Startup Q&A - Vyasa Analytics Utilizes AI to Gain Insight Into the Life Science and Healthcare Fields
If you have seen one of the many Marvel Comics films, you’ve probably watched a few scenes of Tony Stark/Iron Man talking to his AI JARVIS, who seems to have all the answers regarding Stark’s technology.
In the non-superhero world, but similar to the fictional JARVIS, there a several startups developing AI platforms that have a certain level of knowledge in a particular subject.
Newburyport-based Vyasa Analytics has developed an AI platform called Cortex that is able to provide healthcare and life sciences workers with the findings and research they need. The startup has recently exited stealth mode and is ready to assist teams and individual workers in those two fields.
We spoke with the company’s CEO Dr. Christopher Bouton about the Cortex platform and how their software works. Bouton also went into detail on how AI is developing in general and how it’s affecting the world.
CB: I’m a big fan of the phrase “origin story.” What are the origins of Vyasa Analytics?
CB: My background is in the life sciences and data analytics space. My Ph.D. is in Molecular Neurobiology from Johns Hopkins University, where I conducted both molecular biological research as well as computational biology. I was the Head of Integrative Data Mining at Pfizer for five years before founding Entagen, a software company for big data analytics in 2008. In 2013, I sold Entagen to Thomson Reuters. I left Thomson Reuters in 2016 and wandered a bit, including building an art car for Burning Man and music festivals called ForestHouse.
In 2016, I founded Vyasa. Right now, we are experiencing a perfect storm of technology coming together in the AI space and I couldn't help but be extremely excited by what's possible with these new technologies in the life sciences and healthcare space.
CB: What is the ultimate goal of Vyasa Analytics?
CB: Our goal is always to build cutting-edge technologies that enable our clients to gain greater insight into life sciences and healthcare verticals.
CB: Explain what Vyasa Analytics does. If it’s a particular software/platform/service/etc. how does it work?
CB: Cortex is a new secure, highly scalable software platform for life sciences and healthcare organizations. Cortex enables project teams to analyze data using deep learning algorithms. These algorithms enable the software to identify patterns, relationships, and concept representations after being trained on what to look for. Cortex also provides a library of deep learning modules for a range of applications in the life sciences and healthcare verticals including chemical design, image processing, and predictive analytics.
Built into Cortex is something we call Neural Concept Recognition. To help describe Cortex, imagine a friend gives you a dictionary, and you can read it and learn about all of the words and concepts in it very rapidly. Your friend then asks you to read a library of books, sentence by sentence, and identify the patterns and relationships between the words and concepts you learned about in them. As you read the books, you’ll also identify novel examples of the kinds of things you've learned about (new companies, people, locations etc.). Finally, the friend asks you to use that knowledge to write a novel about three concepts and how they relate to each other.
You are Cortex, and your friend is a researcher. The researcher gives Cortex some set of information – like examples of drug-like compounds, diseases, and genes – and Cortex uses deep learning to read it in milliseconds. The researcher then asks Cortex to read a library of data – cell images, molecular structures, research reports – and identify the patterns and relationships between them. Finally, the researcher asks Cortex to use that knowledge to identify novel molecules with a specific toxicity.
CB: Cortex sounds incredibly sophisticated. How long was the development process on Cortex? What kind of tools did the engineers use to develop it?
CB: Since founding Vyasa, we have been in stealth mode building our technologies. Vyasa has 13 software engineers, UX/UI designers and data scientists on the team. Cortex is proprietary software built on highly scalable big data technologies combined with cutting-edge deep learning frameworks.
CB: How big is the team? Looking to hire any particular position in the upcoming months?
CB: Vyasa is headquartered in Newburyport, MA with 13 employees, but the team is primarily based in Boston. We are always in search of great data scientists, software engineers, and UX/UI designers.
CB: Which biotech/digital health industry is Vyasa Analytics looking to target? Who will be the end users?
CB: Vyasa Analytics targets the life sciences and healthcare industries. The end users of Cortex are typically project teams, comprised of a project lead, data scientists and a range of other individuals researching and making decisions on novel therapies, clinical trials, hospital efficiencies and/or other challenges.
Vyasa Cortex works in conjunction with Vyasa's core AI technology, which is called Vyasa Layar. Vyasa Layar provides API access to all of the technologies that are available in Cortex.
CB: Is the company bootstrapped or seeking investments?
CB: The company is privately funded with an initial round of $500,000. We are not currently seeking venture funding.
CB: I’m always interested in how a startup came up with its name. How did Vyasa Analytics get its name?
CB: I lived in India for four years as a boy, from age four to eight. While I lived there, I developed a great respect for the many belief systems and writings of Hindu and broader Indian culture. Vyasa is a highly-revered figure in Hindu dharma. As the key compiler and storyteller of sacred Hindu texts, Vyasa brought together knowledge from across many sources. I loved that idea and wanted to relate it to how AI systems can also help us compile information from disparate sources. Our data today has the ability to tell us important valuable stories, and novel technologies such as deep learning can help us unlock those stories.
CB: Any comments you’d like to make about AI?
CB: Despite the current hype cycle regarding AI, I believe that humanity is in fact just in the very beginning stages of learning how to apply these technologies. I use the analogy of the first time that you saw Netscape Navigator and you thought to yourself, "Huh, this is interesting. I can click on links and seepages. I wonder what this will be useful for?" There is incredible untapped potential with these deep learning technologies, and I believe that they will be powerful tools for us to use to improve the lives of millions. I also believe that they will enable further employment for people who learn how to use these systems, instead of replacing jobs. Our jobs as humans will move up the value chain, bolstered by the capabilities of AI systems to handle more of the mundane tasks we now do.
Another analogy I often use is that of navigation systems in our cars. Navigation systems help us move through a complex information space to enable us to get to an end goal faster and more efficiently with the ability to update and re-route along the way. AI systems will enable these same types of capabilities through larger and larger information spaces, but it will remain the human that envisions and achieves the end goal, as well as strategically identifies what the end goal needs to be.