The prospect of precision medicine, driven by machine learning, in tailoring the diagnostics used in individualized treatments to improve patient health and outcomes is very promising. While many life sciences and healthcare organizations are excited about this prospect, they are still faced with data challenges of accessing patient data and generating clinically actionable insights necessary to generate hypotheses, diagnose disease, and classify patient risk. Cambridge Healthtech Institute’s Inaugural Data Science, Precision Medicine and Machine Learning symposium explores these challenges and accelerated solutions. Through interactive sessions and panel discussions, leading researchers and thought leaders will discuss efforts of AI and ML-driven technologies in accelerating R&D and optimizing therapies based on data. The symposium will also feature experts outside of life sciences who are implementing and adopting data models and AI technologies. Hearing their case studies and learning their best practices may be helpful in your own work.   

Who Should Attend: Directors, Managers, Researchers, and Scientists from Pharma, Biotechs, Academia, Government and Healthcare Organizations working in AI, Bioinformatics, Biomarkers and Personalized Medicine, Business Development, Computational & Systems Biology, Data Operations, Data Science, Diagnostics/Diagnostics Research, Machine Learning, Precision Medicine Technology, R&D, and Statistical Science

Final Agenda

Thursday, March 14

7:00 am Registration Open and Morning Coffee (Continental Foyer)

DATA SCIENCE OVERVIEW
Continental 2

8:25 Chairperson’s Opening Remarks

Eleanor Howe, PhD, Founder and CEO, Diamond Age Data Science

8:30 Decoding Data Science

Jena_GrishmaGrishma Jena, Cognitive Software Engineer, IBM Watson Customer Engagement

Data Science is one of the biggest buzzwords out there. Companies are jumping on the data science bandwagon, hoping to predict diseases, create precision medicine, process records efficiently and do many more tasks. But what exactly does Data Science mean? What is the need for it and what questions can it answer? What is the Data Science pipeline and how to use it to answer those questions? Join me to unravel these mysteries using relevant examples from the field of Life Sciences.

9:00 Training Your Data Scientist

Peled_LotemLotem Peled, Data Science and NLP Lecturer

Data Scientists are among the most sought after professionals in the market today. A Data Scientist requires a combination of strong programming skills, deep mathematical and statistical knowledge, and an ability to communicate that knowledge outwards. I will discuss the Data Science profession, training opportunities to become a Data Scientist, and how to build your own in-house Data Science training program for suitable employees.

9:30 Machine Learning for Healthcare Diagnostic Developer and Payer Orgs

Saripalli_PrasadPrasad Saripalli, PhD, Vice President, Data Science, edifecs

This is a technology and application overview in depth on ML and AI applications in Healthcare Analytics. It is designed to help Payers and Providers understand the role and use cases of Machine Learning for Data Analytics and their applications to Molecular Health, Digital Medicine for the Diagnostic Developer and Payer Orgs. To this end, we will first provide an in depth introduction to Machine Learning, its essential methods and algorithms, and the tools used such as R, Spark, Storm and Hadoop, and the relationships among ML, AI, Statistics and the traditional Business Intelligence (BI). This will help one to understand the essential innovative value and novelty of ML and AI methods in the context of HIT. Using a few specific Payer and Provider Use Cases, we will discuss how ML and AI can be used to enrich the analytics practice for Healthcare. We will then conclude with a discussion on how to critically evaluate the ML and AI use cases, apps and start-ups for Healthcare, and identify the ones which could be profitably deployed in the near-term, intermediate term and long term.

10:00 Lessons Learned from a Diabetes Dataset: Data Science Meets the Real World

Joshi_SanjaySanjay Joshi, Industry CTO Healthcare, Dell EMC

This talk provides a summary overview of the data science process of engineering, DevOps, transformation and continuous model tuning, understanding the topic & disease, explainable machine learning, and regulatory process.


10:30 Coffee Break in the Exhibit Hall with Poster Viewing

DATA SCIENCE FOR PERSONALIZED MEDICINE:

APPROACHES, OPEN CHALLENGES, FUTURE DIRECTIONS
Continental 2

11:15 The Role of Data Science in Healthcare, Life Sciences & Precision Medicine: Key Application Areas for Growth

Basu_SudeepSudeep Basu, PhD, Practice Leader, TechVision-Innovation Services, Frost & Sullivan


11:45 Streamlining New Industry Models and Precision Medicine Collaborations

Ramos_KennethKenneth S. Ramos, MD, PhD, PharmB, Associate Vice President for Precision Health Sciences; Executive Director, Center for Applied Genetics and Genomic Medicine; Professor of Medicine, University of Arizona Health Sciences

 

 

 

12:15 NEW: How SAFE is your AI?

vettrivel vishnuVishnu Vettrivel, CTO, R&D, Wisecube


12:45 Enjoy Lunch on Your Own

1:15 Session Break

MINING AND VISUALIZING ASSOCIATIONS OF DATA AND KNOWLEDGE
Continental 2

1:55 Chairperson’s Remarks

Itan_YuvalYuval Itan, PhD, Assistant Professor, Department of Genetics and Genomic Sciences; Member, Charles Bronfman Institute for Personalized Medicine, Icahn School of Medicine at Mount Sinai, New York, NY


2:00 New Methods to Detect Pathogenic Mutations and Genes Using Machine Learning

Itan_YuvalYuval Itan, PhD, Assistant Professor, Department of Genetics and Genomic Sciences; Member, Charles Bronfman Institute for Personalized Medicine, Icahn School of Medicine at Mount Sinai, New York, NY

Whole exome and whole genome sequencing provide hundreds of thousands of genetic variants per patient, of them only very few are pathogenic. Current computational methods are inefficient in differentiating pathogenic mutations from neutral genetic variants that are predicted to be damaging and cannot predict the functional outcome of mutations. We will present deep learning approaches and machine learning methods in the role of detecting pathogenic mutations. Visualization tools for better utilizing NGS data will be presented to understand human disease genomics.

2:30 Scaling Ontology and Knowledge Graph Development for Childhood Health, Cancer, and Other Domains

McCusker_JamesJames McCusker, PhD, Director, Data Operation, Rensselaer Polytechnic Institute


3:00 Using Knowledge Graphs to Model the Domain Expert’s View of Data

Sequeda_JuanJuan Sequeda, PhD, Co-founder and Senior Vice President of Technical Sales and Research, Capsenta, Inc.


3:30 Refreshment Break and Poster Competition Winner Announced in the Exhibit Hall

EFFICIENT COMPUTATIONAL AND COMPUTER-BASED MODELING METHODS
Continental 2

4:15 Computational Identification & Screening for Deleterious Mutants

Kloczkowski_AndrzejAndrzej Kloczkowski, PhD, Computational Biologist; Principal Investigator, Battelle Center for Mathematical Medicine in the Research Institute of the Nationwide Children’s Hospital; Tenured Professor, Pediatrics, Department of Pediatrics, The Ohio State University College of Medicine

The practice of precision medicine relies on individual patient genome sequences, among other data. The challenge for the interpretation of genome sequences is to distinguish between the normal and the disease-related mutations. The human proteome contains over 20,000 structures, and the consideration of possible mutations increases the number of potential protein structures enormously. This presentation discusses the analysis of big sequential and structural data and application of machine learning lead to identification of deleterious mutants and to the advancement of precision medicine.

4:45 Text-Based Knowledge Scanning for Drug Development

Priami_CorradoCorrado Priami, PhD, Founder and CSO, The Microsoft Research - University of Trento Centre for Computational and Systems Biology (CoSBi)

We have recently developed a natural language process technology to mine scientific literature, patents and clinical trials and to graphically visualize the outcome to help scientists and practitioners being up-to-date with the latest findings in their fields. An application of the technology will be discussed.

5:15 Turning the Tide of Failed Neurology and Psychiatry Trials: Using Quantitative Systems Pharmacology-Based Modeling of Individual Virtual Patients to Support Clinical Trial Design

Geerts_HugoHugo Geerts, PhD, MBA, BA, CSO, Computational Neuropharmacology, In Silico Biosciences; Adjunct Associate Professor, Perelman School of Medicine, University of Pennsylvania

Mechanistic Disease Modeling or Quantitative Systems Pharmacology is a completely new approach for supporting Drug Discovery and Development. By integrating domain expertise in a formalized way, the computer model can go beyond associations and correlations that is often the outcome of bio-informatics studies with Big Data. With the emphasis on ‘Pharmacology’ the effects of therapeutic interventions on the underlying pathology can be simulated in a more quantitative way and key issues can be detected and corrected long before they become a problem in clinical trials.

5:45 Reception in the Exhibit Hall with Poster Viewing

6:45 Close of Day

Friday, March 15

8:00 am Registration Open and Morning Coffee (Continental Foyer)

DEVELOPING AND USING AI TECHNOLOGIES FOR PRECISION MEDICINE IN HEALTH & CANCER TREATMENTS
Continental 2

8:25 Chairperson’s Remarks

Sanjay Joshi, Industry CTO Healthcare, Dell EMC

8:30 Building a Computational Model of the Human Anatomy Using Medical Images

Miolane_NinaNina Miolane, PhD, Postdoc & Lecturer, Statistics for Neuroimaging, Stanford University; Laureate of the 2016 French L’Oréal-Unesco Fellowship for Women in Science

Neurodegenerative diseases, like Alzheimer’s, are characterized by brain shape changes observable on magnetic resonance imaging. Likewise, many diseases can be detected by looking at organ shapes on various medical imaging modalities. We present the promising results of computational anatomy, a field building a computational model of the human body, with applications in diagnosis.

9:00 Open Source Machine Learning for Predicting Optimal Cancer Drug Response

Vannberg_FredrikFredrik Vannberg, PhD, Assistant Professor, Georgia Institute of Technology

Machine learning is transforming medicine and is slated to improve patient outcomes over the next decade. To date there has been widespread sharing of machine learning frameworks and code related to a number of disciplines but proportionality there has been less sharing of code and models for health-related research. This talk will discuss the importance of open source efforts within health, including our work on cancer drug prediction, and how we can continue to catalyze innovation in this space through open and transparent sharing of code.

9:30 Medical Big Data Efforts in China: An Integration of Medical Informatics, Bioinformatics, and AI

Mo_WeikeWeike Mo, PhD, FAACC, Vice President, Precision Medicine, Digital China Health

China National Cancer Center is building a centralized medical data platform for all Chinese cancer patients with Digital China Health. To date, we have collected data for more than 6 million patients from over 30 oncology specialty hospitals. We have applied medical informatics, bioinformatics, and artificial intelligence technologies to clean up, structure, and utilize all the data. Although we are at an early stage of understanding the big data set, we see potential to revolutionize precision medicine practice in hospitals, insurance companies, and pharma companies.

10:00 KEYNOTE PRESENTATION: 10 Misconceptions of AI in Medicine & Healthcare

Chang_AnthonyAnthony Chang, MD, MBA, MPH, MS, Medical Director, Heart Failure Program and Chief Intelligence & Innovation Officer, Founding Director of The Sharon Disney Lund Medical Intelligence and Innovation Institute (Mi3), Children’s Hospital of Orange County (CHOC); Author of Medical Intelligence: Principles and Applications of Artificial Intelligence in Healthcare and Medicine (releasing in Spring 2019)

Medical intelligence is derived from pairing machine intelligence with human intelligence. As AI and machine learning approaches and methods continue to evolve, they impact how physicians practice and perform their tasks. As a practicing cardiologist and chief intelligence and innovation officer with a data science background, Dr. Chang will share how he is cautiously embracing this paradigm shift in practicing medicine and the “hype vs. reality” in healthcare’s adoption of AI.

10:30 Coffee Break in the Exhibit Hall with Poster Viewing

IMPLEMENTATION AND ADOPTION OF DATA MODELS AND AI TECHNOLOGIES: CASE STUDIES FROM OUTSIDE LIFE SCIENCES
Continental 2

11:15 Health Analytics Platform-as-a-Service

Burton_MatthewMatthew Burton, MD, Vice President Clinical Informatics, Apervita

Dr. Burton will be discussing the use of Health Analytics Platform-as-a-Service capabilities that enable AI to be fully leveraged across clinical, operational, and broader industry use cases at greater speed and with incremental cost and effort. Use cases including Clinical Pathways, CDS Rules, Demand Forecasting, Scheduling Optimization, Cost Avoidance, Plan Contracting, and Performance Management will be presented to improve quality of care.

11:45 From Neurons to a Brain: Integrating AI into Your Startup’s Workflow

Dresner_EmilyEmily Dresner, CTO, Upside Travel

In travel, we have flight data, weather data, flight equipment stats, hotel information, hotel ratings and none of this data is consistent in what data it provides, let alone format or how it correlates across data points. We’ve found three areas to focus on organizationally to launch our program and integrate ML throughout the machine learning engineering, data pipelining and the ETL to make data useable and useful, and product management. This talk explores the three areas, focusing on the last two—our dirty data challenges and how product thinking helps to fuel data hypothesis and research.

12:05 pm Challenges of Data Handling and Interface Designs

Gatta_ShannonShannon Gatta, Flight Software Engineer, NASA Langley Research Center

In my role, I analyze software requirements and architectures while assisting in documenting control and data interface designs. I also create designs that develop and test software modules assisting in the integration and testing of software subsystems while providing documentation such as reports, presentations or graphs. I will discuss some of the challenges of data handling and interface designs.

12:15 Interactive Panel Discussion: Challenges of Implementing and Adopting Data Models and AI Technologies

Chairperson & Moderator:

Sanjay Joshi, Industry CTO Healthcare, Dell EMC

 

Panelists:

Matthew Burton, MD, Vice President Clinical Informatics, Apervita

 

Anthony Chang, MD, MBA, MPH, MS, Medical Director, Heart Failure Program and Chief Intelligence & Innovation Officer, Founding Director of The Sharon Disney Lund Medical Intelligence and Innovation Institute (Mi3), Children’s Hospital of Orange County (CHOC); Author of Medical Intelligence: Principles and Applications of Artificial Intelligence in Healthcare and Medicine (releasing in Spring 2019)

 

Emily Dresner, CTO, Upside Travel

 

Shannon Gatta, Flight Software Engineer, NASA Langley Research Center

 

12:45 Close of Symposium