Anushka Ashish Kore
PhD Candidate
Anushka Kore is a PhD candidate in the Diagnostic Image Analysis Group (DIAG) at Radboudumc, supervised by Merel Huisman and Henkjan Huisman. She works on TWIN-X (Multimodal Digital Twins with Generative AI for Explainable Precision Medicine), a European consortium of 19 partners from 13 countries that aims to build trustworthy, explainable AI-driven digital twins of patients to support clinicians in treating cardiovascular and oncological diseases. Within TWIN-X, Radboudumc is a clinical validation partner with a focus on cardiovascular imaging.
Before starting her PhD, Anushka was a research scientist in the RTC AI team (2025–2026). There she worked with the Prevention Hub at the Department of Primary Care on extracting and classifying lifestyle information from unstructured clinical notes. She validated open-source large language models (LLMs) for lifestyle factors such as smoking, alcohol, and drug use, enabling large-scale, privacy-preserving use of lifestyle data from Radboudumc and the RTC Health Data GP database for public health and prevention research.
Anushka obtained her B.E. in Biomedical Engineering from the University of Mumbai in 2022 and then spent a year in the biomedical industry as a Product Specialist. She went on to complete her M.Sc. in Biomedical Engineering at Technische Universiteit Delft, specializing in Medical Physics.
During her Master's, she interned at the Biomedical Imaging Group Rotterdam (BIGR), embedded within IGIT (Image Guidance in Interventions and Therapy) and the ICAI Stroke Lab at Erasmus MC, where she assessed the feasibility of deep learning for automated occlusion detection in stroke DSA images. Her master's thesis, "Spatio-Temporal Deep Learning for Vascular Occlusion Detection on DSA", defended in August 2025, built on this work by developing a deep learning framework that combines spatio-temporal modeling with single-frame spatial networks to automatically detect arterial occlusions in stroke DSA sequences. This work was published as "OccluNet: Spatio-Temporal Deep Learning for Occlusion Detection on DSA" in the Proceedings of the SWITCH Workshop at MICCAI 2025.
- Current research projects:
- Member of: