Penn Center for Innovation

I-Corps Blog: Proton Imaging

Posted on July 17, 2015 — Kat Hinkel

Week 1 

We are developing a proton imaging system that improves proton therapy treatment accuracy and protects healthy tissue from damage. We reached out to potential end-users (e.g., researchers, oncologists, and medical physicists) in radiation oncology for initial thoughts on whether or not they’d value our idea.  We received feedback that we put into a customer validation survey that we can use for structured data collection. The survey allows the people we interview to provide information on customer demographics, value proposition, potential use cases, potential benefits, and potential problems. We can use the information we collect to build a case around the value of our product to the customers.


Week 2 

In our interviews, we received feedback that our idea would be key for treatment of areas with lots of toxicities (e.g., brain, head, neck), but that it would not be as useful in regions with low toxicity, particularly where higher intensity modulator radiation therapy is effective (e.g., breast, prostate).  Similarly, we learned that up to 1 cm of normal healthy tissue surrounding a tumor will receive a full dose, so protecting regions of high toxicity is important.  We learned that X-ray CT images are created with 1/1000 of a dose, so if we could use 1/1000 of a dose to create proton CT images, it may be comparable.


Week 3 

We learned that the imaging modality would probably not affect insurance billing because Penn receives one bulk fee for daily imaging.  We also learned that double scattering is for tumors that move (e.g., lung/upper GI) where pencil beam can cause under treatment, but that double scattering is falling out of favor and new systems are increasingly pencil beam dedicated.  We learned that oncologists would like the value translated into a clinical endpoint, for example, what is the potential dose to organs at risk that can be eliminated (e.g., reduce risk of heart events by x percentage).  We learned that it’s important to know where planning error is the largest (where steep dose gradients are needed), for example, in the cases of unresectable pancreas cancer causing dose problems to the duodenum, liver cancer in people with cirrhosis, and dose to the small bowel for anal/rectal cancer patients.


Week 4 

We learned that physicians would like a slide that shows how a physician could incorporate this idea into their daily workflow, along with example images.  We learned that there could be value if proton CT images could allow for the use of anterior beams because things like bowel gas exchange change on a daily basis and where protons would have once stopped, they could now pass through.  We also learned that there may be value in having daily proton CT images during the treatment phase, but the challenge is that extracting an action from these images because there’s a cost/benefit to changing the treatment plan daily as the tumor shrinks, and may be unrealistic.


Week 5

We learned that there are applications in breast cancer, particularly with left post-mastectomy when interior mammary lymph nodes and the heart may be affected.  We learned that it will be important to interview the influencers and recommenders between our proposed end users (oncologists and medical physicists) and buyers (system manufacturers), such as the operations managers, chief financial officers, and insurance billers for our health system and others.  We also learned that there is value in knowing just where protons go within the body, because nobody knows with much accuracy, so we reached out to other people who may find value in knowing just where protons go outside of medical physics in case there are external applications.

Category: i-corps, i-corps-team-posts