Past Events

VIEW RECORDING (requires Duo VPN)

VIEW RECORDING (requires Duo VPN)

AI and ML techniques are inreasingly used in acute and critical care. But substantial challenges are persisting. This presentation offers a candid and critical reflection on over a decade of work at the intersection of artificial intelligence and perioperative/critical care medicine. Drawing from real-world case studies, including pioneering efforts in waveform-based prediction, sentiment analysis in the ICU, and reinforcement learning to guide vasopressor initiation, this talk explores the promises, pitfalls, and practicalities of deploying AI in high-stakes clinical environments. Through a mix of technical insight and personal experience, Dr. Pirracchio will highlight how data-driven tools can augment clinical judgment, the challenges of bias and generalizability, and what it takes to build trust in AI-assisted decision-making. The talk will conclude with forward-looking strategies for safely and effectively integrating AI into personalized perioperative and critical care.

Click below for recording:

Webinar recording

VIEW RECORDING (requires Duo VPN)

During this talk, Dennis P. Wall, PhD, FACMI, covered a few pressing health issues that AI is and could address, various current and future embodiments of and potentials for empathic AI, and current realities and opportunities for managing uncertainty in AI outcomes.

Click below for recording:

Webinar recording

In many health contexts, we think of the core question as what treatment or intervention to recommend to each patient (or perhaps, relatedly, determining the diagnosis). However, there's a "micro" problem hidden inside this "macro" problem: how do we maximize the efficacy of the human+AI team? The machine learning algorithms need to expose appropriate information about themselves such that the user can confirm alignment and understand the strengths and limitations of the AI. The interaction needs to be designed such that the user does not fall into their own cognitive biases such as over-relying on the machine advice. Dr. Doshi-Velez will discuss how we can design models that are both performant and inspectable, as well as steps toward personalizing the interaction to the user's specific cognitive needs. Dr. Doshi-Velez will also speak more broadly to the rich design space and opportunities when it comes to creating human+AI teams to help people achieve their health and wellness goals. 

Click below for recording:

Webinar recording

“From Data to Knowledge: Integrating Clinical and Molecular Data for Predictive Medicine"

Marina Sirota, PhD

Alzheimer’s disease (AD) remains one of the most pressing medical challenges. Recent advances in computational biology and artificial intelligence (AI) together with availability of rich molecular and clinical data, offer new opportunities to address these challenges by integrating molecular, clinical, and systems-level insights. This talk will highlight complementary approaches that illustrate the power of combining real-world clinical data, knowledge networks, and systems pharmacology to advance precision medicine for AD. This work highlights a paradigm shift toward AI-enabled, data-driven strategies that bridge molecular discovery and clinical application, ultimately informing novel therapeutic interventions and improving patient care.

“The Future of Evidence-Based Medicine is Data-Driven Medicine: Leveraging Data and AI in the Cancer Clinic"

Julian Hong, MD, MS

The integration of real-world data, artificial intelligence (AI), and clinical informatics has the opportunity to transform oncology, creating new opportunities to improve patient outcomes through data-driven precision medicine. In our recent studies, we work to realize Dr. Butte's vision of “data-driven medicine" and “scalable privilege,” building on our experience in applying artificial intelligence (AI)-directed treatment strategies to mitigate treatment-related toxicities in a randomized controlled trial. We leverage Dr. Butte’s transformative contributions to UCSF infrastructure to advance our efforts in generalizing actionable predictive clinical models, generating real world evidence, and integrating large language models (LLMs). These efforts build a realistic pathway towards realizing broadly accessible, data-driven clinical care.

Click below for recording:

Webinar recording

VIEW RECORDING (requires Duo VPN)

Dr. Sendak will synthesize key learnings from over 10+ years of work at the Duke Institute for Health Innovation (DIHI) to develop, evaluate, and implement AI in healthcare. He will describe the motivation for launching Health AI Partnership (HAIP) in 2022 and share best practices from across the AI product lifecycle surfaced by the national network of 30+ healthcare delivery organizations in the US. He will share preliminary results of the HAIP Practice Network program, the country’s first AI technical assistance program targeting safety net settings. He will conclude the presentation by highlighting the complementary value of research, education, and commercialization efforts to advance the safe, effective, and ethical use of AI in healthcare.

Click below for recording:

Webinar recording

As AI continues to transform the healthcare landscape, it is essential to balance innovation with patient safety and ethical responsibility. This talk introduced an example of using AI for chest pain patient risk stratification and explore the challenges and opportunities of integrating AI in healthcare, focusing on the importance of patient safety, the ethics of AI, and the limitations of emerging technologies. It also addressed ethical guidelines and considerations that must be followed, along with practical strategies for the safe and responsible use of AI in healthcare. The aim is to provide a comprehensive understanding of how to implement AI in ways that benefit patients while maintaining safety and ethical integrity.

Click below for recording:

Webinar recording

This educational presentation provided an overview of the AI pathology field including a primer on the main ML methods and examples of clinical and precision medicine algorithms from the current literature. Also discussed was the AI/ML methods driving the future of the field, including vision language models and their potential role in pathologist decision support.

Click below for recording:

Webinar recording

There is immense enthusiasm about the potential of large language models to support clinical and administrative workflows in healthcare. However, a barrier to effective and safe clinical translation is the lack of standardized approaches to evaluate and monitor the knowledge quality, reasoning ability, and risks of these models. In this lecture, Dr. Bitterman discussed current limitations of language model knowledge representation in the context of high-impact clinical applications. She presented research into measuring language model risks, including misinformation and logical reasoning errors, in ways that are systematic, generalizable, and clinically-relevant.

Click below for recording:

Webinar recording

DoC-IT hosted a series featuring thought leaders at the intersection of digital health, informatics, and technology. Each hybrid event brought together experts from academia, industry, and healthcare to share insights, innovations, and emerging trends that shape the future of patient care and medical research.

Click below for each recording:

Jiyeong Kim, PhD, MPH

BAIHAN LIN, PHD

ZEESHAN SYED, PHD

While AI is seen as the frontier for healthcare innovation, robust evaluations of its value remain under-described. This talk reviewed evaluations of health system-scale AI implementation in Kaiser Permanente Northern California and consider how to assess value.

Click below for recording:

Webinar recording

Dr. Poon shared exciting progress in the frontier of multimodal generative AI in precision health, where we can pretrain powerful patient embedding by harnessing multimodal, longitudinal real-world patient journeys and unlock population-scale patient-like-me reasoning and real-world evidence for optimizing patient care and accelerating biomedical discovery.

Click below for recording:

Webinar recording

The AMIA 2024 Annual Symposium brought thousands of clinical informaticians to San Francisco this year. We were delighted that so many DoC-IT colleagues were featured during presentations and sessions throughout the event. 

Click below to find resources we created to welcome visitors and Bay Area locals alike to AMIA, as well as a comprehensive list of sessions spotlighting DoC-IT experts.

Learn More 

The age-friendly health systems (AFHS) movement is one of the largest efforts to improve care delivery for older adults, with thousands of healthcare systems recognized as “Age-Friendly.” However, significant questions regarding this movement remain, including how best to implement evidence-based practices and the effect of Age-Friendly transformation on the outcomes of older adults.

Inquiry: The Journal of Health Care Organization, Provision, and Financing hosted a special issue on AFHS. This Special Collection seminar invited AFHS experts to discuss their research in this area with a focus on issues of implementation and intervention effectiveness of AFHS redesign efforts, furthering the field.

Julia Adler-Milstein, PhD, chief of DoC-IT, was among the invited speakers who shared about her ongoing research to build an evidence base for AFHS, supported by the John A. Hartford Foundation. 

Learn More View the Recording

Lisa Rotenstein, MD, MBA, MSc, assistant professor at the University of California—San Francisco and Director of the Center for Physician and Practice Excellence, joins the Vital Signs podcast to discuss and explore some possible reasons why women in medicine are struggling and more importantly, what to do about it. Her passion is ensuring the well-being of the healthcare workforce with particular focus on women in medicine. 

 LISTEN TO PODCAST

Julia Adler-Milstein, PhD, joins the Digital Health Section Podcast to discuss AI & the Limits of Clinician Vigilance. Conversation topics include: 

• Lessons from the aviation and automobile industry
• Should AI be checking clinicians' work rather than the other way around?
• Potential solutions to improve clinical vigilance ability
• The importance of considering behavior science and systems design thinking
• Is there an ethical risk of demanding superhuman performance from AI tools?

LISTEN TO PODCAST

Raman Khanna, MD, MS, program director for the Clinical Informatics Fellowship Program and professor of clinical medicine at UCSF, presented at the June Health Informatics Grand Rounds and shared his reflections and experiences using digital health tools at UCSF.

VIEW RECORDING (requires Duo VPN)

UCSF seminar series from January to June 2024, focusing on topics related to the implementation and evaluation of AI in real-world scenarios. Topics covered include:

• Generative AI for Healthcare: Panel Discussion

• Shaping the Responsible Adoption of AI in Healthcare

• Artificial Intelligence (AI) and Data Science in the Medical Imaging of COVID and CANCER: MIDRC to the Real World

• Artificial Intelligence (AI) Readiness at UCSF: Current State and Future Directions

Seminar sponsors include the Bakar Institute for Computational Health Science, Division of Clinical Informatics and Digital Transformation, UCSF and UC Berkeley Computational Precision Health Program, and the Department of Epidemiology and Biostatistics.

Learn More VIEW RECORDINGS

Julia Adler-Milstein, PhD, Andrew Auerbach, MD, and Gurpreet Dhaliwal, MD, discussed the fundamental issues in diagnosis, each seen through their specialized lenses of research, informatics, and education.

VIEW RECORDING

Presented by Julia Adler-Milstein, chief of DoC-IT, at the monthly UCSF Benioff Children's Hospital Oakland's Grand Rounds.

VIEW RECORDING

A planning committee of the National Academies of Sciences, Engineering, and Medicine (the National Academies) planned and hosted a public workshop, on April 4th and 5th, on the variety of different experiences with the U.S. healthcare system common to individuals facing barriers*, including members of racial or ethnic minorities, and the consequences of those different experiences on an individual’s health status and medical record, which is relevant to SSA in disability determinations. The workshop included presentations with a focus on how individual's different experiences can manifest in records, as well as medical advances, developments, and research related to health inequities in the United States.

Julia Adler-Milstein, PhD, presented on two panel sessions: "The Health Record in Depth" and "Approaches to Advancing Medical Records to Address Disparities in Disability Determinations."

Learn More VIEW RECORDINGS

Lucas Zier, MD, MS, associate clinical professor at the UCSF Interventional Cardiology and Pulmonary Hypertension Program and co-founder and co-director of the PROSPECT Lab at ZSFG, presented on "Utilizing Digital Automation, Machine Learning and Artificial Intelligence to Improve Health Equity and Outcomes in a Safety Net Health System."

VIEW RECORDING (requires Duo VPN)