Patent Grant
8116430
The present invention pertains in general to therapeutic radiology. In particular, the invention involves imaging devices.
An objective of radiation therapy is to maximize the amount of radiation to a target volume (e.g., a cancerous tumor) and minimize the amount of radiation to healthy tissues and critical structures. The process of identifying the precise location of the target volume immediately prior to a dose of therapeutic radiation is key to the objective. Since each patient is treated over 30 to 40 fractionated sessions, then the time allowed for each session is relatively short, e.g. 10 to 15 minutes, so the process must be fast as well as accurate.
In the case of electronic portal imaging, megavolt therapeutic X-rays emerging from the patient can be used to generate images. However, this method of target location generates images of low contrast and quality, in addition to incidentally damaging healthy tissue. As a result, imaging with megavoltage (MV) radiation is used primarily for portal verification, that is, to confirm that the treatment volume is being radiated.
Radiotherapy simulator machines have been used to perform the pre-treatment analysis of the target volume before a radiotherapy treatment machine applies the therapeutic radiation. However, traditional radiotherapy simulator machines use bulky image intensifier tube detectors to capture images of the treatment volume. These image intensifier tube detectors have the disadvantage of being very large relative to their imaging area. They also have image spatial distortions from their spherical shaped input surface and the orientation of the intensifier tube with the Earth's magnetic field.
A multi-mode cone beam computed tomography radiotherapy simulator and treatment machine is disclosed. The radiotherapy simulator and treatment machine both include a rotatable gantry on which is positioned a cone-beam radiation source and a flat panel imager. The flat panel imager captures x-ray image data to generate cone-beam CT volumetric images used to generate a therapy patient position setup and a treatment plan.
The present invention is illustrated by way of example and not limitation in the figures of the accompanying drawings, in which like references indicate similar elements and in which:
In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the present invention. It will be evident, however, to one skilled in the art that the present invention may be practiced without these specific details. In some instances, well-known structures and devices are shown in gross form rather than in detail in order to avoid obscuring the present invention. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention, and it is to be understood that other embodiments may be utilized and that logical, mechanical, electrical, and other changes may be made without departing from the scope of the present invention.
A clinical therapy simulation machine having a cone-beam computed tomograpy (CT) radiation source and a flat-panel imager is described. The clinical therapy simulation machine is capable of manipulating the flat-panel imager and the cone beam CT radiation source to generate x-ray images for determining patient setup/alignment and a clinical treatment plan to be implemented by a clinical treatment machine.
A treatment couch 218 is positioned adjacent to the gantry 202 to place the patient and the target volume within the range of operation for the radiation source 204 and the imager 206. The couch 218 may be connected to the therapy simulator rotatable gantry via a communications network and is capable of translating in multiple planes plus angulation for positioning and re-positioning the patient 205 and therefore the target volume.
The gantry 202 can rotate 214 about an isocenterline 215 to place the radiation source 204 and imager 206 at any position 360 degrees around the target volume, for example, to generate CT scan image data. As will be described, cone-beam CT image data can be used to generate a three-dimensional representation of the patient anatomy and the target volume. The image data may further be used to generate a treatment plan to tailor a dose of therapeutic radiation to the target volume.
In one embodiment the flat panel imager 206 is a real-time digital x-ray imager incorporating a large-area amorphous silicon sensor array with a high-sensitivity cesium iodide (Csl) scintillator. The flat panel imager may include a receptor module that incorporates the amorphous silicon sensor array, which accepts incoming X-ray photons and converts them to a digital video signal. The X-ray to light conversion may be provided by a thin or thick columnar Csl:Tl (cesium iodide: thalium doped) scintillator The scintillator may be vacuum deposited in a thin (e.g. 0.6 mm) layer or include individual CsI crystals (e.g., being approximately 9 mm thick with an approximate 0.38 mm×0.38 mm pixel pitch) supported in a housing with an aluminum window (e.g., approximately 1 mm thick). The top of the thin CsI scintillator may be coated with a reflective powder/epoxy mixture. Five sides of each thick crystal may be coated with a reflecting powder/epoxy mixture. The sixth side may be in contact with and face the flat-panel sensor. Alternatively, the scintillator components may have other dimensions.
The receptor module may also include a power supply module (e.g., 24 VDC power), interconnecting cables (e.g., fiber optic control and data cables), and drive and readout circuits followed by digital data conversion and transmission capabilities well known to those of ordinary skill in the art.
It should be appreciated that the flat panel imager may be a two-dimensional large flat panel imager that can operate, for example, at 15 to 30 frames per second (fps) over a wide range of dose. In this way, fluoroscopic, radiographic and cone-beam CT imaging can all be achieved with the same flat panel system. Typically, 300-900 projections may be collected during a single rotation of the gantry depending on the image resolution and dose requirements. Fewer projections allow for a faster collection of cone-beam CT image data (e.g., in 20 to 40 seconds depending on gantry speed limits), thereby, allowing for lower dose cone-beam CT images with less patient motion artifacts. Alternatively, the images may operate at other frame rates.
In one embodiment, the flat panel imager has a landscape orientation, an active area of 39.7×29.8 cm2 with 194 micron pixel pitch, and a pixel count of 2048×1536 pixels. It can operate at a frame rate of 7.5 fps in full resolution mode and at a frame rate of 30 fps in 2×2 binned mode—where the pixel count is reduced to 1024×768 pixels2. For example, the flat panel imager may be an amorphous silicon (a-Si) imager available from Varian Medical Systems of Palo Alto, Calif., under the tradename PaxScan™ 4030A. The PaxScan™ 4030A detectors are each 40 cm×30 cm. The detectors may be coupled to signal processing circuitry comprising a preamplifier stage with dynamically controllable signal gain, as described in U.S. Pat. No. 6,486,808, filed on Oct. 16, 2001, assigned to the assignee of the present invention and incorporated by reference, herein, to improve contrast resolution and dynamic range.
The readout electronics may also be located out of the path of the primary cone-beam CT radiation source 204. The flat panel imager 206 may also employ a split data-line where the top half of the array and the bottom half of the array are read out simultaneously. This allows the imager 206 to read out more rapidly and reduces the parasitic capacitance of the data-lines, which in turn reduces the noise gain of the readout charge amplifiers. It should be appreciated that only half of the frame time is used to read out the pixels. During the rest of the frame time, the sensor can be irradiated without generating any interference patterns due to the pulsing of the cone-beam CT radiation source 204. In addition, it should also be appreciated the control system of the flat panel imager 206 allows an external synchronization signal (from the computer 220) to initiate the readout of a frame. This allows the user to externally control when the imager will acquire an image.
In one embodiment, a command processor module 225 manages the receptor module, processes the digital video, and provides interfaces to other components of the simulator 100. The command processor module 225 may include a microcontroller-based, single board computer running a real-time operating system with acquisition, control, and interface software. Also, included in the command processor may be a high-speed digital video interface card, a dedicated image processor card to perform real-time image corrections, a system interface card, and a parallel output to transmit image data to an external image processor and display. Scan-converted digital and analog video may also be provided.
The captured cone-beam CT image projection data may be delivered and stored to a computer 220. As shown in
The cone-beam CT image projection data may also be transferred to a cone-beam CT reconstruction computer 221 that includes software designed to achieve rapid cone-beam CT image generation. The computer 221 can merge or reconstruct the image data into a three-dimensional representation of the patient and target volume. In one embodiment, cone-beam CT reconstruction software may allow for full-cone and partial-cone input data that can produce cone-beam CT images (e.g., approximately 26 to 48 cm diameter) at a specific source-to-imager distance (e.g., 140-150 cm). In addition, in this way, the clinical simulator machine 100 and cone-beam reconstruction software may also allow for large diameter (e.g., approximately 48 cm) axial image volumes.
In one embodiment, the cone-beam CT reconstruction software may transform the image projection data into volumetric CT image data. The volumetric CT image data may include full-fan and/or partial cone image data to reconstruct head size (e.g. 26 cm diameter×17 cm length) and body size (e.g. 48 cm diameter×15 cm length) volumes. For example, the partial-cone method may be used to obtain body size scans when the flat panel imager is not long enough to image the full body in each projection. If the 15 or 17 cm axial section is not wide enough and therefore does not cover sufficient anatomical volume, then multiple scans can be performed. For example, in the two scan case, the patient may be moved axially by 15 or 17 cm couch movements between scans and the reconstructed image volumes may then be merged to provide a 30 to 34 cm axial view.
In one embodiment, prior to reconstruction, the image projection data is preprocessed to account for x-ray beam and detector properties and the system electronic and geometric properties. The algorithm and its implementation is similar to that used in single slice computer tomography in reconstruction of fan beam data obtained with a one-dimensional detector. For partial cone beam reconstruction, the partial cone image projection data is extended to full cone beam image data and then reconstructed using a full cone beam reconstruction algorithm well known to those of ordinarily skill in the art, such as, for example, the Feldkamp cone beam reconstruction technique. It should be understood that the extension of the partial cone beam image data is performed using techniques similar to those used for the extension of partial fan data in well known single slice fan beam computer tomography.
In one embodiment, using the shape and distance data determined from the generated dimensional representation, the target volume may be automatically identified by the computer system 221 and/or by the inspection of an oncologist. The identified target volume may be applied to a radiotherapy planning computer system 220, which creates a treatment plan to be implemented by a clinical treatment machine. The visualization of the data along arbitrary planes, e.g. sagital, coronal, axial, beams eye view, etc., can be displayed to assist the oncologist. To further enhance the visualization, averaging of volume image data perpendicular to the plane of view, i.e. rectangular voxels may be used.
A treatment couch 418 is positioned adjacent to the gantry 402 to place the patient and the target volume within the range of operation for the radiation source 404 and the imager 406. The couch 418 can be capable of translating in multiple planes plus angulation for positioning and re-positioning the patient 405 and therefore the target volume.
The gantry 402 can rotate 414 about an isocenterline 415 to place the cone-beam CT radiation source 404 and imager 406 at any position 360 degrees around the target volume. The resulting megavoltage cone-beam CT image data can then be used to tailor a dose of therapeutic radiation based on at least the generated pre-defined treatment plan.
The flat panel imager 406 is similar to the flat panel imager 206 including the corresponding interconnects with a command processor module 425, a computer 420, a monitor 422, and a cone-beam reconstruction computer 421, corresponding with the command processor module 225, the computer 220, a monitor 422, and the cone-beam reconstruction computer 221, as described above. However, in one embodiment, the flat panel imager 406 may have its electronics unfolded from beneath the imager 406 and the input screen coating may be thicker (e.g., 9 mm vs. 0.6 mm). An example of a flat panel imager that may be used in the present invention is described in U.S. patent Ser. No. 10/013,199, now U.S. Pat. No. 6,800,898 B1, filed on Nov. 2, 2001, assigned to the assignee of the present invention and incorporated herein by reference.
The imager 406 may also interface with an accelerator interface control box 451. The accelerator interface control box 451 interfaces with an accelerator control computer 450 to provide synchronization and gated control between the imager 406 and the cone-beam CT radiation source 404 during treatment based on the generated treatment plan. As shown in
In one embodiment, the accelerator interface control box 451 includes a timing interface. The timing interface coordinates acquisition by the flat panel imager 406 and pulsing of the cone-beam CT radiation source 404. With this interface, as little as one radiation pulse (0.028 cGy at the isocenter) can be used to form projection images.
In one embodiment, the timing interface includes a National Instruments PCI 6602 data acquisition card from National Instruments Corporation of Austin, Tex. USA, that contains hardware and firmware for counting and timing pulses; computer software that provides control logic; and a user interface for the interface system. Alternatively, other cards may also be used.
A master clock signal is derived from a sync signal of the cone-beam CT radiation source 404, which may operate at 360 pulses/s (6 MV) or 180 pulses/s (15-18 MV), according to one embodiment. Using a counter on the National Instruments PCI 6602 card, the sync signal is divided down to produce a master clock, and hence are timed relative to the production of cone-beam CT radiation pulses from the cone-beam CT radiation source 404.
The master clock may be used to generate two control pulses, one that gates the cone-beam CT radiation source 404 on and off and the other that triggers the flat panel imager 406. In one embodiment, the frequency of these pulses is user selectable, and may be any value below 30 pulses/sec. The relative timing of the two pulses may also be user selectable. When the flat panel imager 406 is triggered there is a period, while the image is being read out (half a frame time) during which no beam from the cone-beam CT radiation source 404 is desired. A user can enter the appropriate delay that will prevent irradiation during the frame readout period of the imager. The length of the gate pulse of the cone-beam CT radiation source 404 is also user selectable. By adjusting the width of the gate pulse, the user can control the number of beam pulses emitted by the cone-beam CT radiation source 404 during each gate pulse.
It should be appreciated that the MV cone-beam CT flat panel imager 406 has a high quantum efficient 9 mm thick CsI:Tl screen (e.g., approximately 10% efficient at 6 MV), which preserves spatial resolution and minimizes dose to the patient by at least a factor of 5 over a standard 1 mm thin copper plate and less than 1 mm GOS (gadolinium oxysulfide) screens used in standard flat panel and screen-camera portal imaging. Therefore, images with as low as one 6 MV accelerator beam pulse (e.g., 0.028 cGy) per frame may be collected. In addition, a low patient dose of 8 to 16 cGy per cone-beam CT data set may be yielded for 300 to 600 CT image frames or projections per data set. The lower dose of the MV cone-beam CT radiation allows for more frequent use on each patient during the typical 30 to 37 fractionated treatment sessions. Moreover, reduced spatial resolution on the MV cone-beam CT scans can be afforded for faster processing time using the cone-beam reconstruction software on the CBR computer 421 to achieve rapid image generation.
It should be appreciated that a separate kV cone-beam CT radiation source (not shown) and another opposing flat panel imager (as described above on the simulator but not shown here) may also be coupled to the gantry 402 to perform a diagnostic cone-beam CT scan. For example, the kV cone-beam CT radiation source and opposing flat panel imager may be coupled to the treatment machine gantry 402 at an off axis of e.g. forty-five or ninety degrees from the MV cone-beam radiation source 404 and opposing imager 406. As before, software in the computers 420 and/or 421 may generate the three-dimensional representation of the patient anatomy and target volume from the cone-beam CT image data provided by the kV radiation source. The clinical treatment machine 400 may use the kV cone-beam CT image data to make any necessary adjustments to the treatment plan based on identified movement of the target volume or to determine the amount of patient repositioning required by the treatment couch 418 or collimator movements. In this way, the kV cone-beam CT radiation source and flat panel imager share a common axis of rotation with the MV cone-beam CT radiation source 404 and provide additional information for aligning the patient to the generated simulation treatment plan.
It should also be appreciated that in this way, either the clinical simulator machine 100 and/or the clinical treatment machine 400 diagnostic cone-beam CT image data can be used as a reference for applying the MV radiation beams.
It should also be understood that it is not necessary for the therapeutic radiation to be applied from the exact position(s) where any of the previously generated CT images were taken since the computer software can provide virtual two-dimensional representations for any desired radial location in-between the images.
It should be understood that although the clinical treatment machine 400 has been described as having a cone-beam radiation source, in alternative embodiments beam shaping, along with intensity modulation, may also be implemented based on the generated treatment plan by directing a therapeutic beam through a dynamic multileaf collimator. The multileaf collimator may include a series of stacked metal shims having a center of shim pairs where each shim of the pairs may be individually moved to create a shaped opening capable of shaping the therapeutic beam. To be effective, the radiation field should be large enough to radiate the entire tumor while at the same time minimize radiating healthy tissue. The collimator may be dynamic in that the shims can rapidly move to reshape the beam, which results in blocking the therapeutic beam from striking certain areas of the target volume based on the treatment plan. Such dynamic shaping may result in different areas of the tumor receiving different amounts of radiation over the time that a radiation dose is applied.
It should be appreciated that more or fewer processes may be incorporated into the methods illustrated in
It should be appreciated that a clinical simulation machine having a cone-beam radiation source and flat-panel imagers, as described, allows for identification of a target volume via fluoroscopic, radiographic, and cone-beam CT imaging. In this way, the generation of the treatment plan via the clinical simulation machine prior to the application of therapeutic radiation, increases the accuracy of treating the tumor target. Furthermore, embodiments of the invention as described above may capture images while the gantry is continuously rotating versus traditional systems that stop and shoot every, approximately, four degrees around the patient, thereby further lessening the time for completion.
It should also be appreciated that the cone-beam volumetric reconstruction software can utilize image projection data at non-uniformly spaced gantry angles. Thus the data collection does not require a precise gantry speed of rotation. There is a normalizing detector at the radiation source, which is used to correct for system output variations. In one embodiment, the support arms for the images 206 and 406 are sufficiently precise in mechanical stability during gantry rotation that no compensating spatial corrections are required.
Although the present invention has been described with reference to specific exemplary embodiments, it will be evident that various modifications and changes may be made to these embodiments without departing from the broader scope of the invention as set forth in the claims. Accordingly, the specification and drawings are to be regarded in an illustrative rather than a restrictive sense.
This application is a continuation of U.S. patent application Ser. No. 10/324,227 filed Dec. 18, 2002.
| Number | Name | Date | Kind |
|---|---|---|---|
| 3133227 | Brown et al. | May 1964 | A |
| 3144552 | Schonberg | Aug 1964 | A |
| 3193717 | Nunan | Jul 1965 | A |
| 4149247 | Pavkovich et al. | Apr 1979 | A |
| 4149248 | Pavkovich | Apr 1979 | A |
| 4208675 | Bajon et al. | Jun 1980 | A |
| 4209706 | Nunan | Jun 1980 | A |
| 4521808 | Ong et al. | Jun 1985 | A |
| 4593967 | Haugen | Jun 1986 | A |
| 4675731 | Takasu et al. | Jun 1987 | A |
| 4679076 | Vikterlof et al. | Jul 1987 | A |
| 4726046 | Nunan | Feb 1988 | A |
| 4741621 | Taft et al. | May 1988 | A |
| 4825393 | Nishiya | Apr 1989 | A |
| 4853777 | Hupp | Aug 1989 | A |
| 4868844 | Nunan | Sep 1989 | A |
| 5014292 | Siczek et al. | May 1991 | A |
| 5027818 | Bova et al. | Jul 1991 | A |
| 5039867 | Nishihara et al. | Aug 1991 | A |
| 5080100 | Trotel | Jan 1992 | A |
| 5099505 | Seppi et al. | Mar 1992 | A |
| 5117445 | Seppi et al. | May 1992 | A |
| 5168532 | Seppi et al. | Dec 1992 | A |
| 5247555 | Moore et al. | Sep 1993 | A |
| 5262649 | Antonuk et al. | Nov 1993 | A |
| 5335255 | Seppi et al. | Aug 1994 | A |
| 5394452 | Swerdloff et al. | Feb 1995 | A |
| 5438991 | Yu et al. | Aug 1995 | A |
| 5471546 | Meier | Nov 1995 | A |
| 5537452 | Shepherd et al. | Jul 1996 | A |
| 5661773 | Swerdloff et al. | Aug 1997 | A |
| 5692507 | Seppi et al. | Dec 1997 | A |
| 5751781 | Brown et al. | May 1998 | A |
| 5949811 | Baba et al. | Sep 1999 | A |
| 5956382 | Wiener-Avnear et al. | Sep 1999 | A |
| 5999587 | Ning et al. | Dec 1999 | A |
| 6031888 | Ivan et al. | Feb 2000 | A |
| 6041097 | Roos et al. | Mar 2000 | A |
| 6104778 | Murad | Aug 2000 | A |
| 6104780 | Hanover et al. | Aug 2000 | A |
| 6144875 | Schweikard et al. | Nov 2000 | A |
| 6219403 | Nishihara | Apr 2001 | B1 |
| 6222901 | Meulenbrugge et al. | Apr 2001 | B1 |
| 6292526 | Patch | Sep 2001 | B1 |
| 6307914 | Kunieda et al. | Oct 2001 | B1 |
| 6325537 | Watanabe | Dec 2001 | B1 |
| 6325758 | Carol et al. | Dec 2001 | B1 |
| 6370421 | Williams et al. | Apr 2002 | B1 |
| 6381302 | Berestov | Apr 2002 | B1 |
| 6385286 | Fitchard et al. | May 2002 | B1 |
| 6385288 | Kanematsu | May 2002 | B1 |
| 6429578 | Danielsson et al. | Aug 2002 | B1 |
| 6463122 | Moore | Oct 2002 | B1 |
| 6480565 | Ning | Nov 2002 | B1 |
| 6508586 | Oota | Jan 2003 | B2 |
| 6590953 | Suzuki et al. | Jul 2003 | B2 |
| 6661870 | Kapatoes et al. | Dec 2003 | B2 |
| 6744848 | Stanton et al. | Jun 2004 | B2 |
| 6842502 | Jaffray et al. | Jan 2005 | B2 |
| 6865254 | Nafstadius | Mar 2005 | B2 |
| 6888919 | Graf | May 2005 | B2 |
| 6914959 | Bailey et al. | Jul 2005 | B2 |
| 20010001807 | Green | May 2001 | A1 |
| 20010008271 | Ikeda et al. | Jul 2001 | A1 |
| 20020066860 | Possin | Jun 2002 | A1 |
| 20030007601 | Jaffray et al. | Jan 2003 | A1 |
| Number | Date | Country |
|---|---|---|
| 4223488 | Jan 1994 | DE |
| 19614643 | Oct 1997 | DE |
| 0062941 | Sep 1984 | EP |
| 0205720 | Dec 1986 | EP |
| 0480035 | Nov 1994 | EP |
| 0948930 | Oct 1999 | EP |
| 2 269 745 | Nov 1975 | FR |
| 2 551 664 | Mar 1985 | FR |
| 1328033 | Aug 1973 | GB |
| 63294839 | Dec 1988 | JP |
| 5057028 | Mar 1993 | JP |
| WO 8503212 | Aug 1985 | WO |
| WO 9014129 | Nov 1990 | WO |
| WO 9200567 | Jan 1992 | WO |
| WO 9220202 | Nov 1992 | WO |
| WO 0160236 | Aug 2001 | WO |
| WO 0213907 | Feb 2002 | WO |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 10324227 | Dec 2002 | US |
| Child | 11891505 | US |
