Postdoctoral Research Position in Multimodal Foundation Models for Pediatric Imaging (PI Sergios Gatidis)
Title: Postdoctoral Research Position in Multimodal Foundation Models for Pediatric Imaging
Location: Stanford Radiology
PI: Sergios Gatidis, Associate Professor of Radiology
Duration: 2 years
Application Deadline: Oct 31st, 2025
Position Description:
We are seeking a highly motivated Postdoctoral Research Fellow to join our research group at Stanford Radiology, focusing on the development of large language models (LLMs) and vision-language models (VLMs) for the analysis of pediatric imaging data and clinical information. The successful candidate will contribute to innovative research aimed at enhancing diagnostic accuracy and workflow efficiency in pediatric radiology through advanced AI methodologies. This position offers a unique opportunity to work at the intersection of machine learning, pediatric imaging, and clinical application.
The fellow will lead the design, implementation, and evaluation of LLMs and VLMs that integrate imaging data with clinical narratives to improve diagnostic processes and patient outcomes. The role includes both retrospective model development and prospective clinical validation.
Key Responsibilities:
- Develop and validate LLM and VLM frameworks that leverage pediatric imaging data alongside clinical information for enhanced analysis and interpretation.
- Design workflows for the integration of AI-driven insights into pediatric imaging practices, focusing on automating the extraction of relevant clinical information.
- Build and test predictive models that utilize multimodal data to assist in disease diagnosis and treatment planning.
- Collaborate with pediatric radiologists and clinicians to ensure the clinical relevance and usability of developed models.
- Lead the preparation of publications and present findings at major medical imaging and AI conferences.
Qualifications:
- Ph.D. in Computer Science, Biomedical Informatics, Electrical Engineering, or a related field.
- Demonstrated experience with machine learning, particularly in LLMs, VLMs, and multimodal data integration.
- Strong programming skills in Python; experience with ML libraries such as PyTorch, TensorFlow, or HuggingFace Transformers.
- Strong publication record and ability to work independently in a clinical research environment.
What We Offer:
- A collaborative environment at the forefront of AI and pediatric radiology innovation.
- Direct mentorship from faculty and access to Stanford's extensive computational and clinical infrastructure.
- Opportunity to influence real-world clinical workflows and gain experience working with industry partners.
- Professional growth through publications, conference presentations, and interdisciplinary collaboration.
How to Apply: