Scholar
Magdalini Paschali
Google Scholar ID: KsGo-_QAAAAJ
Postdoctoral Scholar, Stanford University
Citations & Impact
Citations
1,406
H-index
18
i10-index
23
Publications
20
Co-authors
12
list available
Publications
Resume (English only)
Academic Achievements
- Published a comprehensive review on foundation models in radiology; developed methods to evaluate and enhance the robustness of neural networks in medical imaging; contributed to the development and evaluation of foundation models for radiology, particularly those that can interpret both textual and imaging data; developed an instruction-tuned foundation model for chest X-ray interpretation and report generation (CheXagent project); evaluated a comprehensive model for CT scan analysis with EHR codes and reports (Merlin: 3D Vision Language Model); developed approaches bridging data-driven deep learning with hypothesis-driven testing, focusing on biomarker identification using combined imaging and non-imaging data.
Research Experience
- Until August 2023, she was a Postdoctoral Scholar in the Computational Neuroimage Science Laboratory (CNS Lab) with Prof. Kilian M. Pohl, working on multi-modal machine learning models that can improve the understanding, diagnosis, and treatment of neuropsychiatric disorders.
Education
- Earned her Ph.D. Summa Cum Laude in Computer Science from TUM in 2021, advised by Prof. Nassir Navab. Her dissertation was about 'Learning Robust Representations for Medical Diagnosis'.
Background
- Postdoctoral Scholar at the Department of Radiology, Stanford University, focusing on evaluating the performance and robustness of large-scale AI models for medicine and identifying early disease biomarkers. She is part of the Machine Intelligence in Medical Imaging Research Group (MIMI) and AI Development and Evaluation Lab (AIDE).
Miscellany
- Her husband, Walter Simson, is a Senior AI Engineer at NVIDIA.
