Editor’s Note: This is the first in a series of interviews with Drs. Jeff Saffer and Vicki Burnett, founders of Quertle LLC.
Janet Delicata: Jeff and Vicki. Congratulations on the new release. This is huge. It has been reported that the foundation of this release is Artificial Intelligence Technology. Hence your trademarked term BioAI™. Why is bio-specific artificial intelligence required?
Jeff Saffer: There are many artificial intelligence methods that have been developed, but they are generic. Each problem to be solved really requires the appropriate tool for that job. You just can’t say, “I’m going to use a screwdriver.” You have to know if you need a flat head, a Philips head, etc.
AI is no different. Biomedical and biological literature has unique challenges. There is specialized terminology, some of which like gene and chemical names can be easily confused with other terms. And, the general way physicians and biomedical researchers write is different than, say news and social media, for which many AI methods have been developed.
VB: (Laughing) Maybe not the most elegant analogy, but, yeah, we needed and have created a bio-screwdriver of sorts.
JD: Why is investigating the biomedical literature a big data problem and what is the role of visual analytics?
JS: The biomedical and biological literature encompasses more than 40 million authoritative documents – journal articles, patents, treatment protocols, and much more – containing billions of informational nuggets. Many traditional searches lead to 10,000 or more documents and even with a team of people, it is not feasible to go through those results effectively. So, like general web searches, you keep narrowing down your query until the results on the first page look interesting. Meanwhile, you may have just eliminated the most important document of the bunch. Artificial Intelligence overcomes this kind of problem.
VB: This is where predictive visual analytics really shines. The AI-based visualizations that are part of our platform enable the user to hone in on critical information quickly, uncover hidden connections, and predict what areas are becoming more important. You simply cannot get that information from a list.
JD: This is a really exciting advancement. Thank you for taking the time to discuss your BioAI™ platform with me.