Chief Information Officer

Pfizer Senior Data Scientist Explores AI and Machine Learning in the Pharma Sector

Peter Henstock, Senior Data Scientist – Machine Learning, Software Engineering, Statistics and Visualization at Pfizer, offered insights into the use of artificial intelligence (AI) and machine learning in the pharmaceuticals sector during his keynote presentation at the 2018 Information Technology & Security Forum in Boston on December 12. In his presentation, “Capturing the Value of AI & Machine Learning: A Perspective from the Pharma Sector on Overcoming the Roadblocks and Riding the Trends,” Henstock explained how Pfizer successfully integrated AI and machine learning into its day-to-day operations.

The push for AI and machine learning is increasing among global organizations. Meanwhile, an organization that realizes the benefits of AI and machine learning could use these technologies to accelerate its revenue growth, drive product and service improvements and much more.

AI and machine learning are revolutionary technologies, and they frequently help organizations speed up and improve their operations. However, choosing the right AI and machine learning tools sometimes proves to be difficult.

At Pfizer, the organization prioritized drug development and delivery. Pfizer wanted to retrieve and analyze data and transform it into meaningful insights. Next, Pfizer could use these insights to find ways to provide individuals with safe, effective drugs.

“We’re trying to make [drugs] cheaper and get them to patients faster, and we really hope AI is a solution,” Henstock indicated. “We want to help patients and create better and safer drugs faster … but we’re also dealing with data integration, a lack of talent and a lack of AI investment.”

The sheer volume of data available to Pfizer and other organizations that leverage AI and machine learning can be overwhelming. In some instances, organizations struggle to differentiate timely, accurate and relevant data from all other information. And in these scenarios, organizations risk missing out on actionable insights.

Furthermore, data analysis is a complex process. Data scientists may commit significant time and resources to mine unstructured and structured data sets in the hopes of generating meaningful insights. But despite data scientists’ best efforts, it may be difficult to transform large collections of data into insights that an organization can use to achieve its desired results.

“We want to take all of these data sets that we are gathering as businesses … but there is so much data that humans really can’t [analyze it],” Henstock noted.

Differentiating AI from machine learning is crucial, too. AI provides a system with the ability to learn from data and identify patterns and trends hidden within data sets. Comparatively, machine learning in integrated into a system and gives this system the ability to learn on its own without programming.

“All artificial intelligence is essentially machine learning, automatically using data to come up with ideas and models for data,” Henstock stated.

An organization that understands how to properly leverage AI and machine learning tools in conjunction with one another may be better equipped than others to optimize the value of the information at its disposal.

Thanks to machine learning, an organization can establish processes to streamline data collection and analysis. Plus, machine learning can help an organization quickly gain meaningful insights and use these insights to make data-driven decisions.

“It’s really difficult to make rules,” Henstock pointed out. “Now, machine learning has come in and allows us to do things that we previously couldn’t do with rules.”

Additionally, machine learning helps an organization reap the benefits of deep learning and other state-of-the-art data analysis technologies.

Using machine learning with myriad data analysis technologies ensures that an organization can obtain the insights it needs, exactly when it needs them. Then, an organization can use these insights to explore ways to bolster its everyday operations and get the most out of its time and resources.

“Machine learning takes us from a software engineering approach … to a new model,” Henstock said. “It also allows us to use deep learning and other approaches to take instances and create software without making rules.”

There is no shortage of opportunities available to organizations that leverage AI and machine learning, either. AI and machine learning tools can be integrated into the operations of an organization in any industry. Also, these tools can help an organization study its own operations, as well as assess the competitive landscape across its respective industry.

“AI is doing something remarkable for companies and bringing in a lot of value. And it is transforming entire industries,” Henstock said.

The future of AI and machine learning looks bright for Pfizer and other organizations. With an ongoing commitment to innovation and improvement, an organization could identify new ways to incorporate AI and machine learning into its everyday operations. As a result, this organization could generate actionable insights that lead to long-lasting innovation and improvement.



Peter Henstock is the machine learning and artificial intelligence (AI) technical leader within Pfizer’s informatics organization. He is working to push machine learning approaches throughout Pfizer, especially across the research domain.


Visit Argyle Executive Forum's CIO LEADERSHIP FORUM - Toronto in Toronto, ON on Aug 31, 2020

right arrow icon

Next Article:
Finance Leaders Discuss Human Capital in a Digital Paradigm