Geisinger Health
1 week ago
CDTnet PhD Fellowship F2: Generating Synthetic Data for Valve Condition Analysis GE HealthCare in United States
Degree Level
PhD
Field of study
Computer Science
Funding
Available
Deadline
Sep 30, 2026
Country
United States
University
Geisinger Health
How do I apply for this?
Sign in for free to reveal details, requirements, and source links.
More information can be found here
Keywords
Suggested positions
About this position
CDTnet PhD Fellowship F2: Generating Synthetic Data for Valve Condition Analysis
This PhD project is hosted by GE HealthCare in Norway and is part of the CDTnet / Horizon Europe MSCA training network. The research sits at the intersection of artificial intelligence, medical imaging, and cardiovascular medicine, with a specific focus on mitral valve disease and mitral regurgitation.
The doctoral candidate will investigate how generative AI can be used to create synthetic echocardiographic data that improves automated image analysis. The work will span multiple echocardiography modalities, including b-mode and color flow, and will explore applications such as valve segmentation and automated assessment of mitral regurgitation. A key theme is understanding how patient-specific information can be incorporated into generative models to improve accuracy, consistency, and clinical relevance.
The project will use both proprietary clinical datasets and public resources such as the CAMUS echocardiography dataset. The candidate will evaluate whether synthetic data can improve the robustness and performance of machine-learning models compared with training on real-world data alone. The developed methods are intended to be integrated into clinical software and assessed in a prospective study of automated mitral valve evaluation.
The fellowship offers a strong industrial research environment with close collaboration across Europe. Planned secondments include Oslo University Hospital in Norway, IDIBAPS in Spain, and King’s College London in the United Kingdom, providing exposure to retrospective analysis, prospective clinical study design, and mechanistic cardiovascular modeling.
Eligibility emphasizes a Master’s degree or equivalent, preferably in machine learning, computer science, statistics, applied mathematics, or electrical engineering, together with strong Python and deep learning skills. Medical image analysis and echocardiography experience are advantageous. Applicants must also satisfy MSCA mobility and eligibility rules, including not having lived or worked in Norway for more than 12 months in the last 3 years and not already holding a doctoral degree.
The application deadline is 30 September 2026 at 23:59 (Europe/Brussels). Applications must be submitted through the CDTnet application portal.
Funding details
Available
What's required
Applicants should have a Master’s degree or equivalent (minimum 120 ECTS credits), preferably in machine learning, computer science, statistics, applied mathematics, or electrical engineering; other degrees may be considered if the candidate has formal competence in machine learning and/or image analysis/computer vision. A strong background in modern deep learning, machine learning, mathematics, linear algebra, and/or statistics is required, along with solid documented experience in Python and relevant ML frameworks such as PyTorch or TensorFlow. Fluent oral and written English is required. Experience with medical image analysis and echocardiography is a plus. MSCA Mobility Rule applies: the applicant must not have lived or worked in Norway for more than 12 months in the 3 years before recruitment. MSCA Eligibility Rule applies: the applicant must not already hold a doctoral degree and must be eligible to enrol in the PhD programme at the University of Oslo.
How to apply
Apply online via the CDTnet application portal. Use the provided application website and submit the required materials before the deadline. Review the CDTnet site for full instructions and project details.
Ask ApplyKite AI
How do I apply for this?
Sign in for free to reveal details, requirements, and source links.