As is always the case at WIMP, no stone was left unturned. Across a total of 35 presentations, speakers covered everything from motion management to the radiobiology of the Bragg peak, from disease-activated radiation therapy to ways to inspire the next generation of physicists and engineers. Uniquely, the conference offers another big draw: extensive midday breaks that allow participants to share their enthusiasm for winter activities in the beautiful Rocky Mountains.
Breast imaging
One highlight of WIMP 2010 was an afternoon session on breast X-ray imaging. Opening with a talk on "Physics and the human body" from Andrew Bukovitz of Radiological Physics Service, subsequent speakers focused on the future of breast X-ray imaging.
Loren Niklason, director of Hologic's breast tomosynthesis clinical trials and first author on the seminal paper on breast tomosynthesis (Radiology 205 399), presented a comprehensive overview on the principles of digital breast tomosynthesis (DBT). Hologic markets a DBT unit in Europe and has several installations. While not commercially available in the US, DBT units are installed at a number of research institutions and clinical trials are underway comparing digital mammography and DBT.
Niklason discussed the experiences of Hologic's users in Europe and the ongoing clinical trials in the US. He presented a number of 2D and DBT comparisons in which cancers could not be seen or were difficult to see using conventional mammography, but were readily visible in the DBT images.
Next up, Tony Siebert, professor of radiology at the University of California, Davis, and president elect of the AAPM, gave the presentation "Breast tomosynthesis acceptance testing". Siebert – who has contributed extensively to the digital X-ray imaging scientific literature – reviewed his experience in acceptance testing a Hologic DBT unit. In addition to the basic tests performed on a digital mammography unit, Siebert conducted a number of additional tests: slice thickness, x-y spatial resolution, and checking that the tomographic movement of the X-ray source and image receptor are diametric and aligned with the detector rows.
The keynote speaker at this session was John Boone, professor of radiology at the University of California, Davis, who discussed "The future of breast X-ray imaging: cone beam CT or tomosynthesis?". For the past several years, Boone has been developing and building cone-beam breast CT (CBCT) scanners and working with clinical colleagues to assess the efficacy of the technology.
An engaging speaker, Boone gave a fascinating overview of CBCT, the evolution of the scanners that the UC Davis team has constructed, and preliminary clinical results. He presented a number of cases in which cancers that could not be seen with digital mammography were readily evident on CBCT images.
Boone and colleagues are in the unique position of being able to compare digital mammography, DBT and CBCT. While Boone implied that it was premature at this time to postulate whether DBT or CBCT may be the future of breast X-ray imaging, he discussed the differences between the two.
• DBT has better x-y spatial resolution than CBCT and fuzzier z or slice-thickness resolution;
• DBT is currently performed at lower X-ray tube potentials than CBCT and can image microcalcifications better;
• CBCT has the advantage of virtually isotropic spatial resolution allowing users (and CAD algorithms) to compare three different orthogonal slices of suspicious areas, which will likely result in increased sensitivity and specificity.
Questions alluded to but not addressed in detail included dose comparisons of the two technologies and mammography (doses are most likely higher with DBT and CBCT than with mammography), as well as practical considerations such as cost, patient throughput and reading time. The overall impression was that both DBT and CBCT are much better mouse traps than digital mammography: the superposition of over and underlying structures is removed and, as with CT, the images are easier to read.
Expert advice
This year's WIMP also included a new type of session, comprising a keynote presentation followed by a moderated panel discussion. The topic under discussion was "Entrepreneurship in medical physics and beyond."
The session was moderated by Richard Stark, founder of Argus Software and current executive vice-president of ViewRay, developer of an MRI-guided radiotherapy system. The keynote speaker was serial entrepreneur Mark Gehring, co-founder of Geometrics (Pinnacle treatment planning), UltraVisual Medical Systems and Sharendipity.
The panel, meanwhile, was filled out with Josh Levy, president of The Phantom Laboratory; Ken Brooks, vice-president of new technology development at Hologic; and Gary Barnes, a University of Alabama professor and founder of X-Ray Imaging Innovations.
Entrepreneurship in medicine can take many forms, and the symposium saw physicists discuss their experiences and propose ways in which they might capitalize on their intellect and experience to further the field of imaging and cancer therapy. Many questions were raised and discussed: What does it take to start a new company? Should I build a business or license my intellectual property? What is my appetite for risk? How should I finance my new venture? What is my exit strategy? How can I assure success?
Of course, there is no one answer. There are, however, a few things that the panel has learned through experience. Several speakers expressed the importance of understanding the true unmet need that one seeks to fulfill, and the need to estimate the market potential for the idea. They emphasized that financing must scale with market potential, otherwise it is unlikely that investors will earn a return.
Licensing intellectual property was an appealing idea to many in the audience because it potentially allows the inventor to see an idea bear fruit while remaining in academia or clinical practice. Questions were answered regarding how to secure the intellectual property legally and how to approach potential buyers.
Levy discussed his approach to building a product company that focuses on conservative growth with lowered risk taking. The key to his success: applying a personal approach to designing products simply by listening to customers.
Gehring and Stark spoke about raising larger amounts of capital for ideas with big market potential, as well as the risks and rewards of such a strategy. One key to this approach is avoiding the "Valley of Death" that many companies fall into. It's not enough to have a great idea and a successful R&D effort. One must also successfully commercialize the technology. Typically people expect success when early adopters buy. However, there's still the necessity of "Crossing the Chasm", as Geoffrey Moore puts it, to get the majority of buyers to acquire.
The response to this mini symposium was positive, with the audience posing many more questions than there was time to answer, even after running significantly over schedule. Thoughts are now being given to the creation of an entrepreneurship workshop for medical physicists, to bring together inventors, entrepreneurs, investors and strategic partners to discuss ways in which commercialization of new ideas can be accelerated.
• WIMP is endorsed by the American Association of Physicists in Medicine and provides continuing education credits. Planning for WIMP 2011 has already started. For more information visit www.GoWIMP.org in the coming months, or contact Marc Kessler.