treat a human or animal body. Radiotherapy is commonly used to treat tumours within the body of a patient or subject. In such treatments, ionising radiation is used to irradiate, and thus destroy or damage, cells which form part of the tumour. A radiotherapy device typically comprises a gantry which supports a beam generation system, or other source of radiation, which is rotatable around a patient. For example, for a linear accelerator (linac) device, the beam generation system may comprise a source of radio frequency energy, a source of electrons, an accelerating waveguide, beam shaping apparatus, etc. In radiotherapy treatment, it is desirable to deliver a prescribed dose of radiation to a target region of a subject and to limit irradiation of other parts of the subject, i.e. of healthy tissue. In view of this, a radiotherapy device may comprise one or more imaging devices for capturing images of the patient before and/or during a radiotherapy treatment, which can be used to make adjustments to machine parameters or patient location. Such image-guided radiation therapy (IGRT) can improve the accuracy of radiotherapy treatments through aiding delivery of an intended dose in an intended location. Images captured before the radiotherapy treatment begins may provide reference images for the shape and/or location of the patient, and/or may help in positioning the patient in an intended position. Images captured during the radiotherapy treatment may be used to verify that the patient remains in an intended position. Discrete, gross or large-scale movements of a subject may include shifting position, coughing or sneezing. The subject may also undergo cyclical, physiological movement. For example, the subject may undergo respiratory motion due to their breathing cycle. The subject may also undergo cardiac motion based on beating of their heart. In response to determination of such movements using the captured images, the radiotherapy treatment may be halted or adjusted to compensate, for example through gating or tracking of the radiotherapy beam. These techniques may improve clinical outcomes through ensuring that a prescribed dose is delivered to a tumour and that irradiation of healthy tissue such as organs at risk is limited. A radiotherapy device may comprise components configured to perform MV imaging and may comprise components configured to perform kV imaging. A treatment beam source of a radiotherapy device may emit MV radiation for treating the patient. This treatment beam source may be used as the MV beam source for MV imaging. An MV detector may be disposed diametrically opposite the treatment beam source, with the subject therebetween. A radiotherapy device may comprise a kV imaging source, and a kV detector arranged diametrically opposite to the kV imaging source with the subject therebetween. The kV imaging source and the kV detector may be arranged in a different plane (i.e. at a different angle) to the treatment beam source and the MV detector. MV or kV image acquisition during radiotherapy treatment (i.e. during MV radiation delivery) results in an artefact in the obtained images. This effect is present when images are read out while MV radiation is being delivered. The MV radiation is delivered in a pulsed manner such that pixels of a detector are affected by the artefact if the reading out of their data overlaps with the delivery of the MV radiation. The effect is seen even when the relevant detector is not in the radiation beam, and is thought to be related to electromagnetic interference. This artefact manifests in 2D captured images as bright or dark lines (depending on inversion of the images), or as rings in reconstructed 3D volumes. This artefact reduces the accuracy of captured images since, due to the artefact, some of the pixels of the image are not accurate or reliable representations of the regions imaged. When decisions to halt or adjust radiotherapy are taken based on such images, this in turn will reduce the accuracy and reliability of the radiotherapy treatment. One approach for addressing this issue is to use a hardware solution to synchronise when radiation is on and when an image line is read out such that image lines are only read out when radiation is not being delivered. However, this restriction slows down the imaging and requires specific hardware to put it into effect, for example a pulse control circuit, pulse synchronisation circuitry or a detector control board (DCB). In addition, this hardware solution does and will not work for continuous variable dose rate (CVDR) treatments, for which the pulsing rate of the radiation delivery changes continuously. It would be advantageous to provide more accurate and more reliable imaging during radiotherapy treatments.