Novartis’ Data Science Enablers

In the effort to get its data “into good shape” and help define the future of digital at Novartis, last year the company recruited Raj Patil, as Head of Data Strategy, and Shahram Ebadollahi as Head of Data Science and Advanced Analytics. As with Novartis’ Chief Digital Officer, Bertrand Bodson, both experts came from outside pharma. Patil joined from investment company BNY Mellon, and prior to that served for two years as Google’s Engineering Manager, while as VP, Innovation & Chief Science Officer at IBM, Ebadollahi was the technical founder of IBM Watson Health. As such, these appointments underline how the potential for data science is set to make difference at “a company that is primed for change”.

Pharm Exec spoke to Patil and Ebadollahi about their recent move to pharma and the task that lies ahead.

PharmExec: What attracted you to Novartis and how does the work differ from your previous roles?

Sharhram Ebadollahi

Ebadollahi: What I am doing in Novartis is very complementary to what I was doing before. Only the perspective has changed. In my previous role(s) at a technology company, my goal was to innovate at the cutting edge and lead teams that devised and innovated methodologies deeply rooted in data science and AI for healthcare and life sciences applications. The primary role there was to bring healthcare specificity and applications into a deep technology company. Being at Novartis provides a completely different angle. Here my role is bringing the thinking, technology and operation of data science and AI practice into a deep and very innovative medicine’s company.

Novartis has great data assets across the pipeline and has a great group of data scientists already embedded and operating in various parts of the organization. There is clarity of vision and tremendous support to really push and lead in the area of data, digital and data science and AI.

Raj Patil

Patil: Having worked previously within finance, I am used to working in highly regulated sectors and the process of interoperability through the data life cycle helps manage that. Agility in this role will be critical. The curation and unification of data in a frictionless way is critical – my mantra is connect as you collect. I was attracted to Novartis by the potential to have impact. We’re focusing on end-to-end value chain optimization, removing friction, and building strong foundations.

Ebadollahi: We have a huge opportunity to use insights derived from data through applications of data science and AI tools and expertise to have a meaningful impact on people’s lives. Let’s not lose sight of why we are doing what we are doing here. What personally drives me is to help make Novartis a “medicine and data science company” so we can more effectively help people who are suffering, and for whom our innovations in medicine can help and provide hope. To do that we need to bring drugs to market faster, work smarter and more efficiently and focus on delivering personalized experiences. Data science and AI have a major role to play there and we are on the mission to make that happen.

What kind of initiatives are you involved in?

Ebadollahi: There are multiple high-profile programs happening as we speak across the broader organization in bringing data science into various core areas across the pipeline. At this stage, I am focused on building an industry-leading data science and AI team and operation, which we are all proud of and can not only support but accelerate our digital transformation and elevate the practice of data science in Novartis and in the pharma industry. We have a lot on our plate, from building up expertise, connecting our expertise internally and externally, filling the gaps in our tools and technologies, and ultimately embedding data science and AI in the core of our operation and workflow.

Patil: I’m focused on helping accelerate our 12 digital lighthouse projects – key enterprise initiatives that are kick-starting our digital transformation. A key priority is also building data and digital capabilities across projects so they can get velocity.

More specifically, we are building capabilities to “connect as we collect” internal and external data to build large-scale knowledge graphs. This will enable automated inferencing and reasoning to “assist and suggest” our 20,000+ scientists and physicians who are involved in the drug discovery and development processes as citizen data scientists.

When do you expect to see your efforts make an impact?

Patil: I expect we will see the early impact of the framework within the next 12–18 months. Working iteratively within an agile framework ensures we can track progress on an ongoing basis, course correcting as needed. Core to what we will be doing involves building capabilities and strengthening expertise which I see as an on-going priority.