July 2008 Archives

Tom Dellinger, medical physicist at Pardee Hospital (Hendersonville, NC) had something to smile about today. He was the lucky winner of medicalphysicsweb's prize draw, and will be heading home from AAPM in possession of a brand new iPod nano (in medicalphysicsweb blue, of course).

Smile: Tom Dellinger at the IOP Publishing booth

And we'll be heading home soon too, swapping the Houston heat for Bristol drizzle. I hope that you've enjoyed reading this blog, and be sure to check out the main site for AAPM follow-up coverage in the next week or two.

I've spent a busy few days pacing the AAPM exhibition show floor. There's just not room here to include all the new products that I've seen, but here are a few more releases that caught my eye:

• Modus Medical Devices of Ontario, Canada unveiled a neat system for turning a standard Catphan phantom into a breathing phantom. Its QUASAR Catphan Shaker can be programmed with patient-specific breathing data (from a Varian RPM) to simulate respiratory motion for quality assurance before treatment. 

The Shaker comes with software for displaying, editing and running motion waveforms

According to the data sheet, the Shaker offers an amplitude of up to 30mm for RPM input; or it can move sinusoidally with speeds from four to 60 breaths per minute at 40 mm amplitude. 

• Standard Imaging (Middleton, WI) had a whole host of new products to promote. First up, MIMI - a new phantom for testing the isocentricity of imaging modalities. The MIMI (multiple imaging modality isocentricity) platform can be used for QA of optical positioning systems, kV cone-beam CT, on-board imaging and laser systems.

Next up, new versions of the company's IMSure QA and PIPSpro software packages. IMSure version 3.1 offers features including Cyberknife verification, support for .decimal compensators and the ability to import DICOM structure sets. Version 4.2 of PIPSpro comes with automated cone-beam CT testing and improved stereotactic accuracy.

Standard Imaging also unveiled its new SuperMAX - a two-channel electrometer with a colour touch-screen interface - as well as an upgraded version of the MAX 4000 electrometer and three new inserts for its Lucy phantom.

• The big push at the Siemens' booth was ARTISTE, a linac for adaptive radiotherapy. According to Siemens, the system's key selling point is its "ultimate flexibility", which gives clinicians control over how to treat each patient each day.

The ARTISTEs were all tied up, so Siemens brought a big photo instead

ARTISTE features a 160-leaf MLC, claimed to be the fastest on the market with a leaf movement speed of 4 cm/s. "The more leaves you have and the finer resolution, the better you can conform to the tumour and the better you can deliver the dose," Siemens' product manager Brett Evans told me.

Using this collimator in tandem with the company's new IM-Confident Plan treatment planning system enables IMRT treatment time to be reduced to less than five minutes, he explained.

One company that caught my eye on the show floor was Aktina Medical, a Congers, NY-based manufacturer of radiation oncology accessories and treatment systems. Aktina was demonstrating a scheme for performing frameless radiosurgery. The company's pin.point localization and fixation system boasts sub-millimetre positioning accuracy for cranial and head-and-neck treatments.

"Previously, people didn't want to move away from using frames because nothing else could provide the same accuracy," Nick Zacharopoulos, Aktina's vice president, told me. "But we can do just as well as an invasive frame."

Pin.point works by using a moulded mouthpiece, vacuum attached to the roof of the patient's mouth and fixed within a Perspex frame. This frame contains metal rods that can be visualized via CT imaging, thus providing accurate information as to the patient's position. The company also offers a software package for autoregistration.

The pin.point system showing the mouthpiece (blue) within the perspex frame

"The therapist can determine the accuracy based on the vacuum pressure," Zacharopoulos explained. "If you see a good vacuum, you know the patient is set-up with sub-millimetre accuracy."

Aktina expects FDA clearance for pin.point early next month.

Researchers at the University of Florida (Gainesville, FL) have come up with a new take on proton-therapy dose verification: imaging gamma ray emission during the treatment process. The distribution of gamma rays - produced when high-energy protons interact with nuclei in the patient's tissue - provides a measure of the location and amount of energy delivered during treatment.

"If we have a way to image gamma rays, we can almost do real-time dose verification," said researcher Yuxin Feng, speaking at an AAPM session entitled Innovative frontiers in medical physics.

To detect the gamma rays, Feng and colleagues designed a Compton camera based on LaBr₃, a new scintillation material with high stopping power and good energy resolution. They used pixellated LaBr₃ crystals for both the camera's scattering and absorbing detectors.

The crystal modules in the front (scattering) layer are 2.5x2.5x0.5 cm in size, segmented into 4x4x2 mm on the surface to enhance spatial resolution. The detectors in the rear layer are 2.5x2.5x3.0 cm, segmented into 4x4x10 mm.

"The prototype Compton camera is able to image gamma rays from 0.5 to 2 MeV," said Feng. He noted that this technique could also be used to monitor high-energy photon-based therapies.

The proposed Compton camera offers a potential angular resolution of about 0.3 radians when imaging 511 keV gamma rays, and 0.05 radians or less for gamma rays above 2 MeV. Bench-top tests demonstrated a spatial resolution of 5 mm (at 662 keV) - sufficient to detect the distal fall-off of a proton beam.

Feng says that the Florida team now plans to test the prototype camera in clinical radiotherapy facilities, using both proton and photon beams.

Also in the pipeline: new configurations, such as dense scattering cameras or multiple cameras with a perpendicular set-up; and acceleration of the camera's 3D image reconstruction speed. Ultimately, it's this fast image reconstruction that could make real-time imaging of proton therapy a reality.
 

Treatment planning appears to be one of the recurring themes at this year's AAPM, with BrainLAB of Germany among those highlighting the latest and greatest in radiotherapy planning software. On show at the company's booth were the latest releases of its iPlan packages: iPlan RT image, for contouring and target definition, and iPlan RT Dose, for treatment dose planning.

"From the imaging perspective, the version 4 software integrates a couple of new things: 4D CT integration and the head-and-neck Atlas," iPlan project manager David Brett told me.

The head-and-neck Atlas provides a nifty way to perform automatic segmentation of pre-treatment CT or MRI scans. The Atlas - a series of around 50 predefined anatomical objects - is fused onto the patient data set using elastic image fusion. And it's fast: iPlan takes less than a minute to find these 50 objects, after which the user can tweak the outlines manually, if required.

"There's a lack of consistency in contouring; studies of five doctors will give five different results," Brett said. "The benefit of Atlas-based segmentation is that within one minute you get a consistent outline. BrainLAB's software team works closely with clinical partners to get the ideal contours that fit the greatest percentage of patients".

On the dose planning side, BrainLAB's big announcement is the FDA clearance of its Monte Carlo dose calculation algorithm - an integral part of iPlan RT Dose version 4.

David Brett extols the benefits of BrainLAB's Monte Carlo dose engine

Brett explained that although Monte Carlo algorithms have been used before for treatment planning, clinical implementation tended to be limited by extreme calculation time requirements, or restrictions to planning for a single machine or treatment modality.

"Monte Carlo is considered the gold standard, especially for extra-cranial treatments, because it considers inhomogeneities," he said." With our Monte Carlo algorithm, we can create a complex IMRT case in less than five minutes."

Such a feat, which Brett claims would previously have taken 20 computers running over night to perform, was enabled by designing a Monte Carlo algorithm specifically for radiotherapy needs. What's more, BrainLAB's algorithm supports a multitude of treatment plans, including conformal beam, static and dynamic arc, and IMRT.

"We believe that the release of the BrainLAB Monte Carlo dose calculation algorithm will have a great impact on the field of radiosurgery," said Brett. "We continue to develop software that extends clinicians' treatment precision to cover more extra-cranial cancers, as well as advanced dose calculation possibilities for lung and head-and-neck tumours."

5.30 AM on a humid morning, a few hundred physicists gathered on dark street corners waiting to head over to Houston's Memorial Park for AAPM's annual 5k fun run, sponsored by Gammex. Sensibly waiting until the sun actually rose, the race kicked off an hour later, with around 175 brave souls lining up at the start.

And they're off: the starting line

The winner - with a time of 17 minutes 35 seconds - was Yevgeney Vinogradskiy, a student at MD Anderson Cancer Center and a first-time runner of this race. "It was a great course and well organized," he told me at the finishing line. "And it's always good to come first."

Simon and Zoe just ran 5k.

As for our own runners, PMB publisher Simon Harris came in at an impressive 22:38, while medicalphysicsweb's Zoe Anderson - a first time runner in this (or actually any) race, crossed the line at 38:10. Well done to you both. As for me… I just watched - well somebody had to take the photos.

The idea of measuring tissue stiffness using ultrasound is nothing new - articles on this technique appeared in the literature back in the early 1980s. But according to Jonathan Ophir, from the University of Texas Medical School (Houston, TX), the area has progressed at pace of late.

"Real-time, hand-held elastography is now a commercial reality, with several companies offering it," Ophir told AAPM delegates at a symposium on ultrasound imaging. "The clinical applications are expected to accelerate."

So why study tissue elasticity? "Many pathological changes are associated with changes in tissue stiffness," Ophir explained. And with some isoechoic lesions (HIFU-ablated tissue, for example, or some prostate cancers) stiffer than surrounding tissue, "you can see things that you don't see otherwise".

But it's not just a case of finding lesions that can't be detected with standard ultrasound. Ophir presented in-vivo sonograms and elastograms, both showing the same breast-cancer lesion. While the images looked similar, the lesion appeared obviously larger in the elastogram. He explained that this discrepancy appears to be a trend, with three or four other research groups also noting the same effect.

"Nobody has proved this yet, but we think it's due to scar formation around cancers that's not typical for benign disease," Ophir suggested. "This could be used as a sign for cancer."

Looking forward, elastography could provide several additional ways to glean more information relating to the mechanics of tissues. One possibility is offered by imaging shear strains at tissue boundaries.

Ophir showed an example comparing the shear strains of breast carcinoma and fibroadenoma. In the benign case, the areas showing the greatest shear strain were tight up against the lesion/tissue boundary, as viewed in a standard ultrasound image.

In the carcinoma, however, these areas appeared much further from the boundary, possibly due to the aforementioned 'scarring' effect. The high-shear-strain areas were also much larger in this case, assigned to the fact that cancers tend to be more strongly fixed between the surrounding tissue layers. "Shear strain may be able to tell us if a lesion is benign or malignant and where the lesion boundaries are," Ophir explained.

He also cited imaging of the Poisson's ratio of a tissue, and its evolution in time, as another future development. Poisson's ratio can be calculated by measuring both the lateral and axial strains, and can be used to study the behaviour of poroelastic (an elastic matrix containing fluid) tissue and provide information on fluid flow in tissues affected by diseases such as lymphoedema.

As the thermometer hit 100 F last night, the intrepid medicalphysicsweb team ventured out into the Houston heat. The reason for this madness? We were off to check out some of the corporate parties, of course, packing in visits to Elekta's Beer Bash and TomoTherapy's drinks reception.

Zoe, Tami, Rose and Paul brave TomoTherapy's helical storms

Thanks goes to both companies for an entertaining evening. But TomoTherapy has to win the prize for the best cocktail of the week (so far) with its aptly named "helical storm" - a lurid blue concoction complete with a helping of dry ice to whip up that aforementioned storm.

Next stop in my quickfire tour of the 131 exhibitors showcasing their wares at AAPM was Philips. Among the company's line up of new developments was the latest version of its radiation treatment planning system Pinnacle³.

New features of the work-in-progress Version 9 (due for commercial release in Q2 next year) include: faster planning speed; enhanced handling of 4D CT image sets; streamlined image import (one click to import multiple image sets); and improved DICOM export.

"We've made a lot of workflow improvements," said Philips' Kathy Wall. "What used to take a lot of clicks now takes just one or two."

Head-and-neck planning with SmartArc

Keeping track with the latest IMRT developments, in particular, volumetric modulated arc therapy (VMAT), Pinnacle³ now also offers a feature called SmartArc. SmartArc is a treatment planning system that's designed to work with all VMAT compatible accelerators. It supports both constant and variable dose rate delivery, and can deal with MLC leaves with 2.5 or 5 mm resolution.

According to Philips, a new dose engine - along with the use of more powerful computing hardware - enables SmartArc to meet the speed requirements of VMAT's more complex treatment plans. And as SmartArc is based on the company's existing IMRT planning software, Wall says that users will easily learn how to use the new system.

Philips was also announcing the forthcoming release of its Gemini TF Big Bore - claimed to be the first commercial PET/CT that offers a bore size of 85 cm for both PET and CT. The new system will also feature a 190 cm table that's capable of supporting up to 500 lb. The company plans to start installation at beta sites at the end of this year and ship the Big Bore commercially next April.
 

If you've read the latest Talking Point article on medicalphysicsweb, you'll have heard how - according to some - MR guidance during radiotherapy could be one of the next big developments in radiation treatment.

Among those currently working on this challenge is Bas Raaymakers of the University Medical Center Utrecht in the Netherlands. I went along to the poster session to talk to Raaymakers about the project's progress.

The Utrecht team is developing a 6 MV radiotherapy accelerator with an integrated 1.5T MRI functionality. The ultimate goal: real-time, soft-tissue-based image guidance during delivery.

Bas Raaymakers at the AAPM poster session

"This started as a system for better position verification, but on the way we realised that if you have this high soft-tissue contrast you can go a step further," Raaymakers told me. He explained that MR guidance during treatment would enable better avoidance of critical structures, and that using such a system to deliver tumour ablation and then treat microscopic infiltrations could "potentially compete with surgery".

According to Raaymakers, it's the design of the MR magnet that's critical to this system working. It's adjusted to have an area of low magnetic field close to the magnet itself. Placing the linac in this low-field region then minimizes the magnetic interference. In addition, all of the magnet components are moved to the edges, leaving an 'empty' area through which the beam can travel with minimal absorption.

The magnet is arriving in October of this year, and a prototype system should be operational in January 2009. The initial prototype will be static with the linac mounted in a fixed position next to the magnet. Ultimately, the system will have the accelerator rotating around the MRI.

"We plan to start simple, and see if we can improve position verification and reduce margins," Raaymakers told me. "In the end, we should be able to image continuously during treatment and track the tumour motion with dynamic MRI."
 

As the AAPM exhibition opened its doors, my first stop was TomoTherapy's booth, where the Madison, WI-based equipment maker was pushing its newest offering - the Tomo Quality Assurance (TQA) tool. TQA is a quality-assurance system that the company has developed specifically for use with its Hi.Art radiotherapy system.

In simple terms, TQA works by automatically gathering data from Hi.Art's integrated sensors and monitoring subsystems, and then making these data easily accessible to the medical physicist. According to TomoTherapy's press release, the application not only streamlines data collection but also simplifies the data review process.

TomoTherapy promotes its new TQA

"In eight minutes, you get enhanced daily QA - much more than is possible using external equipment," Bob Cravens, lead applied physicist and TQA architect, told the gathered crowd. He noted that clinicians already using a TQA prototype tend to rely on TQA more and more, while using their own calibration equipment far less.

The initial TQA release – available as of today – comes with up to seven modules, including basic dosimetry and system monitor modules. A static step wedge module uses a stepped phantom to measure things such as output, energy, couch speed, profiles, field width and laser offsets. There's also a helical step wedge module that provides information on gantry position, a longitudinal beam profile package, and two modules that examine linac alignment. More options are in the pipeline.

"Hi-Art was the enabler for the oncologists and TQA will be the enabler for the medical physicists," said Cravens.

Whilst in Houston, don't forget to pick up your free copy of the medicalphysicsweb review, a print supplement we've put together for conference delegates and exhibitors at the AAPM meeting. We'll be handing out the review in and around the exhibition hall, or you can head over to booth #1241 to pick up your copy there.

If you can't make it over to the AAPM meeting, don't worry, you can always download the PDF instead. Simply click on this link to access the Summer issue of the medicalphysicsweb review: MPWreview.pdf

But that's not all. We're also running a fantastic prize draw at AAPM, open to anyone signing up as a new medicalphysicsweb member. For a chance to get your hands on an iPod nano, see page 5 of the review for details or ask our staff at booth #1241. We'll be announcing the winner on Wednesday afternoon at the show.

On Sunday afternoon, head over to the convention centre to check out the AAPM's annual Young Investigator competition. Ten finalists - chosen by the abstract reviewers - will present their research at a special symposium, held in honour of University of Wisconsin Professor Emeritus John R. Cameron.

This year sees presentation themes ranging from a study of proton therapy collimators made from carbon foam, a material that should  produce fewer neutrons and be more affordable to manufacture, to research on low-dose-rate prostate brachytherapy with intra-operative post-implant dose evaluation.

The Young Investigators Symposium will be held from 4.00 until 6.00 in Auditorium C of the George R. Brown Convention Center. Following the symposium, the best three presentations will be chosen and the researchers recognized during the AAPM Awards and Honors Ceremony on Monday evening. The full line up is as follows:

• Empirical investigation of flat-panel imagers incorporating high-efficiency, segmented BGO and CsI:Tl detectors for radiotherapy imaging Y Wang et al.
• Carbon Foam Collimators for Proton Therapy Y Charara et al.
• Machine Learning Tools for Predicting Clinical Complications in a Multi-Plan IMRT framework H Zhang et al.
• Improved Lesion Conspicuity in Dual-Energy Imaging of the Chest: From NEQ to Observer Performance S Richard and J Siewerdsen
• Low dose rate prostate brachytherapy: a tomosynthesis-based intra-operative post-implant dose evaluation M Brunet-Benkhoucha et al.
• A Novel Approach to Multi-Parametric Dynamic Chemical Shift Imaging B Taylor et al.
• Classification of Breast Carcinoma Subtypes using Computer-Extracted Morphological and Kinetic Features in DCE-MRI N Bhooshan et al.
• Spatial Weighted Mutual Information for Image Registration in Image Guided Radiation Therapy S Park et al.
• Bioanatomic MR Imaging for Characterization of Brain Tumor and Radiation Response in the Rat Brain T Atwood et al.
• The imaging and dosimetric capabilities of a novel CT/MR-suitable, anatomically adaptive, shielded HDR/PDR intracavitary brachytherapy applicator for the treatment of cervical cancer M Price et al.

Welcome to The Main Event, medicalphysicsweb's news blog. There are just a few days to go now until we head over to Houston, Texas for the 50th annual meeting of the American Association of Physicists in Medicine (AAPM). Whether you're attending the conference or not, The Main Event will bring you the news and the views from the AAPM meeting as it happens.

I'll be reporting direct from Houston on the hot topics presented at the technical conference, as well as the new products and the latest technology innovations being showcased at the exhibition. If you have any announcements, must-see products or new clinical results to tell us about - or you just want to find out more about getting editorial coverage on medicalphysicsweb - catch up with us at the IOP Publishing booth (#1309).

If you have any questions about the journal Physics in Medicine & Biology, the publisher Simon Harris will be at AAPM too. Come and see Simon at booth #1309 to find out more about: how to submit a research paper, the service offered to authors, subscribing to receive the journal, or how to access papers online.

We look forward to meeting you in Houston.