In February, UCI launched the Institute for Precision Health. It is a campus-wide interdisciplinary effort that integrates UCI’s strong health sciences, engineering, machine learning, artificial intelligence, clinical genomics and data science capabilities. Its purpose is to identify, develop, and deliver the most effective health and wellness strategies for individuals, and in doing so confront the associated challenges of health equity and high healthcare costs. is.
IPH brings a multifaceted, integrated approach to what many call the next big advancement in healthcare. The Institute is an ecosystem for cross-disciplinary collaboration.
Dr. Peter Chan leads the Applied Artificial Intelligence Research Group at IPH, is Co-Director of the Center for Artificial Intelligence in Diagnostic Medicine, and Assistant Professor of Radiology in the School of Medicine. Dr. Chang’s unique perspective draws on his experience as a radiologist and full-stack software his engineer, with his 10+ years of experience building FDA-approved tools used in hospitals around the world. is born from
Chang came to academia after launching a successful start-up company. Computers He has the distinction of being one of the few doctors in the country who teach in the science department. His work at IPH uses AI and machine learning to design practical solutions to real-world clinical problems to deliver cost-effective, value-based care. Here, Dr. Chang talks about his AI potential and what he sees for the future of the Institute for Precision Health.
For starters, how would you explain why machine learning and AI are important to healthcare today?
It should be understood that the latest form of AI, the deep learning neural network family of algorithms, has completely revolutionized the way machine learning algorithms learn and think. Traditional forms of AI have required humans to carefully examine lists of patterns, rules, and assumptions, and manually incorporate or program that human experience into computers. But modern forms of AI allow computers to extract patterns and make inferences without human a priori assumptions. For example, if you want to teach an algorithm how to play a game of chess, you can explain the rules of chess and have two AIs play against each other.
This paradigm shift is an entirely new way to approach the design of learning algorithms. And interestingly, this strategy has enabled modern AI systems to learn new and interesting information that even human experts might not have previously been able to perceive. In video games, we often think that the AI is losing on purpose, only to find that at the last minute the computer comes back and beats the human by a small but consistent margin each time. is. Of course, for healthcare, AI could potentially discover patterns without bias from flawed human assumptions or explicit programming. That’s where the power really comes into play.
And it is a core component of precision health, healthcare informed by AI and machine learning. However, progress usually comes with drawbacks as well. Are there any downsides to Precision Health?
I don’t know if I would call that a downside, but I would say it has a lot of hype. I mean, expectations are often overinflated and ultimately those expectations aren’t met and people probably know technology is something I know all too well. I’m clearly a proponent of this technology, but there’s a lot we don’t know. It’s still early days in terms of development, especially in the medical field. There is a lot of room for improvement. So we should admit it. And we have to admit that while progress and potential are perfectly there, they may be slow to materialize.
UCI will launch the Precision Medicine Laboratory in February 2022. What excites you most about this lab and what do you hope to achieve?
The IPH team is a strong collaborative team with diverse backgrounds. I believe this is an important part of our unique approach to precision health and big data at UCI. At the same time, although the team is large, my role is very specific. My background is unique, especially in that I routinely build state-of-the-art AI algorithms and also practice as a certified radiologist. From this perspective, we hope to bridge the gap between technical and clinical experts to accelerate the translation of AI research for healthcare.
How rare is it to be a doctor or professor with an AI background?
As of 2022, this combination remains extremely rare. As a good example, I’ve heard that I’m the only doctor anywhere in the country teaching her Technical Deep Learning classes in the computer science department. This course incorporates a hands-on curriculum that uses medical imaging data to build new AI algorithms every week using the same libraries and tools developed by experts at Google, Facebook, and Uber. increase.
Prior to UCI, my real experience with AI began with research in the precision medicine AI field and ended up in a start-up in the radiology deep learning field. As part of the company, I am active in data science and engineering, innovating the latest AI techniques and applying them to medical imaging. In this role, my experience with AI and machine learning comes from building state-of-the-art algorithms using industry-standard tools and regulatory approvals by the FDA, the European CE Mark, and other international bodies. I think all of this experience complements what is normally expected of a doctor.
What made you join the UCI?
When I was looking for a full-time faculty position, I wanted a position that would allow me to continue doing a hybrid clinical and AI job. Interestingly, such a position did not exist three years before him. So the main reason I joined Irvine was because a UCI leader saw an opportunity in me to expand his AI and machine learning capabilities across the healthcare community. More specifically, I was hired here to build the AI Center. It is an integral part of today’s new Precision Health Laboratory. Our team has grown a lot, but I was one of the first faculty members in medical school to dedicate time and career to AI in medicine. Thus, I have invested in precision medicine laboratories from the beginning.
The process of finding your dream job must have been interesting.
If you asked me where I was supposed to be four or five years ago, I probably imagined myself working in industry. But simply put, my only priority is to have access to resources that allow me to pursue impactful work in healthcare AI. If that meant working at Google, I would have been at Google. But the reality is, here at UCI, we have a unique set of tools and resources that even the tech giants in the industry can’t match. was given. For example, even Google, with all its resources and talent in engineering and other data science areas, doesn’t have access to hospitals. In contrast, here at UCI, you can take lab-built tools and power them up the next day in a realistic clinical environment to see if they actually help doctors in their day-to-day work. Connect with researchers in the basic sciences through clinicians who are global experts in treating some very specific diseases. In addition, here at UCI, I continue my medical practice as a radiologist at the hospital one day a week.
What is your job like when you are not working in medicine or education?
By design, I try to immerse myself in the technical details of new AI technologies as much as possible. Working with students in the lab and writing software code is the highlight of my day. Most of the time, I work with engineering and data science teams to build algorithms and figure out how best to incorporate them into clinical practice. My team can be described as the ‘boots in the field’ to bring the application into real practice. And, of course, I spend a good portion of her day with the clinician discussing potential AI solutions to everyday problems. Most of the time the conversation starts like this:
How many people are there in your team?
The Artificial Intelligence Center currently has about seven to eight full-time staff, in addition to a large number of grant-funded students, trainees, and postdocs. Additionally, I try to get as many student volunteers as possible, trainees who just want to get hands-on experience. There is a large community of individuals without formal funding, who just want to learn. I think that’s why the center has become a popular place around the UCI.
Now that IPH has started, what do you want to tackle first?
Prior to the start of IPH, a small group at UCI focused on precision health and omics. In parallel, my team focused on AI and machine learning applied to precision health problems. In that regard, there are several early projects combining our expert backgrounds to explore AI predictive analytics across multiple diagnostic modalities, including electronic health records (EHR), radiology, and omics data. Including DNA, RNA and proteomics. This interdisciplinary work is the true embodiment of IPH and, frankly, impossible without experts like our team to help guide you along the way. Some important priority areas of research include ALS, dementia, stomach cancer and COVID, among others.
Any final words about the future of precision health?
AI and precision health is an exciting new area of research, but for now, we recommend keeping your feet on the ground and being patient. With so many unknowns to explore and understand, a balanced perspective is needed to truly advance these technologies in a way that ultimately benefits researchers, clinicians, and patients.
For more information on supporting this and other activities at UCI, please visit the Brilliant Future website at: https://brilliantfuture.uci.edu. The Brilliant Future campaign, which opened to the public on October 4, 2019, aims to increase awareness and support for the UCI. With 75,000 alumni and $2 billion in philanthropic investment, UCI aims to reach new heights of excellence in student success, health and wellness, research and more. UCI Health Affairs plays a key role in the campaign’s success.Learn more by visiting https://brilliantfuture.uci.edu/uci-health-affairs/.
About the UCI Institute for Precision Health: Established in February 2022, the Institute for Precision Health (IPH) is a multifaceted, integrated ecosystem for collaboration, combining the collective knowledge of patient datasets with the power of computer algorithms, predictive modeling and AI. to maximize IPH combines UCI’s strong health science, engineering, machine learning, artificial intelligence, clinical genomics, and data science capabilities to deliver the most effective health and wellness strategies for each individual, and in doing so, , to tackle the related challenges of health equity and high standards. cost of care. IPH is part of UCI Health Affairs and is co-directed by Tom Andriola (Vice President for Information, Technology and Data) and Leslie Thompson (Donald Bren Professor of Psychiatry and Human Behavior, Neurobiology and Behavior) . IPH has seven: SMART (statistics, machine learning-artificial intelligence), A2IR (applied artificial intelligence research), A3 (applied analytics and artificial intelligence), Precision Omics (facilitating the translation of genomic, proteomics, and metabolomics research results). consists of one field. Collaborator for Health & Wellness (provides an ecosystem that fosters cross-disciplinary collaboration through the integration of health-related data sources) Education and Training (provides data-centric education for students and healthcare professionals, maximizing licensure) make it available).