Health and our healthcare systems are staggeringly complex. Advances in computation have transformed industries such as finance, journalism, and transportation, but the full promise of computation is yet to be realized for health care. As a result, medical breakthroughs fall short of their potential impact, and people continue to get sick and die needlessly. To transform personal and public health through computation, we need breakthroughs in computation that use the massive data now available to do better: better disease detection, better treatment selection and monitoring, better decision support for medicine and public health, better use of resources.
The UCSF UC Berkeley Joint Program in Computational Precision Health (CPH) is a new graduate training and research program that will change health care practice and policy locally and globally. Our mission is to apply computation to real-world settings to improve the quality, efficiency, and equity of medicine and public health.
A New Paradigm
Computational Precision Health is a new paradigm for developing and deploying adaptive precision interventions for real-world impact. CPH uses novel computational algorithms, tools, and infrastructures to formulate precision problems, develop precision solutions that adapt to time, place, and context, and deploy those solutions more effectively.
Precision Problem Formulation
- New computational methods in causal inference enable us to ask deeper and more precise questions than ever before. We can do more than predict. We can begin to understand why things happen, how to intervene, and when to change our actions.
- Leveraging rich multimodal data streams such as wearable sensor data or environmental data, causal inference can move us beyond blunt average findings towards sharper findings that are tailored to individual people or communities within their specific time, place, health, and socioeconomic context.
- Health is holistic. Four in ten adults in the US have more than one chronic disease, each of which will wax and wane over time. CPH approaches consider all the conditions a person has, going far beyond static one-size-fits-all interventions that target only single diseases.
Computational Precision Solutions
- CPH develops and applies cutting-edge causal inference, machine learning, statistical and other computational methods to develop personalized interventions that incorporate uncertainty, time, and multiplex contexts
- Algorithms and solutions are tailored to the problem at hand and honed in real-world contexts to provide robust solutions that work
- Algorithms are embedded in the real world to allow them to adapt to time, place, context and data streams
- CPH solutions augment human intelligence – computational approaches that model human cognition at the individual level will promote humans and machines to work together, leveraging rather than replicating the strengths of each
- Precision solutions not only detect hidden disparities early but actively intervene to reduce
- To achieve broad impact on real lives, computational advances are deployed in real-world settings in all their technical, economic, operational, and social complexity.
- Standards-based facilitate efficient data capture, data and systems integration, and systems deployment in privacy-preserving and extensible ways
- Real-world deployment must deliver fair and equitable outcomes, are demonstrably safe and effective over time in each and every setting, and are accountable to the most vulnerable
CPH work will begin and end in real-world settings to achieve broad impact and transform personal and public health through computation.
Why UCSF & UCB?
No other place in the world matches the combined strengths of UC Berkeley and UCSF in computer science, engineering and statistics coupled with world-class health care, unparalleled access to health care data from over 7 million patients from all 5 University of California medical centers, and a long and storied history of medical and computing breakthroughs that have repeatedly changed the world.
The CPH program unites these two research and education powerhouses to transform health beyond what either could achieve alone.
CPH’s novel bi-campus structure blends top computational and health faculty from both institutions into a singular unparalleled intellectual community, and recruits new world-class faculty dedicated to CPH. Graduate programs that bridge the two campuses will train a new class of talent to live and think at the intersection of health and computation, providing an engine for innovations and a powerful draw for the best faculty in the world.
The core of CPH is a new and novel Augmented Graduate Group (AGG) that functions as a bi-campus department with philanthropically supported endowed faculty in addition to participating faculty multiple UCSF Clinical departments and multiple UCBs Departments such as Electrical Engineering and Computer Science, Public Health, Statistics, and many others. The AGG offers a PhD program and Designated Emphasis program.
Supporting CPH’s activities is an exceptional high-performance computing and data infrastructure that provides investigators ready access to the latest AI tools and platforms, as well as de-identified health information from hundreds of millions of clinical encounters and curated data resources in both traditional and graph theoretical databases representing biomedicine from genetics and proteins all the way to population-level characteristics. The deep partnership with UCSF Health provides a real-world laboratory for testing and deploying AI and machine learning tools in clinical practice.
The CPH program aims to grow to 10-12 endowed faculty over the next 5 to 7 years with a steady state of approximately 60 graduate students within the PhD program, and a robust Augmented Graduate Group totaling 60-80 faculty.
The program is also developing partnerships with a broad range of stakeholders to ensure relevance and applicability to real-world problems, and to enable translation of CPH advances to evidence-based policies in healthcare, government, industry, and other sectors for maximum impact.
Diversity, Equity, & Inclusion
The promises of computation for health should be available to all persons regardless of personal or group characteristics. For CPH training and research impacts to be equitable, we commit to promoting diversity, equity and inclusion in the realms of research, teaching, and/or service. We recognize the intrinsic relationship between diversity and excellence in all our endeavors and embrace open and equitable access to opportunities for learning and development as our obligation and goal.
Governance & Leadership
CPH is led by faculty Directors Dr. Maya Petersen (UCB) and Dr. Ida Sim (UCSF), under the auspices of the Chancellors of each university. The graduate training program and the Augmented Graduate Group are under the Graduate Divisions. Faculty who are appointed in the CPH Joint Program are within the Division of Computing, Data Science, and Society at UCB and the School of Medicine at UCSF.
Health care accounts for almost 20% of America’s GDP. However, the promise of artificial intelligence and machine learning (AI/ML) have yet to impact the daily delivery of medicine and public health.
The CPH program grew out of a series of collaborative discussions between UCSF and UC Berkeley between 2019 and 2020 that identified the intersection of health and computation as a singular opportunity for a cross-campus initiative. A Joint Program Working Group refined the idea in 2020-2021. CPH was launched in Fall 2021 with a significant 8-figure gift to recruit 4 endowed faculty and to establish a PhD and a Designated Emphasis program.
The Program is now fully established with 3 new endowed faculty hires, a Graduate Group comprising 42 faculty across UCB and UCSF, an inaugural class of Designated Emphasis students, and open applications for our first PhD class and for an additional endowed faculty position.