InetSoft Webinar: Healthcare Data Science Platform

Below is the continuation of the transcript of a Webinar hosted by InetSoft on the topic of Healthcare Machine Learning Analytics. The presenter is Abhishek Gupta, Product Manager at InetSoft, and the guest is Jim Reynolds, CTO at Health Analytica.

Abhishek: Okay, so we have a sense of your healthcare data science platform. The challenges, let's talk a little bit about how you now take this and apply it to the problem that your healthcare sector clients have. What are you doing in terms of being able to create recommendations to make healthcare data solutions available, and the speed, how does that factor. So I guess I'm trying to get to what are the requirements that people have that exploit the technology that you put in place when it comes at being actionable.

Jim: Sure, so let's start from the user's perspective, and then I'll work a little bit backwards more towards the technology. From a user perspective, people involved in, any an intelligence environment where you're trying to figure out whether some behavior was good or bad. They're just inundated with lots of little questions, and the longer that those little questions take to answer the harder their job is to get it done.

So what we provide by leveraging the column store technology in Vertica is the ability to rapidly cycle through data and do measurements in an interactive fashion and the column stores provide very, very fast answers, and so we can let people filter data down and get new metric calculations on the fly, and this allows them to essentially self serve by getting answers to questions.

So we allow the users to filter data, and as they filter their data they get new measurements, and new metrics coming back on a very, very rapid fashion. That allows them to answer questions very, very fast, and they don't lose the context.

We also allow them to do this on very large amounts of data which is another problem that they face, is that it's easier to answer questions on small amounts of data. By the time it's in the spreadsheet pretty much most of the decisions have been made, but we allow them to do what they used to do in spreadsheets but on the full size of the data, and that's been a big shift to a lot for effective cadence in field investigations.

Jim: So that's key here for that democratization of getting more people to self serve as you describe it, and not too long ago many customers complain that they'd had to get in line and wait for an analyst or the scientist to take their query and convert it into the right language and using the right tools. Do you still do that? Do you still have an interface between your business type user and your scientist, or are you extending this interface using different languages, using different user experience benefits, to actually allow people to get at this data directly?

Abhishek: So in general we shield people from dealing with the complexity of the data. Most of the time people who are doing investigations are not data scientists. They're not SQL experts. They are experts in their field whether it would be in healthcare or healthcare billing, healthcare coding. They're clinicians. So there's a variety of people, and to be a master both at healthcare and data science is what people typically call the unicorn. Those are very rare people.

View a 2-minute demonstration of InetSoft's easy, agile, and robust BI software.

So we have one set of tools that allows people to get out the data in an interactive fashion and self serve without having to be an expert in SQL or anything of that nature. We also do have data science tools that allows people to interact with and build new data and new data sets that they can go and experiment with, and these are more oriented towards the data science type grade that allows them to go through and do population studies, to do a risk management, large scale crunching of data on risk management.

I can touch more on that there if you like, but that's a separate interface, and then if you really need access to it you have your standard other types of database tools or visualization tools that you'd be able to hook up because Vertica supplies those types of APIs.