Hoboken, New Jersey
18th July 2019
Artificial Intelligence and the Digital Transformation of Healthcare
Stevens Institute of Technology recently hosted a PHUSE Single Day Event (SDE) in Hoboken, NJ. The SDE was a great success, with 75 attendees. The theme of the SDE was “Artificial Intelligence and the Digital Transformation of Healthcare”. With today’s changing landscape of the healthcare industry where organisations are challenged to achieve milestones with limited resources, it becomes important to leverage artificial intelligence (AI) and machine learning (ML) techniques to increase efficiency. The SDE focussed on how AI and machine learning methodologies are being applied to analyse huge datasets, perform repetitive tasks and handle routine cases in healthcare data analytics.
The event began with a keynote from Gregory Prastacos, Dean of the School of Business at Stevens. He emphasised how Stevens has a tradition of innovation, which made it a perfect venue for the SDE. Chris Asakiewicz from Stevens and Frank Corvino of Genesis Research shared some examples where industry–academic partnerships have been used to solve challenging problems. They also described four models of effective industry–university collaboration and presented some basic concepts of AI/ML. This was a good segue to the second presentation of the day, from Dr Marko Zivkovic of Genesis Research and Dr Rong Liu of Stevens, which described the use of Natural Language Processing (NLP) techniques for data identification and extraction from pooled research resources.
Technical presentations followed including the presentation from Richard Bryant of MMS, where he demonstrated how an open-source programming language named Julia, developed for data science and incubated in an academic setting at MIT, could be used for more powerful, faster data processing. Rebecca Laborde from Oracle Health Sciences then showed how customer experience and AI can be leveraged to improve the patient experience. Dr Laborde described the “White House Challenge” to create an AI-driven solution to support patients in the process of matching to appropriate clinical trials and Oracle’s solution to create a “Digital Assistant”. Chris Hurley finished the morning off with a PHUSE update, where he discussed some of the ongoing/upcoming PHUSE events and initiatives.
The afternoon sessions followed a delicious hour filled with good food, plenty of interesting conversation and a fabulous view of New York City across the Hudson River. A presentation by Iraj Mohebalian of Bayer described how he became a data scientist as a result of an unexpected job transition. Iraj provided a roadmap to data science by describing the skills required and tools/technologies one needs to learn. Continuing the afternoon session, we looked further into how we can leverage statistical learning methods such as “Random Forest” to build a model to identify potential fibromyalgia patients, from the presentation by Birol Emir of Pfizer.
The audience then had the opportunity to listen to Richard Baumgartner and Weifeng Xu from Merck, who described “phenotyping” using population-based administrative databases and their applications in cancer, heart failure and frailty. They also emphasised how linked databases SEER-Medicare are a key requirement for successful phenotyping. The event concluded with a Q&A session, where there was lot of discussion about how industry–academia collaboration can be used to solve challenging problems and how industry can come together to develop infrastructure to use open-source packages such as R for regulatory submissions.
This was a very nice venue and the volunteers, presenters and sponsors made this a special day for our attendees. We thank them all for their contribution and we appreciate the attendees for coming in such strong numbers to our event.
Industry–Academic Partnerships (Genesis Research and Stevens School of Business) – Chris Asakiewicz, Stevens Institute of Technology and Frank Corvino, Genesis Research
Using NLP Techniques for Data Identification and Extraction from Pooled Research Resources – Marko Zivkovic, Genesis Research and Rong Liu, Stevens Institute of Technology
"Hello Julia" – An Interactive Look at My First Experience with the MIT Incubated Open Source Programming Language – Richard Bryant, MMS
A White House Challenge: Empowering Patients in Clinical Trial Matching Using CX and AI to Improve the Patient Experience – Rebecca Laborde, Oracle Health Sciences
A Roadmap to Data Science – Iraj Mohebalian, Bayer
Construction and Implementation of a Predictive Model to Identify Potential Fibromyalgia Patients – Jack Mardekian and Birol Emir, Pfizer
Phenotyping the Population-based Administrative Databases: Applications in Cancer, Heart Failure and Frailty – Richard Baumgartner and Weifeng Xu, Merck
View the event brochure.
Photos of the day:
Gregory Prastacos, Dean of the School of Business at Stevens Institute of Technology, kicked off the Hoboken SDE.
Chris Asakiewicz described how industry/academic collaboration is a path forward through an interconnected health data ecosystem.
Frank Corvino and Marko Zivkovic, Genesis Research, and Rong (Emily) Liu, Stevens Institute of Technology, discussed extracting data and utilising natural language processing from literature.
The networking time was a great opportunity to discuss volunteering opportunities and take a selfie at the SDE with the PHUSE Americas Director, Chris Hurley, and Steven Huang from Eisai.
Richard Bryant (right), MMS, and the Health Analytics Collective started the afternoon presentations with an introduction to the Julia programming language.
The SDE attendees listening to the Julia programming language presentation, Richard Bryant.
Iraj Mohebalian, Bayer, discussed a roadmap on how to become a data scientist.
Birol Emir, Pfizer, shared a predictive model to identify potential fibromyalgia patients.
Richard Baumgartner, Merck, explored the use of population-based administrative databases and electronic medical records for real-world evidence generation.
Weifeng Xu, Merck, delved into the challenges of machine learning implementation and software.