How AI Is Accelerating Biomedical Discovery: A Q&A with Dr. Monica Bertagnolli
Dr. Monica Bertagnolli, a surgical oncologist who has served as Director of the National Cancer Institute and the National Institutes of Health, has joined Renaissance Philanthropy as a Senior Advisor. Recently, Renaissance Philanthropy CEO Tom Kalil caught up with Dr. Bertagnolli to discuss her work on harnessing AI to improve health outcomes for patients and accelerate the pace of biomedical research.
Tom Kalil: As the former Director of NCI and NIH, you have seen many promising areas of biomedical research. What made you decide to focus on AI?
Dr. Monica Bertagnolli: AI has matured to the point where we can now integrate multiple types of health data to train multimodal models. For example, I am working with hospitals and academic medical centers to develop a “copilot” for breast cancer by integrating data from electronic health records, clinical notes, radiology images, and histopathology slides. With patient consent, doctors will be able to use these AI copilots to ensure that the treatment options they recommend are consistent with expert guidelines, which are designed to guide care that maximizes benefit for patients.
AI-enabled tools will also be able to monitor the medical records of individual patients on an ongoing basis and provide updated recommendations as their needs change.
TK: How could this reduce the gaps in the quality of care between urban and rural communities?
MB: Many rural communities have fewer specialists in fields such as oncology. These copilots could help healthcare professionals practicing in all communities make better recommendations for their patients.
TK: What is enabling these types of advances?
MB: Better technologies are helping us make progress on the privacy, security and interoperability challenges of sharing personal health information. This allows hospitals with different electronic health record (EHR) systems to share data for research and health care improvement, while protecting patient privacy. These methods can also allow patients to decide how they want their health information to be used and to receive information on what results were achieved when their information was included in research. This is very important, as it can go a long way toward earning a patient’s trust in health care and in the research process.
TK: In addition to helping doctors make decisions that are consistent with expert guidelines, what are other benefits of sharing health data?
MB: Medicine is constantly changing, and guidelines must be continually updated. Working with health systems, we can use technology to comprehensively track results for many patients over time and study different aspects of clinical care, allowing us to better understand what works best for patients. The results obtained in this manner can be considered by the expert panels who develop guidelines, creating a loop where a stream of current information from clinical care continually informs better future care. This means that we will finally be able to implement a learning healthcare system that improves results for people over time.
It’s also going to enable breakthroughs in areas of biomedical research that have been very challenging, such as early detection, identifying subtypes of different diseases, and understanding the many environmental and social influences on disease. There are so many important questions. For example, how does treating obesity with GLP-1 drugs affect the results of cancer treatment? What is the best way to monitor a person to diagnose an early, treatable cancer? How do we personalize care for individual patients? Our ability to diagnose, treat, cure or prevent a given disease will increase considerably if we can access and analyze a much broader range of information from individual people and their communities.
TK: Is this approach just relevant to cancer?
MB: Not at all. I am in active conversations with healthcare providers and researchers interested in applying this same approach to maternal health, autism, neurodegenerative diseases, and multiple sclerosis. Working with Renaissance Philanthropy, I’d be delighted to partner with philanthropists and foundations that are interested in applying these approaches to the diseases that they care about.