" Hi, my name is Jeff Jaskolka, and I am the Vascular Interventional Radiologist, Adjunct Professor at University of Toronto, and a board advisor to SapienSecure. In some ways, it feels like I have been interested in AI and neuroscience for my entire life. I never really had aspirations to make a career of it, but always found myself in university classes and med school electives somehow pulled towards computer science, neurology and, of course, AI. When this so called “third wave” of AI started to take off a few years ago, I could see immediately the kind of profound impact it was going to have on the world at large, including medicine and particularly as it related to Radiology. As a speciality, Radiology is heavily computer and image driven with massive repositories of data. I speculated early on a big disruption was coming, and I wanted to be part of it.
My initial attempts to understand AI came primarily from a business perspective, though wanting a deeper understanding of new concepts, I found that perspective too superficial, and felt to truly understand AI would require a technical knowledge. I attempted to take an online course, and quickly noticed I did not have the requisite math or coding knowledge, so I decided to take a python course. Soon after I took an intro to machine learning course, then an even deeper learning course.
You could say I caught the bug, and then took a specialization in Natural Language Processing (NLP). One day I looked up and astonished myself that I could train neural networks to do fun and interesting things.
I felt drawn to NLP in particular since I had already discovered how hard it was to pull non-imaging data from hospital information systems for research or operational improvements. I could see the potential of incredibly rich information locked away in the unstructured data of the electronic medical record systems, and radiology information systems, and I wanted to figure out a way to access it. As soon as I was introduced to SapienSecure it was as if someone had turned on the light bulb, and excited to find a group of people with similar vision and ideas. I joined the team almost immediately, and for the last year I have greatly enjoyed working directly with the founders, Dr. William Parker, Brian Lee and Dr. Savvas Nicholaus, on developing technology to easily and securely understand and extract information from health data to for doctors such as myself to ultimately improve patient care.
• Adjunct professor at University of Toronto
• Vascular Interventional Radiologist
• Research in AI/NLP of extraction from unstructured radiology reports and quality improvement initiatives