How the world’s first open source MRI happened

You wouldn’t think that a paragliding accident could start a revolution in medicine. But when a teenager fell 150 feet into a lake several years ago, the subsequent events that sparked a revolutionary new diagnostic method, what I am calling the first open source MRI. The story is a fascinating look at what one family can accomplish if they keep asking questions and getting the right people into a room to talk to each other.

MRIs are big proprietary machines made by a dozen large equipment manufacturers (such as Siemens, Hitachi and Toshiba) and cost plenty: think of them as what IBM mainframes of the 1980s were when PCs first started entering the corporate ecosystem. You have to train people how to operate them and interpret their scans, and once they are setup, you have to run them more or less continuously to realize the return on your investment. If your hospital or imaging center buys equipment from one vendor, you are pretty much locked into that vendor’s line of accessories and upgrades. And they are designed for very specific diagnostic situations. One of them that they aren’t very good at is looking at spinal cord abnormalities, because of the mass of tissues and bones that block a clear view.

This is what Spencer Stein faced when he hit that lake after his paragliding accident. His fall ended up giving him four spinal fractures, which he found after a week of living with back pain and going to the doctor. That was to be expected, given the height of his fall: he was lucky to be mobile. But what the doctors also found was a cavernous angioma, a series of blood clots inside his spine that he had since birth. The defect wasn’t related to the fall, just something that had been with him all his life, silent and symptom-free. The clots could stay that way, or they could bleed out and paralyze him instantly from the chest down.

Spencer had a tough choice; “I faced a dilemma. Leaving my defect alone, without surgery, could mean permanent loss of all bodily sensation, hideous neurological pain, or loss of functioning and movement in the lower half of my body. Yet an operation on this ultra-important, all-but-inaccessible part of my body could cause any of the same problems. It seemed like a no-win situation.” He wrote this for an article in the local San Diego newspaper this past summer chronicling his adventures.

As Spencer says, the fall was really a gift to learn about his birth defect. Without his fall, it would have never been discovered. So should he get the operation, or leave it alone? It wasn’t an easy decision. First, he had to find a surgeon who did many of these operations: this is a very highly specialized field, and only a few people actually do more than a few spinal operations a year of this nature. Second, many of the doctors Spencer saw didn’t want to do high-risk surgery on someone so young and otherwise so healthy, and said they wouldn’t dare operate. Spencer’s parents left the decision up to him. Can you imagine having to make the call of opening up a spinal cord on an extreme athlete?

Spencer with his parents visited a few of the doctors who specialize in this six-hour procedure, and eventually found one who was willing to proceed with the surgery in Arizona. That was great. But to really understand the structure of the clot, this doctor wanted better, higher-resolution pictures.

Back to the MRIs. Most of the machines that are available to the general public use 1.5 to 3 Tesla magnets (this is a measurement of magnetic force, named after the scientist that did a lot of early work, not the contemporary electric car company). Stein had a series of these MRIs to see what was going on inside his spine. But these images only told part of the story. The way to get to higher res images is to use a bigger magnet, just like our camera phones have a bigger image sensor in them.

It so happens that there is a 7 Tesla magnet at NYU Medical Center, but it was used for medical research unrelated to spinal imaging. No one had ever thought to use this research magnet in this way before.

That’s where Spencer’s dad came into our story. His dad made a few phone calls, and knew some of the leading edge diagnosticians in medical imaging. He managed to bring together people from bio-imaging, neurosurgery, and other clinicians to collaborate on how to use this magnet to take the right kind of spinal pictures.

Spencer’s dad is Lee Stein, someone that I have known for decades for his work in tech. Lee was part of the initial digital payments efforts in the mid-1990s, before there was a PayPal, before there was a Square, when paying for something online using a credit card was new. Now we take these things for granted, but back then Lee and a group of Internet pioneers assembled an elegant way to pay for things using nothing more than email. The effort never took off, but the founders of the firm laid the groundwork for how we pay for things online today, and eventually sold their patent rights to eBay.

I have written about Lee before, in another effort that he has been working on for several years called Prize Capital. As a result of this work, Lee became familiar with the TED Med community. (You can watch his TED Talk here🙂

He is a very unassuming and modest man for someone who has led some pretty amazing technologic efforts over the years. “Spencer was no longer a patient, but a catalyst for forming a new center at NYU.” When I met up with Lee a few months ago, he told me that he just made a couple of phone calls. Yeah, right. The scan brought together many specialists who were able to cobble together this open source solution in real time, rather than years that it would have normally taken. As you can see in the TED video, the doctors are sitting together, hunched around the MRI monitor, sharing knowledge and ideas on the spot.

As a result of Spencer’s scan and his parents’ efforts, more than $22 million dollars was raised for this new center, including contributions from Siemens, the MRI vendor used at NYU. “It has been a remarkable experience for us as a family to see how NYU has come together to do this,” Lee says in his TED lecture.

The Steins brought together these very highly specialized doctors at NYU, and now they are figuring out ways to use the equipment for a variety of diagnoses of different diseases. It is quite thrilling.

However, the irony of all this is that the 7 Tesla images didn’t help Spencer’s situation. Turns out, the large blood clot concentrates iron molecules in one place, and the large amount of iron deflects the higher-intensity MRI beams so no additional information could be gleaned from these higher res pictures.

But Spencer went ahead with the operation, and fortunately it was successful. He is back in school at Cornell. The Stein family, being who they are, created a prize of $5,000 as part of an ongoing annual business plan competition held by the school.

Their prize is given for the business plan that best addressed issues of disability or environmental sustainability. The first one was given to an idea called EcoFishFeed that raises forager fish for feed at salmon aquaculture farms. The Stein Family Prize will continue to be a part of future Cornell business plan competitions.

It is a fitting testimonial to Spencer’s fall from the sky.

0 thoughts on “How the world’s first open source MRI happened

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.