-
the United States as well as clinical imaging and testing data from Stanford. Project themes will include developing models using EHR data to predict outcomes in ophthalmology and glaucoma, as well as investigating
-
NIH- and foundation-funded projects focused on treatment pathways, chronic pain phenotyping, pharmacoepidemiology, and biomarker-based prediction across autoimmune rheumatic diseases. This position
-
labs at Stanford to tackle emerging clinical questions in oncology, utilizing various AI methods, predictive modeling approaches, and large language models. Specific areas of interest include but are not
-
systems. Includes establishing medical reasoning benchmarks and automated / scalable evaluation methods. Developing recommender algorithms to predict specialty care with large-language model based user
-
transmission modeling, statistical modeling, spatial data analysis, and cost-effectiveness analysis. In parallel, we conduct research on vaccine-preventable infections, developing and evaluating predictive
-
predict disease progression? How can we better measure and mitigate the impact of biased training data for downstream clinical uses? Can we improve the factuality and reasoning of foundation models by
-
or crowding in predicting reading skills. In regards to EF, the field is progressing towards a consensus that EFs are important for reading and that EF deficits are common in dyslexia. But due to the limited
-
measures to identify biotypes of disorder, understand how these biotypes relate to symptoms and behaviors, and predict personalized treatment outcomes. This work is supported by the lab’s shared framework
-
) cancer screening modeling, (2) microsimulation, (3) decision analysis/health policy modeling, (4) survival data under competing risks, (5) dynamic risk prediction modeling (e.g., landmark model), and (6