Home Thesis

Joint Ph.D. · Maastricht University + Harvard-MGB AIM · Cum Laude

Evaluation and Implementation of Large Language Model-based Applications for Electronic Health Records

AcknowledgmentsPersonal acknowledgments and thanks Cover art for Chapter 1Chapter 1General Introduction, Outline and Glossary Cover art for Chapter 2Chapter 2Use of AI chatbots for cancer treatment information Cover art for Chapter 3Chapter 3LLM-assisted responses to patient messages Cover art for Chapter 4Chapter 4NLP extraction of esophagitis severity★ First paper Cover art for Chapter 5Chapter 5LLMs for social determinants of health extraction Cover art for Chapter 6Chapter 6Agentic AI for immunotherapy toxicity detection Cover art for Chapter 7Chapter 7ChatGPT-family evaluation for biomedical reasoning Cover art for Chapter 8Chapter 8Cross-care: pre-training data and language model bias Cover art for Chapter 9Chapter 9Fragility to drug names in biomedical benchmarks Cover art for Chapter 10Chapter 10Helpfulness backfires: sycophancy risks Cover art for Chapter 11Chapter 11WorldMedQA-V multilingual multimodal evaluation Cover art for Chapter 12Chapter 12MedBrowseComp deep research benchmark Cover art for Chapter 13Chapter 13General Discussion and Future Perspectives Cover art for Chapter 14AppendicesSupplementary publications and contributions