The findings, published in npj Digital Medicine, suggest this technology could help surgeons refine their techniques and improve patient outcomes.

The AI system, called Frame-to-Outcome (F2O), analyzes video recorded during the nerve-sparing portion of robot-assisted prostate surgery, when surgeons work to preserve the nerves responsible for sexual function. Rather than relying on experts to manually evaluate each procedure, the system automatically identifies patterns in a surgeon’s movements—called “surgical gestures”—and uses them to predict patient recovery.

Prior studies of these surgical gestures demonstrated a strong relationship between the gestures performed by the surgeon—such as the sequence of instruments used or the speed of stretching nerves to move them aside—and the patient’s outcome. By analyzing the gestures used during surgery, the AI system made determinations about whether the patient will be more or less likely to have a good outcome. Until now, this type of analysis required labor-intensive review by trained human observers.

“Our goal isn’t simply to predict who will recover,” said Andrew Hung, MD, corresponding author of the study and professor of Urology at Cedars-Sinai. “We want to identify the surgical techniques with the best outcomes so surgeons can learn, refine and improve care for our future patients.”

Investigators found that F2O matched expert human reviewers in predicting patient outcomes while dramatically reducing the time required to analyze surgical performance. The technology could eventually provide surgeons with objective feedback on the techniques most closely associated with successful patient recovery.

To develop the system, researchers trained the AI system using videos annotated by human analysts from 294 surgeries from 23 surgeons across four international centers. Then they tested it on an additional 29 surgeries and found its predictions closely matched those of expert human reviewers.

“By identifying and interpreting the gestures that result in positive patient outcomes, we can offer surgeons insights that can help improve surgical performance and patient care,” Hung said.

The study was a collaboration between the Department of Urology, the Department of Computational Biomedicine and the Center for Artificial Intelligence Research and Education (CAIRE) at Cedars-Sinai.

“This work highlights what we can achieve when surgeons and computational biomedicine experts work together toward a shared clinical goal," said Jason Moore, PhD, chair of the Cedars-Sinai Department of Computational Biomedicine and director of CAIRE. "Bringing together these disciplines allowed the team to build something that is technically rigorous and genuinely meaningful for surgeons—and their patients."

Additional Cedars-Sinai authors include Xi Li, Nicholas Matsumoto, Jay Moran, Miguel E. Hernandez, Cherine Yang, Jeanine Kim, Jasmine Lin, Peter Wager, Ujjwal Pasupulety and Atharva Deo.

Other authors include Alvin C. Goh, Christian Wagner and Geoffrey A. Sonn.

Funding: Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health under Award Number R01CA273031. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Cedars-Sinai Health Sciences University is advancing groundbreaking research and educating future leaders in medicine, biomedical sciences and allied health sciences. Learn more about the university.

The model addresses a longstanding challenge in hospital care. Low blood sugar, also called hypoglycemia, is a common and potentially life-threatening complication among hospitalized patients, including those receiving diabetes treatment, those who are fasting before procedures or those in critical care. However, there are no widely used tools for predicting which hospitalized patients may develop hypoglycemia.

“Today, most hospital care for hypoglycemia is reactive, and we respond after a patient’s blood sugar drops,” said Roma Gianchandani, MD, senior author of the study and vice chair of Quality and Innovation in the Department of Medicine and program director for Diabetes.  

The AI model developed by Cedars-Sinai investigators analyzes patterns in medications, lab results, meals and other data from patients’ electronic health records. It collects the information in four-hour intervals over a five-day period and uses it to predict whether a patient will develop hypoglycemia within the next 24 hours.

Researchers developed and tested the model using data from more than 143,000 adult hospital admissions across three Cedars-Sinai Health System hospitals between 2014 and 2025. Investigators also tested the tool using prospective hospital data to confirm their initial findings.

“The AI model is designed to alert patient care teams before a patient experiences low blood sugar and identify the key factors driving that risk,” said Amanda Momenzadeh, PharmD, lead author of the study and a project scientist in the Meyer Research Lab at Cedars-Sinai. “By offering actionable insights to care teams, it also aims to support hospital diabetes management programs.”

Researchers estimate the tool could help prevent about three to four cases of low blood sugar at a large hospital each day. Extrapolating across all hospital beds worldwide, the impact could be substantial.

“What’s exciting is that this isn’t just a theoretical model, but instead, it is built and validated to work prospectively in real time using data hospitals already collect,” said senior author of the study Jesse Meyer, PhD, assistant professor in the Department of Computational Biomedicine at Cedars-Sinai. “By identifying patients at risk earlier, we have an opportunity to reduce preventable complications and improve patient safety.”

If widely adopted, the model could lead to more proactive, data-driven care for hospitalized patients with diabetes and other conditions that affect blood sugar.

Additional Cedars-Sinai authors: Caleb Cranney, Dennis Chen, and Elizabeth Nguyen.

Funding: NIGMS R35GM142502, NIH National Center for Advancing Translational Science (NCATS), and UCLA CTSI Grant Number UL1TR001881

Disclosure: Jesse Meyer, Amanda Momenzadeh, and Caleb Cranney are listed as inventors on a patent application related to this AI tool.

Cedars-Sinai Health Sciences University is advancing groundbreaking research and educating future leaders in medicine, biomedical sciences and allied health sciences. Learn more about the university.

But is agentic AI smart enough to replace actual scientists?

The Cedars-Sinai Newsroom sat down for a conversation with Jason Moore, PhD, chair of the Department of Computational Biomedicine at Cedars-Sinai, to tackle the pluses and minuses of agentic AI. Moore is corresponding author of a new paper, published in Nature Biotechnology, that examines where agentic AI is today and where it is headed.

What is agentic AI? Why is it called ‘in silico team science’?

Conducting biomedical research requires a team of specialists with expertise in different aspects of the medical issue being studied, from physiology to data collection and analysis, study design and writing. 

Agentic AI replicates this approach in a computer (in silico) by coordinating the activities of a “team” of AI solutions dedicated to completing specific tasks.

How does agentic AI help biomedical scientists?

As a researcher, I have a lot more ideas than I can actually pursue in my laboratory. Agentic AI is opening the door for me to explore more scientific questions than I otherwise could. It has allowed researchers in my lab to complete complex software-engineering and computer-programming projects in days rather than months, and we're seeing mind-boggling levels of productivity and efficiency.

These benefits are coming along at a particularly useful time. As the healthcare industry faces rising costs and reduced reimbursements, agentic AI can help labs be more efficient and survive with smaller teams.

Does this mean agentic AI will replace human scientists?

There's a lot of discussion about this in the AI community. I don't yet have confidence that AI can fully replace anybody in my research lab—and maybe it never will.

Humans do many things that AI may not be good at, such as managing people, displaying emotional empathy, and coming up with new hypotheses and creative solutions to problems. Those things are important if you want a research lab that functions well.

Trust is also a barrier. We know how to trust a human collaborator, but with potentially dozens of AI agents doing very complex things very rapidly, how do you know what they're doing? How can you be sure that what they've done is accurate?

From a broader perspective, how do we design ethical guardrails that ensure we put the human subjects of our scientific research first? And how do we find an environmentally sound way to provide the enormous amount of energy to fuel the computing power that AI requires?

Given these challenges, what do you view as the future of agentic AI?

The genie's out of the bottle. This technology is here, and it’s going to affect absolutely everything we do in our professional and personal lives. It is going to turn things upside down.

The exciting thing for biomedical science is that agentic AI will allow each person to be 10 or 20 or 100 times more efficient. And assigning tasks to AI allows us to focus more on the human skills, the creativity and the emotional side of what we do. It has the potential to accelerate scientific discoveries and the translation of those discoveries into better healthcare practices.

I try not to make specific predictions because all of this is moving so quickly and unpredictably. But I think the one prediction I can make about agentic AI is that everything's going to be different a year—or even six months—from now.

Cedars-Sinai Health Sciences University is advancing groundbreaking research and educating future leaders in medicine, biomedical sciences and allied health sciences. Learn more about the university.

Using a powerful technology called single-cell proteomics, these detectives track molecules called proteins, cell by cell, to shed new light on how the human body works and how diseases develop. Their tenacity has helped make Cedars-Sinai one of the world’s leading institutions in this rapidly evolving frontier of medical science.

Proteomics, the study of the complete set of proteins expressed by an organism, provides a powerful window on what is happening in the human body. That’s because these hard-working molecules carry out nearly every bodily activity.

The challenge is that a typical laboratory sample of human tissue contains thousands of cells, and each cell is packed with multiple copies of thousands of different proteins.

To connect each protein to the correct cell, single-cell proteomics investigators rely on equipment called mass spectrometers. These machines, which are like highly sophisticated scales, use magnetic forces to separate proteins by their molecular weight or mass. With the proteins sorted, investigators can identify them and match them to the cells that contained them.

Using mass spectrometry, Cedars-Sinai investigators have uncovered new types of heart cells. They are unlocking secrets of how our arteries work. And they are developing proteomics technology that can analyze thousands of cells in minutes instead of hours or days.

More discoveries are on the way. At the Cedars-Sinai Board of Governors Innovation Center, mass spectrometers operate around the clock, seeking, sorting and scrutinizing proteins. The machines are part of the center’s Alfred E. Mann Single Cell Precision Medicine Center.

But it’s really the people, not the machines, who are most responsible for Cedars-Sinai’s leading role in this nascent discipline. Below are profiles of three investigators who rank among the top international experts in single-cell proteomics.

The Visionary

It is impossible to conduct a serious discussion of proteomics without mentioning Jennifer Van Eyk, PhD, who helped pioneer this discipline and recently served as president of the international Human Proteome Organization, the field’s premier scientific association. In 2024 she was listed by The Analytical Scientist as one of the world’s 20 most impactful analytical scientists in human health.

“I started working in proteomics before the word ‘proteome’ was coined in the 1990s,” Van Eyk said in a recent interview. “At that time, I was among just a few scientists around the world who were trying to accurately measure proteins on a large scale.”

Van Eyk’s passion is clinical proteomics, which applies scientific discoveries to patient care. At Cedars-Sinai, she directs the Smidt Heart Institute’s Advanced Clinical Biosystems Research Institute, which she founded in 2014 to foster collaboration among scientists, physicians and biotechnology companies.

One of Van Eyk’s most recent achievements was to co-lead a study revealing that cardiomyocytes, the muscle cells of the heart, are not all identical. Using single-cell proteomics, she and her colleagues discovered two new hybrids of cardiomyocytes that produce both heart- and neuron-related proteins. The team is now exploring whether gender differences in cardiomyocytes could affect how a person responds to medications.

Given that drugs generally target proteins, Van Eyk sees single-cell proteomics as a critical new tool for troubleshooting disease treatments and testing new ones.

“Suppose you have a drug that works 50% of the time,” she said. “Does that mean it's working 50% in every cell, or is it working 100% in 50% of the cells? The answer is important because it allows you to fix the problem. Single-cell proteomics can provide that information.”

As director of Basic Science Research in the Barbra Streisand Women’s Heart Center and the Erika J. Glazer Chair in Women’s Heart Health, Van Eyk has a special focus on cardiology. But she also collaborates on studies of conditions as diverse as pulmonary hypertension, breast cancer and amyotrophic lateral sclerosis, also known as ALS.

Van Eyk envisions a bright future for single-cell proteomics.

“We're finding such unexpected things that you couldn't even have thought about before using these methodologies,” she said. “And the discoveries will continue until we've done enough, and then we'll do the next breakthrough in the technology.”

The Racer

Before she was a scientist, Sarah Parker, PhD, was an aspiring Olympic speed skater. Having narrowly missed that goal in 2002, she now applies the Olympic motto of “Faster, Higher, Stronger – Together” to proteomics.

The together aspect is important. Parker, an associate professor of Cardiology and Biomedical Sciences, co-directs the Proteomics and Metabolomics Core at Cedars-Sinai with Van Eyk. During her decade-long career at Cedars-Sinai, she has been a versatile team player while heading her own laboratory that has produced breakthrough studies in atherosclerosis, cancer and proteomics applications.

Parker is also striving to make single-cell technology faster and stronger through higher volume. With colleagues, she is perfecting so-called “high throughput” techniques that enable a mass spectrometer to analyze proteins in tissues from multiple people. The bigger the sample, the better the chance of finding rare cells and discovering something new about the body.

“The challenge is that to perform this analysis, you need to quickly turn and burn through a lot of cells in a reasonable amount of time,” Parker said.

In 2023, she co-led an influential Cedars-Sinai study that devised a novel solution: Get rid of the dead time between loading one sample and acquiring the data from a subsequent sample by toggling back and forth between the two processes. Using this system, a mass spectrometer can identify more than 1,000 proteins in individual cells in 15 minutes, allowing nearly 100 cells to be measured each day. This rate was double the industry standard at the time.

In current research funded by the National Institutes of Health, Parker is using single-cell proteomics to investigate how hormones influence aortic aneurysms, the bulges in the wall of the main artery from the heart that can rupture, with life-threatening results. The potential clinical application is to design better drug treatments for both males and females with this serious condition.

Parker’s longtime interest in the cardiovascular system grew from courses she took as a college student and athlete preparing for a career in sports psychology.

After retiring from professional sports, Parker learned about proteomics while earning a PhD in physiology at the Medical College of Wisconsin in Milwaukee. As a postdoctoral fellow at Johns Hopkins University in Baltimore, she was mentored by Van Eyk, who later moved her laboratory to Cedars-Sinai and encouraged Parker to join her there.

“I really liked everything about Cedars-Sinai,” Parker said, including the collaborative ethos. “On the proteomics team, we’re on most of each other’s papers—not all of them, but most of them.”

The Data Scientist

To Jesse Meyer, PhD, an assistant professor in the Department of Computational Biomedicine at Cedars-Sinai, health comes down to one number: age.

“Aging is the biggest predictor of most diseases,” he said. “If we can deeply understand aging, then we can potentially delay the onset of many diseases at once instead of spending so much energy targeting each disease separately.”

The quest for that knowledge led Meyer to proteomics and data science, an interdisciplinary field that uses statistics, computer science and mathematics to uncover meaningful patterns in large sets of data.

“What is so cool about proteomics is that proteins are tiny machines in your cells, that we have so many of them and that they are so diverse,” he said.

For data scientists like Meyer, quantifying these tiny machines is a labor of love.

Meyer first encountered protein biochemistry and mass spectrometry as an undergraduate studying plants at the University of Minnesota in Minneapolis. After completing his PhD in chemistry and biochemistry at the University of California, San Diego, he decided to use his skills to help people. That led him to take a postdoctoral research fellowship at the Buck Institute for Research on Aging in Novato, California, where he learned to apply proteomics, and the study of small molecules called metabolites in cells and tissues, to problems related to aging.

Meyer established his laboratory in 2020 and relocated to Cedars-Sinai in 2022, where his research includes developing informatics tools for proteomics, optimizing single-cell workflows, and applying these approaches to single-cell and single muscle fiber proteomics studies of muscle aging.

Based on his achievements in proteomics at such an early career stage, the US Human Proteome Organization presented Meyer with the 2025 Robert J. Cotter New Investigator Award. That year’s advances by Meyer included co-leading the creation of a user-friendly web platform for analyzing mass-spectrometry data that facilitates sharing among multiple collaborators.

In a recent publication, in the peer-reviewed journal Cell Genomics, Meyer and his colleague, project scientist Amanda Momenzadeh, PharmD, offer an overview of the current state of single-cell proteomics. Despite the field’s many challenges, they conclude that single-cell proteomics is poised to transform our understanding of biological complexity.

At this critical moment, you would need a crystal ball to foretell the future of this dynamic science. But judging from their track records, Cedars-Sinai investigators are likely to be at the forefront of the next big innovation.

Cedars-Sinai Health Sciences University is advancing groundbreaking research and educating future leaders in medicine, biomedical sciences and allied health sciences. Learn more about the university.

More than 30% of clinical trials fail due to adverse drug reactions, and the up to $5,054,235.00 contract award by the Advanced Research Projects Agency for Health (ARPA-H) Computational ADME-Tox and Physiology Analysis for Safer Therapeutics (CATALYST) program, will address this longstanding challenge in drug development.

“Each year, many promising drugs fail in trials because animal tests and short-term lab studies cannot predict how medicines behave in real people over time,” said Nicholas Tatonetti, PhD, vice chair of Computational Biomedicine at Cedars-Sinai and the project's lead investigator. “These failures delay lifesaving treatments and drive up drug development costs.”

The new platform, called KronosRx, aims to reduce these failures by applying AI tools to “patient avatars”—sophisticated organoids and organ-on-chip systems derived from human stem cells—to help investigators predict drug toxicity that might otherwise harm clinical trial participants.

The avatars use tiny numbers of cells to mimic the function of whole organs and their immediate response to experimental medications. The AI models in the platform are trained using millions of anonymous patient data points from Cedars-Sinai’s extensive electronic health record network. The resulting platform can forecast an organ’s response to a medication over time—and across the diverse population of patients reflected in the Cedars-Sinai data.

“These AI systems don’t just predict whether a drug is safe or toxic; they model how risk evolves dynamically, accounting for age, a patient’s health, and other medications they might be taking,” Tatonetti said.

Investigators hope this approach will allow better predictive modeling that can evolve over time, reducing reliance on animal studies and improving safety for all patients.

“By creating a more reliable and human-relevant method for safety assessment, the KronosRx project aims to improve clinical trials and to shorten development timelines,” said Clive Svendsen, PhD, executive director of the Cedars-Sinai Board of Governors Regenerative Medicine Institute and an investigator on the KronosRx project.

The Cedars-Sinai KronosRx team includes leaders in computational biomedical innovation, stem cell biology and health informatics.

Tatonetti is leading project integration using biomedical data science and AI-driven drug discovery methods. Svendsen is applying induced pluripotent stem cells and organ chip technologies to better understand how common drugs may cause rare neurological side effects.

Arun Sharma, PhD, director of the Cedars-Sinai Center for Space Medicine Research in the Board of Governors Regenerative Medicine Institute, is using patient-specific cardiac organoid and organ chip systems to assess drug-induced cardiotoxicity. Graciela Gonzalez-Hernandez, PhD, professor and vice chair for Research and Education in the Department of Computational Biomedicine, is advancing the project’s AI and unstructured text data integration to connect molecular and clinical phenotypes.

The ultimate goal, Svendsen said, is to make critical treatments available to patients sooner.

“This approach allows AI to continually refine its forecasts as new evidence emerges, bridging the gap between computational prediction and real-world patient outcomes,” Svendsen said.

Cedars-Sinai Health Sciences University is advancing groundbreaking research and educating future leaders in medicine, biomedical sciences and allied health sciences. Learn more about the university.

Synthetic data platforms use artificial intelligence to produce new data that mimics real-world patient data, without disclosing any private information. This approach allows organizations like Cedars-Sinai to simulate various scenarios and outcomes, providing valuable insights for medical research and clinical decision-making, while still maintaining a high level of accuracy and validity in research findings.

For example, Cedars-Sinai can take real patient data and convert it into a new dataset that reflects patient profiles and treatment scenarios. This synthetic dataset could be generated within one hour, offering a significant advantage in speed and efficiency over traditional methods that require time-consuming processes to retrieve real patient data.

“The use of synthetic data at Cedars-Sinai reflects our pursuit of cutting-edge technologies to advance medical research and improve patient care,” said Craig Kwiatkowski, PharmD, senior vice president and chief information officer at Cedars-Sinai. “This supports our broader strategy of building a more connected data ecosystem—one that gives our teams easier access to a wider range of data that helps drive real-world innovation in healthcare.”

To conduct this work, Cedars-Sinai is partnering with Syntho, an Amsterdam-based company that participated in the Cedars-Sinai Accelerator program in 2022. Syntho provides artificial intelligence-based, privacy-enhancing technology to generate anonymous synthetic data.

Cedars-Sinai also is exploring the use of synthetic data in tandem with the recent launch of a new Digital Innovation Platform, as it develops a comprehensive data platform tool set to address some of the most pressing challenges in healthcare. The initiative will leverage Cedars-Sinai’s clinical expertise, infrastructure, research capabilities and vast data resources to develop companies that tackle healthcare issues, in partnership with Cedars-Sinai staff, investors and venture-builder Redesign Health.

Jason Moore, PhD, chair of the Department of Computational Biomedicine, and Nicholas Tatonetti, PhD, vice chair of Computational Biomedicine, are leading research efforts on synthetic data at Cedars-Sinai. They sat down with the Cedars-Sinai Newsroom to share more about the role of synthetic data in a healthcare setting.

What are the benefits of using synthetic data?

Moore: Synthetic data is generated using artificial intelligence to model various patterns in real data and produce new data that preserves these patterns. This synthetic data does not have the same privacy and security issues as real data, making it easier—and much faster—to use for research. That’s because synthetic data doesn’t require approval from committees or governing bodies like an internal review board.

Synthetic data also makes it easier to collaborate and share information with other internal teams and external institutions.

Tatonetti: The speed, accuracy and sheer volume in which we can access synthetic data opens the door to studying new and complex conditions like rare diseases.  

What is the difference between de-identified data and synthetic data?  

Moore: De-identified data is real data with patient identifiers removed, while synthetic data is completely artificial. This artificial data is generated to preserve the relationships and patterns in the original data, making it useful for research without the privacy concerns associated with real data.

What about the privacy and security of synthetic data?   

Moore: One of the biggest motivators of using synthetic data is to ensure patient privacy and security. Synthetic data eliminates the possibility of re-identifying patients, thus allowing researchers to work with data without the same restrictions as real data.

How will the partnership with Syntho advance our work with synthetic data?

Tatonetti: Through our partnership with Syntho, we hope to achieve three things:

• Lower the barriers to clinical research, allowing more investigators to conduct studies without lengthy approval processes associated with real data
• Speed up the process of launching and dropping studies, enabling researchers to quickly test and iterate on their hypotheses
• Allow students and trainees in the Cedars-Sinai Health Sciences University to use synthetic data, providing them with realistic datasets to learn from, build tools on, and conduct analyses.

What are the limitations of synthetic data?

Moore: Synthetic data has limitations and does not handle all data types well—like discrete genetic data—so it’s imperative we understand these limitations in our workflows. It’s also critical we effectively communicate these limitations to our users to prevent user frustration and fatigue.

Cedars-Sinai Health Sciences University is advancing groundbreaking research and educating future leaders in medicine, biomedical sciences and allied health sciences. Learn more about the university.

In the first study, Cedars-Sinai investigators applied advanced statistical techniques to analyze electronic health records from nearly 100,000 hospital stays. This approach identified drugs that were unexpectedly associated with raising or lowering blood sugar levels of hospitalized patients.

“Our findings offer practical insights to help clinicians anticipate and manage medication-related blood sugar changes, ultimately improving glycemic safety for patients in hospitals,” said Jesse G. Meyer, PhD, assistant professor of Computational Biomedicine at Cedars-Sinai and corresponding author of the study.

In the second study, co-led by Cedars-Sinai, investigators developed a secure method to pool patient data from multiple hospitals for research studies. This method allows hospitals to send statistical summaries of their patients’ characteristics, rather than the healthcare data of individuals, to a central location for analysis by investigators, reducing the risk of inadvertent disclosure of sensitive patient information.

“Our innovative approach opens the door for larger, more diverse studies that better protect patient privacy, improve research quality and support the development of more effective treatments,” said Ruowang Li, PhD, assistant professor of Computational Biomedicine at Cedars-Sinai and co-corresponding author of the study.

“Both studies emphasize our unique approach to using machine learning and big data in academic medicine,” said Jason Moore, PhD, professor and chair of the Department of Computational Biomedicine at Cedars-Sinai and co-corresponding author of the study. “These studies foster collaboration, ultimately leading to patient care and research that are driven by data, overcoming gaps in outcomes and creating healthier lives.”

First study:

Other Cedars-Sinai authors include: Amanda Momenzadeh, PharmD, Caleb Cranney, MS, So Yung Choi, MS, Catherine Bresee, MS, Mourad Tighiouart, PhD, Roma Gianchandani, MD, Joshua Pevnick, MD, MSHS, and Jason H. Moore, PhD.

Acknowledgments: The authors thank Edward Kowalewski, Kevin Japardi and the Honest Enterprise Research Broker for EHR data extraction services. This research was supported by NIH National Center for Advancing Translational Science (NCATS), UCLA CTSI Grant Number UL1TR001881.

Declaration of interests: Amanda Momenzadeh and Jesse G. Meyer have a provisional patent related to this work. Jason H. Moore is a member of Patterns Advisory Board.

Second study:

The other Cedars-Sinai author was Jason H. Moore. Other authors include Luke Benz, Rui Duan, Joshua C. Denny, Hakon Hakonarson, Jonathan D. Mosley, Jordan W. Smoller, Wei-Qi Wei, Thomas Lumley, Marylyn D. Ritchie and Yong Chen (co-corresponding author).

Acknowledgment: Funding sources included NIH R01 LM010098, AG066833, GM148494, LM014344, LM012607, LM013519, AI130460, AG073435, RF1AG077820, R56AG069880, R56AG074604, U01TR003709, R21AI167418 and R21EY034179. MDR was funded by R01HG010067 and R01HL169458.

eMERGE Network (Phase III). This phase of the eMERGE Network was initiated and funded by the NHGRI through the following grants: U01HG8657 (Group Health Cooperative/University of Washington); U01HG8685 (Brigham and Women’s Hospital); U01HG8672 (Vanderbilt University Medical Center); U01HG8666 (Cincinnati Children’s Hospital Medical Center); U01HG6379 (Mayo Clinic); U01HG8679 (Geisinger Clinic); U01HG8680 (Columbia University Health Sciences); U01HG8684 (Children’s Hospital of Philadelphia); U01HG8673 (Northwestern University); U01HG8701 (Vanderbilt University Medical Center serving as the Coordinating Center); U01HG8676 (Partners Healthcare/Broad Institute); and U01HG8664 (Baylor College of Medicine).

UK Biobank. All data for this cohort pertained to project 32133 – “Integration of multi-organ imaging phenotypes, clinical phenotypes, and genomic data.”

Declaration of interests: Jason H. Moore serves as a current member of the Patterns advisory board. The other authors report no competing interests.

Cedars-Sinai Health Sciences University is advancing groundbreaking research and educating future leaders in medicine, biomedical sciences and allied health sciences. Learn more about the university.

The hottest trend on the horizon for artificial intelligence (AI) is agentic AI, according to

Unlike traditional AI that is primarily designed to complete a single task, agentic AI is a new generation of AI models that can independently perform multiple tasks simultaneously to achieve specific objectives.

At Cedars-Sinai, Moore and colleagues are immersed in agentic AI models, creating algorithms that make faster and more accurate decisions while combing through large datasets. Moore, professor of Computational Biomedicine and Medicine, sat down with the Cedars-Sinai Newsroom to explain the potential—and recent boom—in the use of agentic AI.

What is agentic AI and how does it differ from existing AI models?

For the past 10 years we have been developing state-of-the-art AI methods, including deep-learning algorithms and large language models for natural language processing. These methods have been designed to complete one specific task—for example, to analyze an echocardiogram image of the heart to find defects.

Agentic AI, however, assembles teams of AI specialists to complete specific tasks—then collectively brings these teams together to solve a complex problem. Using the same echocardiogram example, an agentic AI model can simultaneously analyze echocardiogram images, laboratory tests, vital signs, medication history and clinical notes to provide a comprehensive picture of a patient in a fraction of the time it would take multiple clinicians to review results.

This technique mirrors the way humans solve complex problems.  

What makes agentic AI the next hottest trend in AI?

ChatGPT has shown we can use powerful algorithms for specific tasks. With agentic AI, algorithms are tailored to specific needs and adapts strategies independently to achieve predefined goals. It can assemble teams of AI agents to handle various tasks.

In my laboratory, for example, we work with big data. So, we need people whose expertise is in cleaning data, preparing it for analysis, building computational models with the data and providing statistical analysis. We also need people who can interpret the data for us; what does the data tell us about biology, clinical care, etc.? We then need someone to summarize all of these results in written form, then prepare graphs and figures to communicate these results.

Agentic AI builds teams of AI agents that handle each of these respective areas, with the end goal of providing understanding of the data and explanation of the results.

The field is advancing in a way that individuals may soon use these methods at home. I expect to see many tools coming out in the next year or so that will make our lives easier. One can imagine an AI agent helping prepare your taxes, your family budget, or preparing your weekly grocery list.

Is there published research happening in agentic AI?

Yes, we are seeing an uptick in published research studies involving agentic AI. Our laboratory recently published a study in Bioinformatics about an agentic AI model we created called ESCARGOT (Enhanced Strategy and Cypher-driven Analysis and Reasoning using Graph Of Thoughts).

The ESCARGOT model combines large language models with a dynamic “graph of thoughts” and biomedical knowledge graphs—an approach that was shown to improve output reliability and reduce inaccuracies. To do this, we inputted existing data we have procured about Alzheimer’s disease, then asked the agentic AI model to provide several things: genes associated with the disease, drugs and therapies that may offer the best treatments for these genetic variations, etc.

We compared these findings to the responses ChatGPT produced and, not shockingly, agentic AI provided answers with 80%-90% accuracy, compared to ChatGPT, which scored about 50%.

We believe strongly in making our models open-access to ensure science progresses. The ESCARGOT model is public, free and available on GitHub.

Cedars-Sinai Health Sciences University is advancing groundbreaking research and educating future leaders in medicine, biomedical sciences and allied health sciences. Learn more about the university.

The findings, published in the peer-reviewed journal Med, demonstrate how AI can improve drug safety, support drug discovery and improve understanding of medication risks.

The database is called OnSIDES (ON-label SIDE effectS resource). It is free and publicly available on GitHub.

“OnSIDES provides the most comprehensive and up-to-date database of adverse drug events from drug labels,” said Nicholas Tatonetti, PhD, vice chair of Computational Biomedicine at Cedars-Sinai and corresponding author of the study. “This work enables researchers and clinicians to systematically study drug safety.” 

Adverse drug events are unintended, harmful events related to the usage of medication and are the fifth leading cause of death internationally. Experts believe half of all adverse drug events are preventable.

“While many drug safety studies are conducted on individual medications during clinical trials and through post-marketing surveillance programs, far fewer studies have studied the occurrence of adverse drug events more broadly,” said Tatonetti, also the associate director for Computational Oncology at Cedars-Sinai Cancer. “The lack of broadscale studies may be attributed in part to the array of medications and the complexity of drug interactions, as well as the lack of standardized data publicly available.”

The OnSIDES model analyzed 3,233 unique drug ingredient combinations extracted from 47,211 labels and identified over 3.6 million pairs of medications and adverse drug events. This work was also expanded to labels from countries outside of the U.S., revealing differences in how adverse drug events are reported internationally.

By using artificial intelligence to extract adverse drug events from drug labels, investigators improved access to structured, machine-readable data, ultimately making it easier to identify drug risks, predict new drug uses and enhance patient safety.

“This resource supports drug repurposing, pharmacovigilance, and AI-driven drug discovery,” said Jason Moore, PhD, chair of the Department of Computational Biomedicine at Cedars-Sinai. “We are hopeful that future research can build on OnSIDES to develop better predictive models, personalized medicine approaches and regulatory insights, ultimately leading to safer medications and more informed clinical decision-making worldwide.”

Additional Cedars-Sinai authors include Apoorva Srinivasan, Michael Zietz, Gaurav Sirdeshmukh and Jacob Berkowitz.

Additional authors include Yutaro Tanaka, Hsin Yi Chen, Pietro Belloni, Undina Gisladottir, Jenna Kefeli, Jason Patterson and Kathleen LaRow Brown.

Funding: R35GM131905 to N.P.T, T32GM145440 to H.Y.C, T15LM007079 to U.G, M.Z, K.L.B.

Cedars-Sinai Health Sciences University is advancing groundbreaking research and educating future leaders in medicine, biomedical sciences and allied health sciences. Learn more about the university.

“Pathologists traditionally identify relevant structures in diseased tissue by visually examining stained tissue images—a method that is costly, time-consuming and typically reliant on a single type of tissue imaging,” said Kyle Coleman, PhD, an investigator in the Department of Computational Biomedicine at Cedars-Sinai, who led the team that developed the MISO tool. “MISO streamlines this process by integrating detailed molecular information with tissue histology, enabling automated and precise identification of disease-relevant structures.”  

The MISO tool can, for example, distinguish regions with different levels of cancer severity within a single colon cancer tissue sample. Additionally, by integrating transcriptomics, metabolomics, and histology imaging, MISO accurately maps detailed structures of the mouse hippocampus at a high resolution. Investigators hope that MISO can lead to a deeper understanding of disease processes, potentially advancing the development of new therapies.

The MISO software is currently available on GitHub.

Additional authors include Amelia Schroeder, Melanie Loth, Daiwei Zhang, Jeong Hwan Park, Ji-Youn

Sung, Niklas Blank, Alexis J. Cowan, Xuyu Qian, Jianfeng Chen, Jiahui Jiang, Hanying Yan, Laith Z. Samarah, Jean R. Clemenceau, Inyeop Jang, Minji Kim, Isabel Barnfather, Joshua D. Rabinowitz, Yanxiang Deng, Edward B. Lee, Alexander Lazar, Jianjun Gao, Emma E. Furth, Tae Hyun Hwang, Linghua Wang, Christoph A. Thaiss, Jian Hu and Mingyao Li.

Funding: This research was supported by National Institutes of Health grants R01HG013185 and R01LM014592.

Follow Cedars-Sinai Academic Medicine on LinkedIn for more on the latest basic science and clinical research from Cedars-Sinai.

A recent opinion column published in the Cell Press journal Patterns lays out what the investigators see as the challenges and opportunities for those who identify as lesbian, gay, bisexual, transgender and queer or questioning (LGBTQ+) to increase their representation in these fields. The column emphasizes the importance of addressing the biases and erasure of gender and sexual diversity in data and computational models.

“Our team works to improve patients’ lives by using computers, computing technologies and data resources; insights from a diverse community are invaluable in helping us do that,” said Jason Moore, PhD, professor and chair of the Department of Computational Biomedicine and an investigator who helped author the opinion article.

Moore said it’s not only a matter of tolerance and acceptance: “LGBTQ+ inclusion is necessary to enhance scientific problem-solving, generate more equitable knowledge and address the health disparities and needs of diverse communities. Not doing so can impact computer models and all patients, potentially leading to harmful conclusions.”

The column’s authors propose strategies and resources for LGBTQ+ inclusion, such as improving data collection and model evaluation, revising policies, creating gender- and sexual diversity-inclusive curricula, and fostering allyship and support networks. They provide examples of conferences, organizations and programs that promote LGBTQ+ participation and leadership in STEM and AI, and they acknowledge the contributions of LGBTQ+ scientists in the history and development of STEM and AI.

To learn more about the team’s perspectives, the Cedars-Sinai Newsroom recently talked with two of the paper’s authors—Pei-Chen Peng, PhD, an assistant professor in the Department of Computational Biomedicine, and Ryan Urbanowicz, PhD, a research assistant professor and director of Cedars-Sinai’s National AI Campus, a U.S.-wide AI and machine learning collaborative and project-based initiative. 

Why was it important to bring this topic to the forefront?

Urbanowicz: Considering the ever-increasing use of AI in clinical care and research, it seemed timely to encourage open discussions about the unique challenges of the LGBTQ+ community and how they overlap with STEM and AI, both from the perspective of encouraging and supporting LGBTQ+ scientific trainees, and in how to tackle the challenges and considerations of LGBTQ+ in STEM and AI research.

Peng: When the Patterns journal invited us to submit an opinion article related to the experiences of queer scientists for its June issue, we saw this as a great opportunity to further highlight the experiences of the LGBTQ+ community and to advocate for queer scientists. As we wrote in the article, LGBTQ+ scientists are underrepresented in the biomedical AI and STEM community, which can cost the scientific community great minds and stifle innovation.

Why is it critical to have LGBTQ+ inclusion in STEM education?

Peng: The awareness of gender and sex diversity in STEM education inevitably influences clinical research methods. Inclusion of LGBTQ+ ensures that the STEM community is reflective of the broader society and reduces biases when we do biomedical and AI research.

Urbanowicz: I think that fostering a diverse community of STEM trainees ensures a more diverse set of perspectives and ideas, essential to our research community’s ability to do its best work as a whole.

Is there an area of opportunity that you are particularly passionate about?

Urbanowicz: Personally, I want to do my part as an LGBTQ+ scientist to be a visible and accessible example/role model to other LGBTQ+ trainees in STEM and AI research. Also, I want to bring awareness to the current limitations of AI, specifically with regard to potential disparity.

Peng: I’m excited about the potential for positive changes. We’ve highlighted the importance of inclusivity and representation, addressing how diverse perspectives can lead to more innovative and equitable scientific advancements, and we encourage all scientists to take initiative in advancing our understanding of LGBTQ+ health issues. I’m also proud that we were able to provide a catalog of existing educational and professional resources that the community can refer to as needed. For example, trainees and researchers will know which conferences are inclusive and promote diversity, and they can choose venues where they can present their research findings confidently.

What are some ways that AI models can ensure health equity for LGBTQ+ people?

Urbanowicz: When it comes to both medical and AI research, awareness of what makes the LGBTQ+ community unique and taking these considerations into account when it comes to data collection, study design and research resources should all improve overall health equity.

Based on the opportunities identified in your article, how can Cedars-Sinai be more inclusive?

Peng: Our National AI Campus is one area that comes to mind. It’s a training program that aims to make AI accessible to a diverse community with training experiences that focus on biomedical projects. Beyond promoting participants’ diversity, the program also could address LGBTQ+ issues by developing or inviting AI and/or machine learning projects that tackle LGBTQ+ health and analytical challenges.

Urbanowicz: I agree with Dr. Peng. National AI Campus is founded on principles of inclusion and accessibility for all students interested in learning about machine learning and artificial intelligence. In the future, in addition to introducing projects that tackle LGBTQ+ issues, we can encourage more LGBTQ+ participants to get involved and highlight the importance of these issues to trainees at large.

Read more on the Cedars-Sinai Blog: A Place at the Table to Shape Cancer Research

The study detailed how the tool, a free and publicly available software platform, analyzes and compiles data and information—including new peer-reviewed studies—to answer researchers’ queries. The key to the tool’s success is a new type of large language model, said Jason H. Moore, PhD, professor and chair of the Department of Computational Biomedicine at Cedars-Sinai and senior and corresponding author of the study.

Large language models are a specific type of AI programs that can distill large amounts of data—like medical studies, books, articles and interviews—and use that data to create new content.

“The large language model approach we developed uses knowledge stored in a special database, called a knowledge graph, that specializes in capturing the relationships between entities such as drugs and genes,” Moore said.

Historically, the main challenge in using large language models to generate content is ensuring quality, accuracy and reliability of the generated responses.

The Cedars-Sinai technique, however, moved past this challenge by using the graph-of-thoughts technique—a framework that allows investigators to break down a problem into subproblems, and turn the information generated by the large language models into a visual graph.  

The Cedars-Sinai tool also incorporates retrieval augmented generation, or RAG, which augments large language models with external data sources that provide relevant facts and context. Together, this powerful tool unearths efficient and accurate data and information about varying conditions and diseases, including Alzheimer’s disease, which was the focus of the research study published in Bioinformatics.

The open-source software, called Knowledge Retrieval Augmented Generation ENgine—or KRAGEN—is publicly available on GitHub, a cloud-based platform that helps developers collaborate and manage code. To date, the software has received more than 400 endorsements from users.

To demonstrate the usability of the database, Moore and team used KRAGEN to generate data on Alzheimer’s disease, including data on genes, drugs and other aspects related to the condition. Investigators asked the database questions like, “What drugs bind to the proteins APOE and PTAU?” And “Which are genes associated with Alzheimer’s disease?”

Instead of receiving a list of data points for their question, investigators received a synthesized summary of information.

“This AI approach is a step toward fully automating the analysis of Alzheimer’s disease data by incorporating knowledge generated from previous biomedical research studies,” Moore said.

As a next step, investigators are working on ways to integrate KRAGEN into Cedars-Sinai’s AI software for automated machine learning analysis of complex biomedical data.

“This advance is a big step toward allowing users to issue spoken commands to perform analyses in minutes, that otherwise could take weeks or months,” said Craig Kwiatkowski, PharmD, senior vice president and chief information officer at Cedars-Sinai, who was not involved in the study. “It’s encouraging to see the potential of this tool to impact AI-driven programs at Cedars-Sinai.”

Other authors involved in the study include Nicholas Matsumoto, Jay Moran, Hyunjun Choi, Miguel E. Hernandez, Mythreye Venkatesan, and Paul Wang.

This work is supported in part by funds from the Center for AI Research and Education at Cedars-Sinai Medical Center and grants from the National Institutes of Health USA (U01 AG066833 and R01 LM010098).

 Follow Cedars-Sinai Academic Medicine on X for more on the latest basic science and clinical research from Cedars-Sinai.

Jason Moore, PhD, an expert in artificial intelligence and professor and chair of the Department of Computational Biomedicine, sat down with the Cedars-Sinai Newsroom to discuss how the team draws on applied mathematics, bioengineering, biomedical informatics, biostatistics and computer science to answer biomedical and clinical research questions.

How do you define computational biomedicine?

Put simply, those who work in computational biomedicine at Cedars-Sinai are interested in improving the lives of our patients by using computers and computing technologies along with data resources.

To go a level deeper, our department uses leading-edge and state-of-the-art methods and algorithms in AI and machine learning for the analysis of clinical data. We partner with clinicians to embark on research projects, and we participate in clinical rotations to ensure our work is most meaningful to the patients and community we serve.

What excites you most about this booming field?

My entire career has been spent in AI, and the most exciting time is right now. Artificial intelligence has matured to the point where it’s both useful and practical, both clinically and for research. For the first time, we can think seriously about putting AI in the clinic to improve patient care and clinical decision-making.

At Cedars-Sinai specifically, our computational biomedicine capabilities are exceptional, thanks to the infrastructure built by our colleagues in Enterprise Information Services (EIS). This infrastructure—coupled with our collaborative efforts among their employees—has created a culture that is accepting of utilizing AI where it’s most beneficial, safe and effective.   

What sets Cedars-Sinai’s computational biomedicine team apart?

We have recruited some of the best professionals in the country, many of whom were interested in embedding themselves within a hospital with direct opportunities to impact clinical care. We also have a strong academic and training component, and our graduate students serve as the glue to bring collaborators together. 

Cedars-Sinai Health System is especially unique because we work alongside one another to improve patient care. Patient care is our end goal; that’s our bottom line. Silos go out the door, innovation is expedited and building bridges among departments is understood as critical.

The Department of Computational Biomedicine team has many experts and areas of focus. Tell us about some of the people and projects within the department.  

• Jesse Meyer, PhD, an assistant professor in the Department of Computational Biomedicine and a research scientist in the Smidt Heart Institute, is providing computational tools that make data analysis possible. He is working alongside Jennifer Van Eyk, PhD, director of the Advanced Clinical Biosystems Institute in the Smidt Heart Institute and a world leader in clinical proteomics, which is the study of the structure and function of proteins within the body.  
  
  While Van Eyk works to bring mass spectrometry—the way we measure proteins—to every patient at Cedars-Sinai, Meyer is developing the computational methods for processing that data. These tools can infer what proteins are present within a sample, then use AI and machine learning to analyze that data. Meyer is also exploring how—and when—proteomics can integrate with clinical data to better predict clinical outcomes.
   
• Graciela Gonzalez-Hernandez, PhD, vice chair of Research and Education in the Department of Computational Biomedicine, is a pioneer in utilizing artificial intelligence for natural language processing. She is on the front lines of using large language models to mine clinical notes, the published literature, and social media data to address key questions in health research.
  
  For example, if individuals are having an adverse reaction to a commonly prescribed medication, they may share their experience online through social media. Gonzalez-Hernandez would collect that online data, create a repository for it, then mine that information for clinically relevant takeaways.
   
• In medical school, trainees learn the “if, then” rule. If a patient is older than 65 years old, then you should check what medications they take. We take the same “if, then” approach in computational biomedicine, but call it rule-based machine learning. The concept is that we build machine learning models from data, then teach the technology “rules” that are intended for clinicians to interpret.
  
  Ryan Urbanowicz, PhD, a research assistant professor on our team, is one of the world’s experts in rule-based machine learning. He has developed powerful systems that mine for data, then present clinical findings in an immediate, and explainable way, for clinicians. 
   
• Automated machine learning is a keen interest of mine and a specialty where we translate data into usable information. At Cedars-Sinai, we are collecting a tremendous amount of data in proteomics, genomics and beyond.
  
  Our experts, including Kyoung Jae Won, PhD, and Nicholas Tatonetti, PhD, integrate computational methods, databases, machine learning and AI to provide translational, clinical information. The end goal with our work is to do a better job of understanding diseases and helping patients make more informed decisions.  
   
• Tiffani Bright, PhD, a national leader in applied clinical informatics, serves as co-director of the Center for Artificial Intelligence Research and Education within the Department of Computational Biomedicine. She is spearheading the development of new AI algorithms and software, and applying those findings into genomic research, personalized medicine and other healthcare research applications. Bright's work ensures our team is diverse, equitable and inclusive in its research and education programs, making certain that innovative solutions are accessible and relevant to all communities.
   
• The primary roles within our department wouldn’t be possible without our biostatisticians, whose complementary job function is critical to our work. Under the leadership of Mourad Tighiouart, PhD, director of the Biostatistics Core and Biostatistics Research Center, these team members apply math and statistics to answer some of the biggest questions at Cedars-Sinai and in the broader healthcare landscape. As a cancer research scientist, Tighiouart ensures our computational biomedicine team collaborates and communicates with the cancer enterprise at Cedars-Sinai—which expedites novel discoveries in the laboratory.

Read more on the Cedars-Sinai Blog: Dr. Tiffani Bright | AI Evangelist and Skeptic

Another is studying how artificial intelligence (AI) technology can help predict disease outcomes through X-rays, CT scans and MRI images.

A third group hopes to predict COVID-19 outbreaks using genomic data that machine learning algorithms understand.

This is a sampling of the eight projects that more than 170 undergraduate and graduate students, postdoctoral students, scientists, medical residents, faculty members and others working in scientific or medical institutions are participating in this spring through Cedars-Sinai’s National AI Campus. This project-based learning initiative is in its second year at Cedars-Sinai, and it brings together AI experts and people from various educational and professional levels to address challenging problems in science and medicine using AI and machine learning.

National AI Campus is part of the Center for Artificial Intelligence Research and Education, a component of the medical center’s Department of Computational Biomedicine. It was developed in 2018 by Xiuzhen Huang, PhD, research professor in Computational Biomedicine, when she was at Arkansas State University. When Huang joined Cedars-Sinai, she brought National AI Campus along. Today, National AI Campus is connected to a nationwide program with participants from diverse academic programs and institutions and also includes high school students.

“National AI Campus makes artificial intelligence accessible to a broad community by offering a collaborative, highly interactive training program in which everyone can learn from each other,” Huang said. “At Cedars-Sinai, because of our focus as a medical center and academic institution, we tailor our program around biomedically related projects. We offer medical imaging and genomic projects, as well as those focusing on business analytics and social sciences.”

National AI Campus is free of charge and open to anyone working at Cedars-Sinai or at other invited institutions—at any experience level, in any degree or area of specialty, and with or without previous programming, machine learning or high-performance computing experience.

“All you need is a willingness to learn the basics and an interest in exploring the leading edge of AI and machine learning in medicine,” said Professor Jason Moore, PhD, chair of the Department of Computational Biomedicine, director of the Center for Artificial Intelligence Research and Education and chair of the National AI Campus Steering Committee.

“Artificial intelligence has great potential to improve modern life by addressing many of society’s major challenges, particularly those related to human health. At Cedars-Sinai, our ultimate goal is to one day move our findings into the clinic to help patients.”  

AI makes it possible for computers to perform tasks that require human intelligence. Machine learning is a subset of artificial intelligence in the computer science field. It gives computers the ability to “learn” with data, without being explicitly programmed, and it recognizes patterns in large amounts and diverse types of data to generate important insights.

There are two phases of the National Campus AI program, and each phase lasts four months. To ensure each phase is successful, Huang enlisted Ryan Urbanowicz, PhD, a research assistant professor in the Department of Computational Biomedicine, to serve as director of the campus program, and Joshua Levy, PhD, director of Digital Pathology Research, as associate director.

During Phase One, small interdisciplinary teams of up to 20 people work individually and as a group—online and on their own time—on a project. The teams are led and mentored by experts who have experience in each project topic. Phase One culminates with a showcase of team presentations, including to the Cedars-Sinai research community. This year’s showcase is planned for the summer.

During an optional Phase Two of the program, participants are selected to be part of a global AI competition or a novel Cedars-Sinai-based collaborative research project aiming toward publication in a peer-reviewed medical journal. Cedars-Sinai’s National AI Campus teams won Phase Two competitions in 2023 and have been recognized nationally and internationally.

More than 50 universities across the U.S. also participate in National AI Campus, including 21 historically Black colleges and universities. Cedars-Sinai experts are working to expand the program to California universities and are helping California State University at Dominguez Hills start its own National AI Campus program.

One participant in Cedars-Sinai’s spring 2024 National AI Campus program kickoff said she works in research administration. She noted that she was eager to learn ways to use AI to make data analysis more precise and efficient in her job and to learn skills she could use in other ways.

This feedback aligns with another goal of National AI Campus, Huang said: to create a strong educational resource for students and faculty members and enhance workforce development in AI.

“AI technology is impacting our lives in a major way,” Huang said. “I compare it to when the steam engine came along 300 years ago, bringing dramatic change and transforming industry. AI is another powerful invention that is making significant changes, and although there is some societal anxiety around how to safely use it, I always emphasize that AI will change our life, but it will never replace our life.

“It’s important that Cedars-Sinai harnesses the potential that AI and machine learning offer, and one way we are doing that is through National AI Campus.”

Read more on the Cedars-Sinai Blog: Physician Shadowing Program Offers Students Rare Access

The findings, published in the peer-reviewed journal Nature Communications, suggest that more was in play than either factor alone.  

“Our results suggest that initially, differences in shared home environment influenced who was infected with COVID-19 more than genetic differences,” said Katie LaRow Brown, MA, first author of the study and a PhD candidate at Columbia University who collaborated with Cedars-Sinai on this study. “Over time, however, the importance of these differences in shared home environment decreased—and the importance of genetics increased—eventually eclipsing shared home environment.”

COVID-19 has infected more than 340 million people in the U.S., underscoring the urgency in conducting therapeutic research and uncovering potential treatments. However, until this study, little was known about how an individual’s environment and genetic background impacted their experience with the virus.

Using electronic health records from New York-Presbyterian/Columbia University Irving Medical Center, investigators identified 12,764 patients who received conclusive results—either positive or negative—from a PCR test for COVID-19. These patients belonged to 5,676 families with an average of 2.5 family members who had a bout of COVID-19. The time frame studied was Feb. 21, 2020, to Oct. 24, 2021.

The investigators’ analysis found that at the start of the pandemic, genetics accounted for 33% of variation in susceptibility. By the second half of the research study, however, genetics accounted for 70% of variation in susceptibility.

When measuring patients’ severity of COVID-19, investigators also found that a patient’s genetics were more of a factor than their home environment. Disease severity was defined by length of hospital stay. Genetics explained 41% of variation while shared environment explained 33%.  

“We were especially surprised by the percentages of susceptibility,” said Nicholas Tatonetti, PhD, senior and corresponding author of the study, vice chair of Operations in the Department of Computational Biomedicine and an associate director of Computational Oncology at Cedars-Sinai Cancer. “Since this is an infectious disease, we assumed that home environment differences would explain most variation for the entirety of the study.”

While Tatonetti says his team of investigators cannot know for certain, they suspect that over time, discrepancies between people’s home environments changed in important ways.

“This work also suggests that the specific genetic factors influencing susceptibility and severity have not been fully identified,” said Tatonetti. “This is very important in terms of directing resources and defining future research goals.”

Jason Moore, PhD, chair of the Department of Computational Biomedicine and a professor of Medicine, said the study provides critical information and insights for future pandemics.

“The age-old debate of what matters most—genetics or your environment—continues through the work of this important study,” said Moore. 

Funding: This study was supported by the National Institutes of Health National Institute of General Medical Sciences R35GM131905.

Other authors involved in the study include Vijendra Ramlall, Michael Zietz, and Undina Gisladottir.

Read more from the Cedars-Sinai Blog: COVID-19 and Flu Shots Provide a Double Dose of Protection

“These newly instated Research Town Halls are intended to create an ongoing dialogue between the research institute and investigators,” said Jeffrey Golden, MD, vice dean for Research and Graduate Education, director of the Burns and Allen Research Institute, and professor of Pathology and Laboratory Medicine at Cedars-Sinai. “Our hope is to create an open forum for faculty to express ideas and needs while also hearing firsthand about research opportunities, new developments and future planning at Cedars-Sinai.”

Golden shared several research-related updates, including an invitation to participate in Research Day, scheduled for Wednesday, March 27, 2024. The keynote lecture will be presented by David R. Liu, PhD, from the Broad Institute of the Massachusetts Institute of Technology and Harvard University.

Golden also detailed a newly established relationship with Tsinghua University in China. Recently, Cedars-Sinai welcomed five medical students from the university to work for two years—and possibly more—in various Cedars-Sinai laboratories. The next cohort of participants will be interviewed in 2024, and Golden expects to select another five to seven students from China to participate.

“In addition to hosting the inaugural five medical students, we are in the process of exploring opportunities for Cedars-Sinai faculty to spend short periods of time studying and training at Tsinghua University with a goal of establishing research collaborations,” Golden said.

Research Town Hall attendees also heard from Graciela Gonzalez-Hernandez, PhD, vice chair for research and education in the Department of Computational Biomedicine at Cedars-Sinai. Her topic—generative AI in healthcare—challenged listeners to dissect and question ChatGPT and other AI tools.

“ChatGPT is designed to please you by providing as many details as possible, but those details often lack context and sources,” Gonzalez-Hernandez said. “The field of artificial intelligence is now shifting toward a reinforcement learning to provide a measure of checks and balance.”

Reinforcement learning perceives and interprets a particular environment, takes suitable action, then learns through trial and error, Gonzalez-Hernandez said.

“Systems utilizing reinforcement learning can learn, grow and evolve—then be rewarded for making smart, solid decisions,” she said. 

Her presentation ended with a caution and a tip.  

The caution, which Golden underscored, is that the National Institutes of Health issued a statement prohibiting any form of ChatGPT from being used when reviewing grant applications. We all must be aware of the risk of using open-source software that will make confidential or protected information publicly available.

“My tip to you is that, instead of ChatGPT, I encourage the use of generative AI with greater transparency, like Perplexity.ai,” Gonzalez-Hernandez said. “This system acts as a next-generation search engine and displays the source of the information it provides.”

The final presentation by V. Krishnan Ramanujan, PhD, director of the Cedars-Sinai Biobank and a research associate professor of Medicine at the medical center, began with an announcement about some long-awaited news.

“We just learned that the Cedars-Sinai Biobank received accreditation from the College of American Pathologists,” Ramanujan said. “This accreditation recognizes our institutional biobank as a national center of quality that meets the highest standards.”

With the accreditation in hand, Ramanujan spotlighted the biobank’s team and vision.  

The Biobank and Research Pathology Resource Program comprises four service areas: biobanking, histology, research pathology, and microscopy and image analytics. Institutional partners include Enterprise Information Services, Anatomic Pathology, Facilities Operations, the Office of Research Compliance and Quality Improvement, and the Office of Research Administration.  

“Looking ahead, we will turn to the research community to take advantage of this state-of-the-art institutional resource and ask that we work together to develop new strategic collections,” Ramanujan shared. “Our team will continue advancing our tissue microarray resources, digital pathology workflows and biomarker validation platforms.”

Follow Cedars-Sinai Academic Medicine on Twitter for more on the latest basic science and clinical research from Cedars-Sinai.

Hung is highly skilled in robotics and advanced laparoscopic techniques to treat urologic cancers. He also is an internationally respected leader in the development of surgery simulation technology, as well as an innovator in developing automated performance metrics and machine learning algorithms to predict surgical outcomes.

“We extend a warm welcome to Dr. Hung, a leader in our field who has an excellent reputation as a surgeon and as a pioneering researcher in medical innovations,” said Hyung Kim, MD, chair of the Department of Urology at Cedars-Sinai and the Homer and Gloria Harvey Family Chair in Urologic Oncology in honor of Stuart Friedman, MD. “Dr. Hung’s leadership and expertise will continue to advance our next-generation clinical and research efforts to benefit urology patients.”    

Prior to joining Cedars-Sinai, Hung was tenured associate professor of Urology and director of the AI Center in Urology at the University of Southern California (USC), where he was instrumental in incorporating artificial intelligence into robotic surgery. 

He has received multiple grants from the National Institutes of Health, and his work, comprising more than 200 papers, has been published in top peer-reviewed journals such as Nature Medicine, Nature Biomedical Engineering, Nature Communications Medicine, JAMA Surgery and the Journal of Urology.

Hung was the first consulting editor on AI for the British Journal of Urology International. He currently serves on the American Urological Association’s Research Grants and Investigator Support Committee and New Technologies Committee.

Hung earned his bachelor of science degree in molecular, cellular and developmental biology from Yale University and his medical degree from Weill Cornell Medical College at Cornell University. He completed his urology residency and robotics fellowship in the Department of Urology at USC.  

“It is an exciting time in urological technology and treatment innovation,” Hung said. “I’m honored to be among the team of world-class Cedars-Sinai surgeons and researchers who are bringing the latest discoveries to patients.”

Cedars-Sinai once again has been recognized by U.S. News & World Report, this year as the #5 program in the U.S. for Urology. Cedars-Sinai is also #1 for Urology in California and Los Angeles (highest U.S. News ranking in the state and region).

Read more on the Cedars-Sinai Blog: Who Should Get a Prostate Cancer Screening and When?

Levy’s research focuses on implementing AI, digital pathology and statistics in various fields of anatomic pathology. Levy’s work has resulted in the development of several computational methods, including tools that measure spatial molecular variations in tumors. Those variations can help physicians more accurately determine the severity of patients’ disease.

“Dr. Levy brings a wealth of expertise in digital pathology and computational biomedicine, which will significantly enhance our research capabilities in these critical areas,” said David Frishberg, MD, chair of the Department of Pathology and Laboratory Medicine.

Levy said his goal is to help build a nationally recognized digital pathology research program at Cedars-Sinai, and to create pathways for student recruitment to ensure a consistent infusion of new talent into the field.

“Joining the Cedars-Sinai team has been a thrilling opportunity,” Levy said. “Cedars-Sinai fosters a collaborative and innovative environment, blending a culturally rich and vibrant community with a diversity of ideas and perspectives.”

Levy joins Cedars-Sinai from the Geisel School of Medicine at Dartmouth-Hitchcock Medical Center, where he was an assistant professor in the Department of Pathology and Laboratory Medicine, along with joint appointments in the Department of Dermatology and Department of Epidemiology. He also held a faculty position in the Quantitative Biomedical Sciences Graduate program, in addition to contributing to the Biostatistics and Bioinformatics and Trace Element Analysis Shared Resources at the Dartmouth Cancer Center. Levy also holds various positions at Veterans Affairs and the Biomedical National Elemental Imaging Resource. Notably, he co-founded and co-directed the Machine Learning arm of the Emerging Diagnostic and Investigative Technologies (EDIT) program.

Levy completed his undergraduate studies in physics at the University of California, Berkeley, and earned his doctorate in quantitative biomedical sciences (data science) from the Dartmouth College Geisel School of Medicine.

In addition to research, Levy is dedicated to education and mentorship. He has successfully launched a national internship program that has mentored more than 150 high school, college and postgraduate students.

“Fostering the next generation of scientists and clinicians is a responsibility I hold dear, and I am committed to contributing to their growth and development in this new role,” Levy said.

Levy's academic background is accompanied by a substantial publication record, including more than 58 peer-reviewed publications. He has received awards from Berkeley, Dartmouth, Modern Pathology, BIOSTEC and the Guarini School of Graduate and Advanced Studies .

Jason Moore, PhD, director of the Department of Computational Biomedicine, said, “Levy's innovative research and commitment to education make him a valuable addition to the Cedars-Sinai team.”

Read more in Discoveries: Living in AI’s Endless Summer

Building on a legacy of innovation, Cedars-Sinai is harnessing rapidly evolving breakthroughs in AI technology to enhance patient care, improve efficiency, advance scientific discovery, and cultivate greater physician and staff wellbeing.

AI at Cedars-Sinai already is having an early impact on clinical and research initiatives. Investigators, for example, are using the technology to identify the earliest signs of pancreatic cancer, predict sudden cardiac arrest and uncover genetic predictors of Alzheimer’s disease risk.

Cedars-Sinai leaders are excited by the possibilities afforded by AI—including the potential to reduce healthcare disparities and costs—as they guide the organization through this dynamic revolution in healthcare.

“AI extends and augments human capabilities and intelligence,” said Craig Kwiatkowski, PharmD, senior vice president and chief information officer. “It holds the potential to transform the ways we envision, plan and deliver care. Because of the vast opportunities, we are moving deliberately in these early stages of the journey.”

The Three Pillars of Artificial Intelligence at Cedars-Sinai

Cedars-Sinai’s AI strategy is built on three foundational strategic pillars: investing and planning, transitioning innovation into adoption, and supporting sound, ethical use of new technologies.

Through its first pillar, Cedars-Sinai is investing in state-of-the-art technology, infrastructure and services while fostering AI fluency across the workforce. The intent is to lay a foundation to meet the organization’s future healthcare needs.

The second pillar calls for the use of AI research and innovation to solve critical, real-world healthcare challenges—accelerating the integration of AI discoveries into clinical practice and delivering benefits to patients, physicians and the healthcare delivery system.

The third pillar focuses on the ethical and responsible use and governance of AI by adhering to regulatory requirements while ensuring that AI tools are used in fair and unbiased ways that protect patients and their privacy.

“The AI journey doesn’t have a finish line, and we must keep our eyes on the road ahead,” said Mike Thompson, vice president of Enterprise Data Intelligence, who works alongside Kwiatkowski to steer the organization’s AI strategy. “We are building a strong foundation for a future of limitless possibilities.”

Artificial Intelligence Council

An essential component of success involves the creation an Artificial Intelligence Council. It brings together cross-functional leaders—from patient care, research, data and technology teams—to review, guide and coordinate AI strategy. The council provides a forum for open dialogue and an ongoing exchange of ideas while setting priorities, evaluating the use of AI tools and identifying measures of success.

“The AI Council is a core piece of our commitment to using AI responsibly,” said Jason Moore, PhD, chair of the Department of Computational Biomedicine and a founding member of the council. “It is critical that we promote responsible AI principles, helping to ensure that AI is deployed safely, effectively and in an unbiased and transparent manner.”

Artificial Intelligence in Action

AI can accelerate scientific discovery by advancing the understanding of disease states and potential treatment pathways. At the same time, AI can improve the efficiency and accuracy of clinical data, freeing providers to spend greater face-to-face time with patients.

Early adoption of artificial intelligence already is making a difference in research and clinical programs at Cedars-Sinai. Examples include:

• Pancreatic Cancer: Cedars-Sinai investigators have leveraged AI to identify the earliest signs of pancreatic cancer, a disease notoriously difficult to diagnose in its early stages. By using advanced machine learning algorithms to analyze medical imaging scans and patient records, the AI system may help prevent deaths through early detection, leading to timely interventions and improved patient prognoses.
• Heart Health: Research led by investigators in the Smidt Heart Institute and the Division of Artificial Intelligence in Medicine in the Department of Medicine is helping clinicians get closer to predicting two common heart conditions: sudden cardiac arrest, which is often fatal, and increased coronary artery calcium, a marker of coronary artery disease that can lead to a heart attack.
• Brain Cell Modeling: Investigators from the Anastassiou Lab—members of the Departments of Neurology and Neurosurgery, the Board of Governors Regenerative Medicine Institute and the Center for Neural Science and Medicine at Cedars-Sinai—have created complex computer models of individual brain cells, unlocking new avenues for understanding brain function and neurological disorders.
• Alzheimer's Disease Research: An $8 million grant from the National Institutes of Health to study Alzheimer’s disease is enabling investigators to study new, leading-edge artificial intelligence methods and to use these to identify genetic predictors of Alzheimer’s disease risk.
• Liver Disease: Cedars-Sinai experts are investigating how ChatGPT may help improve outcomes for patients with cirrhosis and liver cancer by providing easy-to-understand information about lifestyle changes and treatments.
• Obstetrics and Gynecology: AI is helping physicians make headway in predicting the need for cesarean section delivery. By analyzing electronic health records, the Cedars-Sinai AI model can help physicians assess factors influencing the need for C-sections, potentially leading to better outcomes and informed decision-making for parents.
• Spine Surgery: The Department of Computational Biomedicine, in collaboration with Cedars-Sinai’s AI Council and spine surgeons, is using AI and machine learning to predict which patients are most likely to successfully manage their pain post-surgery and which ones might need additional assistance.
• COVID-19: During the height of the COVID-19 pandemic, Cedars-Sinai developed an AI model that helps physicians diagnose the severity of COVID-19 pneumonia. This technology already is being used in multicenter clinical studies.

Although AI initiatives are already having an impact at Cedars-Sinai, those leading the organization’s strategy are busily planning to expand the scope of applications.

“We are only at the very beginning of understanding what AI can do to improve healthcare and quality of life for our patients, our physicians and our staff,” Kwiatkowski said. “We are committed to building a firm foundation as we head into the future.”  

Read more in Discoveries: The Human Factor of Artificial Intelligence