Image Guidance and Localization Technologies for Radiotherapy


Introduction

A key component of safe and effective modern radiotherapy is the ability to align a patient with respect to the radiation beam during treatment. Current means of accomplishing this include 3D visualization of soft tissue structures (via computed tomography (CT) or ultrasound and soon magnetic resonance imaging (MRI)), 2D or 3D localization of bony landmarks or fiducials (planar kV/MV images or implanted marker tracking), and alignment of surrogates such as a bite block or reflective markers. The most extensive use of these techniques is for image-guided radiotherapy (IGRT), which uses imaging on a regular (sometimes daily) basis, typically with a volumetric modality. This chapter focuses on the technical aspects of these on-treatment localization approaches. Clinical motivations and considerations are considered further in Chapter 18 , which includes a comparison of the various advantages/disadvantages of various image guidance systems.

This chapter includes 1) an overview of technologies for imaging and localizing patients during treatment, 2) basic concepts underlying image quality, and 3) principles and practice of commissioning and quality assurance. While the primary literature on imaging and localization technologies is vast, this chapter draws from consensus documents including AAPM Task Groups, standards from the Canadian Association of Provincial Cancer Agencies (CAPCA), and other sources. A resource reference list appears in Table 7.1 in shorthand form. Appendix I has more complete details for each reference.

TABLE 7.1
Resource Documents for On-Treatment Imaging and Localization (See Appendix I for Full Citations)
Report # Date Description
AAPM TG-132 Pending Image registration and data fusion algorithms and techniques
AAPM TG-226 2014 Medical physics practice guideline (MPPG), QA of x-ray-based IGRT systems
AAPM TG-179 2012 QA of CT-based IGRT
AAPM TG-174 2012 QA of non-radiographic localization systems
AAPM TG-154 2011 Ultrasound for external beam radiation therapy (EBRT) prostate
AAPM TG-135 2011 QA for CyberKnife
AAPM TG-148 2010 QA for TomoTherapy
AAPM TG-104 2009 In-room kV imaging
AAPM TG-75 2007 Dose in IGRT
CAPCA QC: EPID 2005 Canadian Quality Control Standards for EPID
AAPM TG-58 2001 EPID imaging
1999 IPEM document, Report 81 (IPEM, 1999); EPIDs
ACR/ASTRO 2009 ACR-ASTRO practice guidelines on IGRT
ASTRO IGRT Safety 2013 Safety considerations for IGRT

Overview of Clinical Imaging Systems

The technology and use of on-treatment imaging systems continue to evolve at a fast pace. A literature search reveals 236 papers published in the Radiation Oncology literature with the phrase “cone-beam” in 2013 alone. Several of the consensus reports attempt to provide overviews, of which AAPM TG-104, The Role of In-Room kV X-Ray Imaging for Patient Setup and Target Localization is notable for its historical perspective dating back to the use of kV sources coupled to 60 Co treatment units in the late 1950s. Currently the most commonly used technologies include the following (see lists in AAPM TG-104/AAPM TG-179 ):

  • Kilovoltage cone-beam CT (kV-CBCT), e.g., On-Board Imaging (OBI) system (Varian Inc., Palo Alto, CA); Synergy X-ray Volume Imager (XVI) system (Elekta, Inc., Crawley, UK). Flat panel imaging device in fluoroscopic mode; 2 projection images per degree with 200° to 360° gantry rotation range; 1 to 2 minute acquisition. These units are also capable of radiographs and fluoroscopy. Reference document: AAPM TG-104 (Sec. II.C).

  • Megavoltage planar images (portal images) used in gantry-based systems. Reference documents: AAPM TG-58 (circa 2001) and CAPC QC Standards for EPIDs. Efforts are now in development to utilize MV imaging to verify delivery of dose either just before treatment or during treatment itself, an approach referred to as in-vivo EPID dosimetry .

  • Fan-beam MV-CT (TomoTherapy/Accuray, Madison, WI). CT using the 3.5 MeV beam, 4 mm collimated beam with pitch 1 to 3. Reference document: AAPM TG-148.

  • In-room CT on rails. Less widely used than the above technologies, although diagnostic-quality images are available. Reference document: AAPM TG-104 (Sec. II.A) .

  • Planar stereoscopic kV-imaging. CyberKnife system (Accuray Inc., Sunnyvale, CA) and Novalis ExacTrac system (BrainLab AG, Feldkirchen, Germany) use paired x-ray tube/flat panel imagers mounted in the floor and ceiling to determine shifts relative to DRRs. Dual-energy kV systems are now becoming commercially available, which enable subtraction imaging to enhance soft tissue visualization. Reference documents: See AAPM TG-104, AAPM TG-135.

  • “Combined systems” (i.e., stereoscopic planar plus kV-CBCT). NovalisTx platform (Varian Inc., Palo Alto, CA). Vero (Mitsubishi Heavy Industries, Ltd. & Brainlab AG), using the MHI-TM2000 accelerator and a pair of x-ray tubes/flat panel imagers mounted in a pivoting O-ring assembly. Reference document: AAPM TG-104.

  • Digital tomosynthesis. Lying somewhere between planar imaging and CT, this technique renders a “thick plane” perpendicular to the beam by acquiring images over a small range of angles (<40 degrees). Reference document: AAPM TG-104 and references therein.

In addition to the above systems, various “add-on” technologies are available for on-treatment localization. Ultrasound has been used to localize the prostate and surgical cavity in breast during treatment. A camera system establishes the geometric relationship between the ultrasound transducer probe and the treatment isocenter. Commercial products include SonArray (Varian Medical Systems, Palo Alto, CA) and Clarity (Resonent Medical, Montreal, Canada) marketed as “I-beam” (Elekta Inc., Crawley, UK). Challenges include interuser variability (and the associated importance of user training), abdominal pressure (displacements of 5 mm have been noted), and intrafraction motion (one study noted 5 mm displacements 3% of the time). Reference documents: AAPM TG-154 for external beam and AAPM TG-128 for ultrasound (US) in brachytherapy.

Several localization systems are available that rely on imaging the surface of the patient and/or superficial markers (i.e., nonradiographic localization systems). Reference document: AAPM TG-147, cf. Table 1, list of systems . One class of systems uses two or more stereoscopic cameras to image infrared reflective markers. These include the FreeTrack system (Varian Medical Systems, Palo Alto, CA), which aligns a bite block for intracranial treatments; ExacTrac (BrainLab AG, Feldkirchen, Germany); DynaTrac (Elekta Inc., Crawley, UK); and CyberKnife (Accuray Inc., Sunnyvale, CA), all of which use reflective markers in some modes of operation. Other systems rely on the projection of structured light patterns onto the surface of the patient (e.g., AlignRT, VisionRT Ltd., London, UK) or projected lasers (Sentinel, C-Rad, Uppsala, Sweden). One fundamental limitation of all such systems is their reliance on features on the surface of a patient, which may not correlate with the position of internal anatomy of interest.

A final class of nonradiographic localization relies on implanted electromagnetic transponders whose location can be determined through a source/receiver coil at or near the surface of the patient. One system, Calypso (marketed by Varian Medical Systems, Palo Alto, CA) uses 8 × 2 mm implanted transponder devices. It is currently FDA approved for prostate radiotherapy and its use in lung is under investigation. In the prostate, three beacons are typically implanted from which gland rotation/translation can be determined.

Other on-treatment imaging systems are under development but are not yet the subject of society-level reports. These include the MR- 60 Co treatment unit (ViewRay Inc., Bedford, OH) and MR-coupled linear accelerators. This is an area of active research on the image guidance aspects and also the effect of the magnetic field on radiation transport (e.g., electron-return effect ). Proton radiotherapy centers also employ on-treatment image guidance, which is briefly described in AAPM TG-104. Almost all proton radiotherapy centers currently rely on kV planar radiography and do not utilize on-treatment CT.

There are few studies that directly compare the IGRT technologies mentioned above and almost none that does this in a prospective manner. This may be an area of future investigation. An example study for prostate cancer is found in Mayyas et al., who evaluated ultrasound, kV planar images, CBCT, and implanted electromagnetic beacons in 27 patients prospectively. A benefit was shown to IGRT (relative to standard three-point setup) in terms of possible margin reduction. Though similar overall, some modality-dependent differences were noted in the different dimensions.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here