INTRO
Meet Dr. Jason H. Moore, Co-Principal Investigator at PennAITech. By day, he serves as the Chair of the Department of Computational Biomedicine and Director of the Center for Artificial Intelligence Research and Education at Cedars-Sinai Medical Center. Dr. Moore's research is geared toward understanding the complexity of human health through the development of artificial intelligence (AI) methods, particularly automated machine learning (AutoML), to predict disease risk. His work aims to democratize AI, making it accessible to everyone, and is especially focused on diseases like Alzheimer's. An elected fellow of multiple prestigious societies, including the American College of Medical Informatics and the American Statistical Association, Dr. Moore brings a wealth of expertise to the field. Dive in as we explore Dr. Moore's insights on the role of AI in tackling the complexities of aging and dementia, and how it can bridge gaps in current research and care.
#1 - Can you share with us a little about your work or research?
My research is motivated by understanding the complexity of human health. We develop artificial intelligence (AI) methods and software for developing predictive models of disease risk. Recent work has focused on the development of automated machine learning (AutoML) methods that can democratize AI by making this powerful technology available to everyone regardless of expertise or training. We are augmenting these AutoML methods for diseases like Alzheimer's using expert knowledge about genes, drugs, pathways, etc.
#2 - What initially drew you to this intersection of AI, AgeTech, aging, and dementia? Is there a personal story or motivation behind your commitment to this field?
I have a strong family history of Parkinson's disease and am thus very interested in neurodegenerative diseases. Further, diseases such as Parkinson's and Alzheimer's are increasingly prevalent in the population as people live longer. Thus, they represent a significant public health problem. These diseases fascinate me as a researcher because of their tremendous complexity. Most common neurodegenerative diseases are characterized by many genes and many environmental factors that interact with each other differently in different people. This makes risk modeling difficult. AI has an important role to play in embracing this complexity.
#3 - In your view, where is the biggest gap in the current landscape of aging and dementia research and care, and how can AI and emerging technologies help bridge this?
Most studies building risk models consider each biomarker independently and assume that their relationship with disease risk is additive. However, this is not how biology works. Biology is driven by biomolecular interactions in time and space. AI has an important role to play in risk modeling because it can consider non-additive interactions among biomarkers and biological and clinical heterogeneity.
#4 - Any words of wisdom for budding startups or researchers eager to dive into the AI and AgeTech space?
I think the most important aspect of developing AI methods for Alzheimer's disease is the incorporation of biological and clinical knowledge into the modeling building process. We as humans don't solve problems in a vacuum. Why should we expect our AI algorithms to solve problems without access to the wealth of knowledge we have about brain health and disease? Knowledge engineering in AI can help with building more powerful, simpler, and more interpretable models. This requires domain expertise.
#5 - Outside of the lab or office, what’s a hobby or activity you're passionate about?
I learned to program in the early 1980s on an Atari 8-bit home computer (the Atari 800). I still collect these old computers and have fun programming and doing hardware modifications. I love programming in Assembly Language for these old computers because you can control exactly what the computer chips are doing. Programming on modern computers is less rewarding because there are many layers of abstraction between the programming language and the hardware. A few years ago I wrote a game for the Atari 2600 game system from the 1970s and 1980s called Gene Medic (genemedic.org) where you play a future doctor editing DNA in a cell to cure a patient. I also maintain a web page where I post fun projects for these old systems (atariprojects.org).
Bonus - What's the most constructive piece of criticism or feedback you've received in your career, and how did it shape your research or business trajectory?
I started my career developing AI methods and software for the human genetics community. I was discouraged from pursuing this line of research because it was believed that all genetic effects on disease risk could be modeled by simple additive models. This was a blessing because it pushed me into the AI and biomedical informatics disciplines where the value of my work was recognized for its impact beyond just genetics. Now genetics is waking up to the importance of AI for modeling genetics-based risk factors for common diseases.
