Doctoral Thesis · Maastricht University, GROW

Large Language Models in
Healthcare and Medicine

Shan Chen · Harvard-Mass General Brigham AI in Medicine Program

AcknowledgmentsPersonal acknowledgments and thanks Chapter 1General Introduction, Outline and Glossary Chapter 2Use of AI chatbots for cancer treatment information Chapter 3LLM-assisted responses to patient messages Chapter 4NLP extraction of esophagitis severity★ First paper Chapter 5LLMs for social determinants of health extraction Chapter 6Agentic AI for immunotherapy toxicity detection Chapter 7ChatGPT-family evaluation for biomedical reasoning Chapter 8Cross-care: pre-training data and language model bias Chapter 9Fragility to drug names in biomedical benchmarks Chapter 10Helpfulness backfires: sycophancy risks Chapter 11WorldMedQA-V multilingual multimodal evaluation Chapter 12MedBrowseComp deep research benchmark Chapter 13General Discussion and Future Perspectives AppendicesSupplementary publications and contributions