As we’ve begun the implementation phase of our fellowship, the KnightCap team has begun clinical efficacy studies at MedStar Washington Hospital Center and National Rehabilitation Hospital. It was an exhilarating experience handing over our device to the first stroke patient study subject, and has made me reflect a lot on how we got here.
One of my favorite parts about developing KnightCap is that it’s a product built out of free thinking and free sharing of ideas. We couldn’t find an EEG headset made for sleep with the channels we needed that didn’t require the user to apply gel to the electrodes every night—or a bunch of other cumbersome steps for use. So, we set out on a journey to build our own EEG hardware, incorporated into a sleep mask, with the necessary electrodes for stroke monitoring.
KnightCap is built on open-source hardware through OpenBCI, which made “an open-source, low-cost EEG system for makers to lower the barrier of entry for brain-computer interface enthusiasts... with a fully programmable Arduino on board.” Working to build our system with an OpenBCI board has connected us with a community of makers in this space. On OpenBCI forums, people have found clever hacks to clipping electrodes to lead wires and we, in turn, have both posted and responded to questions about using OpenBCI in ways unique to our work.
We’ve assimilated into this culture of tinkering and sharing: 3D printing a lot of our parts, frequenting Github for useful code, and learning from different APIs (application programming interface). I think this has been a large part of how we’ve been able to pivot so quickly and iterate through different sensorization approaches for our product.
I used to think that competition would always drive innovation, but we’re seeing now that more collaborative, open products are the ones succeeding in the current market. Companies like Twitter and Google have released more and more open APIs, revealing part of a platform’s underlying code, for independent developers to create more horizontal growth for the company. The success of open, shared innovation diminishes the need to preserve competitive advantage, which can no longer compete with the inefficiencies brought on by intellectual property protection and closed R&D labs. Steven Johnson puts it best in his book Where Good Ideas Come From:
“Like any complex social reality, creating innovation environments is a matter of trade-offs. All other things being equal, financial incentives will indeed spur innovation. The problem is, all other things are never equal. When you introduce financial rewards into a system, barricades and secrecy emerge, making it harder for the open pattern of innovations to work their magic.”
Additionally, KnightCap’s technology would not have been possible without decades of clinical and academic research on EEG detection of stroke. When KnightCap was in its ideation phase, we reached out to a few different professors who had done some of this prominent EEG research. The responses we received were humbling. They took the time to excitedly explain their algorithms to us, and how we might adjust them for our intended purposes. They even offered to share their code and stroke EEG data they had collected during their studies. This sharing of information was enormously helpful for us during the ideation phase, to not only learn from their EEG analysis paradigms but also to evaluate the potential of KnightCap implementation.
Though academia is often talked about as an “ivory tower,” a lot of advancements in science and technology—especially in the healthcare space—arise from university labs. That’s why it’s so important that science findings shouldn’t always be so protected. Often, we find that multiple labs will work on a similar project, and try to beat each other to publication. Or, some research teams spend substantial funding collecting data that someone else had already collected.
Philosopher Michael Polanyi proposed, in 1962, an idealistic Republic of Science, where scientists could participate and operate almost like a political body with “coordinated mutual adjustment of independent initiatives” and freely exchange ideas. Polanyi paints an image in which every scientist has a piece of the jigsaw puzzle, but you have to be able to watch what everyone else is putting down and assembling together to know where to place your piece. As a community of scientists in a particular field collectively creates meaningful advancements for humankind, more real-time collaboration can make that process more efficient, and more publicly accessible dissemination of that knowledge can spur even more user innovation.
I’m not so naïve as to ignore the importance of intellectual property to solidify the value proposition of an innovation; non-disclosure agreements to protect years and years of hard work; or the competition between scientists to secure grants and publication in prestigious journals. All of these may be possibilities for KnightCap.
However, I would be remiss if we did not acknowledge the community of sharers—both hackers and scientists— who have helped make our development possible. Just as importantly, we would be remiss not to give back and share our successes and failures along the way.