Office Hours with… Elias Jaffa

For the past few years, Elias Jaffa worked at a community hospital in rural South Carolina. The job, he says, expanded his perspective on medicine.

Jaffa, an expert in point-of-care ultrasound who recently joined Yale’s Department of Emergency Medicine, is also interested in technology innovation. A self-described “hacker,” he enjoys finding new applications for old technologies to improve medical care and has even developed artificial intelligence models.

We recently caught up with Jaffa for the latest edition of Office Hours, a Q&A series that introduces Yale newcomers to the broader university community.

What role does ultrasound play in emergency medicine?

Elias Jaffa: Over the last 20 years or so, the scope of what’s possible with ultrasound has expanded rapidly and it has been recognized as having much broader applicability. A lot of that has to do with improvements in technology, the shrinking of the machines, and rapidly falling price points.

It is also a very low-risk intervention. You can learn so much about the human body just spending five minutes at a patient’s bedside with an ultrasound probe in your hand. And for someone like myself who is taking care of a wide breadth of problems — oftentimes several in the same patient — being able to rapidly get an understanding of somebody’s internal anatomy, physiology, and pathology is really useful.

You describe yourself as a hacker. What do you mean by that?

Jaffa: I mean it in the traditional sense of manipulating objects to make them do something they weren’t initially designed to do. For example, Raspberry Pi systems are these little, credit card-sized computers that are very low cost. I’ve used them to hijack the video output of older ultrasound systems and convert it into a digital stream that can be viewed live by anyone, anywhere in the world.

Colleagues I’d worked with in Tanzania started calling me to consult, showing me videos on their phones of ultrasound screens, but I had a very difficult time seeing what was going on. So that’s where the idea for this open source project, which I’ve been calling the Sonostreamer, came from — trying to get high-quality source images from ultrasound machines in order to facilitate long-distance, real-time feedback and education.

Does this work play into your research interests?

Jaffa: Yes, so in the last few years, I got involved in software development and have developed a couple of apps now and I even started developing some AI models. Bringing those and projects like Sonostreamer to bear on academic research is what I’m currently exploring. And I’m looking forward to digging into more AI research.

What are your interests outside of work?

Jaffa: I have three kids, five and under, so that’s my interest. I don’t think med school explicitly prepared me to try to explain cardiac physiology or nerve blocks to a four-year-old, but my wife and I are both doctors, so here we are.

Growing up, you had what you call an “extreme middle-child experience.”

Jaffa: Yes, my older brothers are identical twins, which was a surprise at the time since this was before ultrasound was used routinely. When my mother was giving birth, the doctor was like, “I don’t know how to tell you this, but there’s another one in there. Keep pushing.” I’m 14 months younger than them, and then six years later my parents finally got a daughter, but also got her twin brother in the deal. So I’m kind of the ultimate middle child.