“If we had vaccines offering protection not just against a single virus, like SARS-CoV-2 or Covid-19, but against a range of viruses all belonging to the same family, what could that do during a pandemic?” Dr. Charlie Whittaker examines how to model and allocate vaccines during a future pandemic, particularly in scenarios where vaccines are in short supply. His research focuses on helping policymakers decide who should be vaccinated and where vaccination efforts should be concentrated to best contain a pandemic.

Dr. Serina Chang and colleagues describe how AI can accelerate breakthroughs in infectious disease epidemiology, integrate diverse data sources to improve infection control, and enable robust and equitable public health action.

Understanding of how social or structural determinants of liver transplant success has been limited, with information documented only in unstructured clinical notes Drs. Jin Ge, Irene Chen and colleagues use an LLM to analyze unstructured notes in the medical record, combine those findings with clinical and demographic data, to identify factors leading to or impeding successful transplant.

Dr. Sandya Subramanian and colleagues tracked unconscious pain during surgery using non-invasive wearable sensors and new computational models. The study provided an objective measure of “nociception,” (subconscious perception of pain), and opens the way for personalized anaesthesia management.

When does adding more data help, and when does it hinder, desired model performance? Dr. Irene Chen and colleagues show that adding training data can result in reduced accuracy, fairness, or subgroup performance–and introduce heuristics to guide decision-making on which data additions will yield improvements in model performance.

The relational nature of large language models and other forms of generative AI raises additional questions for medicine, posing new ethical challenges beyond existing frameworks like the Belmont principles of beneficence, respect for persons, and justice. Drs. Ida Sim and Christine Cassel call for the establishment of proactive health AI ethics centers and regulatory bodies to ensure ethical standards are met as AI technologies rapidly evolve.