Focus on Pharma and Health Care: Bio-IT World Conference Digs Deep on AI Applications


In theory, artificial intelligence and machine learning can be applied to nearly every process in healthcare, says Bill Evans, Managing Director and CEO, Rock Health. But how best to do that is still being hashed out.

At the Bio-IT World Conference & Expo to be held April 16-18 in Boston, three tracks will be devoted to AI in pharma and biotech, genomics, and healthcare. Speakers from pharma, biotech, and major research and care facilities will share the tools they are using and how they are applying them to bring AI’s promise to bear on drug discovery and patient health.

The AI Trends editorial staff will be there, and we are already marking our agendas for the week. Here are a few of presentations that have caught our attention.

–The Editors

Taking the big picture approach, Edmon Begoli, Chief Data Architect, Oak Ridge National Laboratory, says AI is becoming a key player in the convergence of medical data and computer technologies. Begoli will provide a big picture look at how AI’s role has evolved in precision medicine, what the key drivers and challenges are, and trends to look for during the next 10-15 years. Wednesday, April 17, 11:00am

Several speakers hail from hospitals and research centers. Their patient data sources have been growing exponentially, and AI could be the answer to managing the hundreds of thousands of genetic variants per patient that come from exome and whole genome sequencing.

Current computational methods are inefficient in differentiating pathogenic mutations from neutral genetic variants and cannot predict the functional outcome of mutations, says Yuval Itan of the Icahn School of Medicine at Mount Sinai. Itan will present several machine learning tools in use at Mount Sinai: a deep learning approach to efficiently detect pathogenic mutations, a machine learning method to estimate whether a mutation results in gain- or loss-of-function, a gene burden study to detect genes and pathways enriched with disease-causing mutations, and visualization tools to better use sequencing data. Thursday, April 18, 3:30pm

These tools are exciting, says Sandy Aronson, Executive Director of IT, Partners HealthCare Personalized Medicine, especially when we can apply them to patients. However, realizing the potential of algorithmic-enhanced patient care will involve more than developing algorithms and making them available to clinicians, Aronson argues. The most substantial benefits may come from the dramatic reformulations of the care delivery process these new capabilities make possible. Aronson will discuss how to redesign clinical workflows to take maximum advantage of these new tools. Wednesday, April 17, 11:30 am

Anthony Philippakis, Chief Data Officer at the Broad Institute and a cardiologist at Brigham and Women’s Hospital will outline the Broad Institute’s Data Sciences Platform, a methods development and software engineering group dedicated to maximizing the impact of the data sciences on the life sciences. Philippakis will focus on how DSP engineers, analysts, and designers build applications and capabilities to serve patients, data generators, and researchers. Wednesday, April 17, 1:55pm

Vitaly Herasevich at Mayo Clinic believes we are entering a phase of market saturation for health-IT commercial systems. Competition in this space will lead to innovation and a proliferation of new technologies with difficult-to-predict effects on providers, patients, and health systems. Herasevich will offer a systematic approach to the evaluation of technology in healthcare. We’ll need it, he says, to reliably discriminate between useful innovation and clever marketing. Wednesday, April 17, 4:00pm

Neil Tenenholtz, Director of Machine Learning at Massachusetts General Hospital and Brigham & Women’s Hospital Center for Clinical Data Science, knows all too well the unique challenges for machine learning within the healthcare sector. From protected datasets to the complexity of the physician’s workflow, the required background knowledge is often siloed across multiple domain experts. Tenenholtz will discuss how to achieve effective knowledge transfer between these groups and maximize the likelihood of a developing a clinically successful product. Thursday, April 18, 11:10am

John Mattison, Kaiser Permanente, will look at AI within integrated delivery networks and share Kaiser’s early experiences. Value-based care is best realized through synergy with integrated delivery networks, Mattison says. Machine learning creates a natural synergy. Thursday, April 18, 2:00pm

AI and algorithmic-enhanced care might seem like an exciting leap forward for some, but what do investors think? In a panel led by Viet Nguyen, a physician and health-IT champion, investors will discuss how they invest in and evaluate healthcare start-ups, real-world applications of AI in the healthcare industry, emerging partnerships between integrated health systems and pharma, the impact of AI on future jobs in the healthcare industry, and what can healthcare learn from other industries. Panelists include Sagran Moodley, UnitedHealth Group; James Falkoff, Converge; and A.G. Breitenstein, Optum Ventures. Wednesday, April 17, 4:30pm

On Thursday, Bill Evans, Managing Director and CEO, Rock Health will outline the various responsibilities of entrepreneurs, enterprise leaders, and investors to discriminate between incremental improvements and the 10X improvements that will transform the industry. Evans will share practical, real-world use cases while providing a framework for evaluating impactful technologies. Thursday, April 18, 3:20pm

Pharma a biotech will weight in as well. Pfizer has several presentations on the program; two are exploring how to use AI as a driver of productivity in drug discovery & development. Peter Henstock will describe how his team uses AI to navigate relationships and identify context for pharma data. Our baseline search capability delivered answers from the databases-level through enterprise-level results, he said. The limitations are being able to create the right queries to find all the relevant information without having to craft the perfect queries or sift through 1000s of entries. Wednesday, April 17, 2:25pm Morten Sogaard will give an overview of the impact of AI on productivity within pharma, highlighting process engineering and automation, drug design and manufacturing, and target and biomarker discovery and validation. Thursday, April 18, 2:00pm

Margaret Bray at Alexion is applying AI to rare disease diagnostics. She’ll describe her work as well as a look at the limitation of current methodologies and areas for future growth. Thursday, April 18, 2:30pm

Unstructured data is a significant source of cross-disciplinary insight. However, it’s also a significant hurdle to manually read and digest and extrapolate a holistic view. Robert L. Martin, IBM Watson Health, will explain how AI applies a common framework to the disparate data using NLP and predictive analytics to construct a network of known and inferred connections between biological concepts enabling researchers to make informed, cross-silo decisions. Thursday, April 18, 11:40am