Mehdi Astaraki
Contact
Name and title: Mehdi Astaraki
ORCID0000-0001-5125-4682 Länk till annan webbplats.
Workplace: Department of Physics Länk till annan webbplats.
Visiting address Room R803 053CCK Medicinsk Strålningsfysik Visionsgatan 56 plan 03
Postal address Medicinsk strålningsfysikBox 260, 171 76 Solna
About me
Mehdi Astaraki is an Associate Professor in the Division of Medical Radiation Physics within the Department of Physics at Stockholm University. His research centers on medical image/data processing and analysis, with a primary focus on oncological applications. Specifically, he specializes in developing deep learning (DL) and machine learning (ML) models for cancer diagnosis, prognosis, and treatment planning.
He holds both Bachelor's and Master's degrees in Biomedical Engineering. In September 2022, he earned a dual Ph.D. through a joint program between the Division of Biomedical Imaging at KTH Royal Institute of Technology and the Department of Oncology-Pathology at Karolinska Institutet. Following his doctoral graduation, he completed a postdoctoral fellowship at Stockholm University's Division of Medical Radiation Physics before being appointed to his current role as Associate Professor in November 2025.
He is the primary teacher for the following courses:
Image and System Analysis
AI and Computational Methods in Medical Radiation Physics.
Beyond his current curriculum, he holds extensive teaching experience across a broad spectrum of technical subjects, including medical image processing and analysis, deep learning for medical image analysis, signal and system analysis, linear control systems, numerical methods, and engineering mathematics.
His current research portfolio encompasses several projects in medical image analysis.
A major focus is the development of robust segmentation models for diagnosis and radiotherapy treatment planning.
These projects aim at tackling challenging segmentation tasks, particularly concerning small pathologies such as brain, liver, and bone metastases, as well as lymphomas.
Additionally, his work involves segmenting Clinical Target Volumes (CTV) for Gliomas and Head and Neck cancers.
Beyond segmentation, his ongoing research includes developing multimodal deep learning models for diagnostic and prognostic applications, analyzing longitudinal data for disease monitoring, and imaging biomarkers for non-invasive measurements.
Master's Thesis Opportunities: Prospective Master's students interested in pursuing a thesis project in any of the aforementioned areas are welcome to reach out. Please contact via email with your CV attached.