Kerem Delikoyun

https://cdn.keremdelikoyun.com/profile.jpg

I'm recently a machine learning researcher at TUMCREATE in Singapore and a Ph.D. candidate in Electrical and Computer Engineering at the Technical University of Munich (TUM). I'm passionate about building intelligent systems that uncover the mysteries of life, bridging scientific discoveries with real-world impact. My work lies at the intersection of healthcare, computer vision and Gen AI, developing advanced computer vision pipelines for real-time, high-throughput medical imaging, and combining analytical AI with agentic AI to reveal hidden data patterns through multi-agent reasoning and automated grounding. I've published in high-impact journals and conferences and hold several issued and pending patents in computer vision, agentic AI, and biomedical imaging. Beyond research, I'm deeply motivated by the entrepreneurial and economic potential of AI and fascinated by how it will transform our world and society in the near future. Always open to new perspectives. Feel free to reach out!

News

Selected Publications

  1. TriAgent: Automated Biomarker Discovery with Deep Research Grounding for Triage in Acute Care by LLM-Based Multi-Agent Collaboration
    Kerem Delikoyun, Qianyu Chen, Win Sen Kuan, and 3 more authors
    arXiv, 2025
  2. Real-Time Deep Learning Phase Imaging Flow Cytometer Reveals Blood Cell Aggregate Biomarkers for Haematology Diagnostics
    Kerem Delikoyun, Qianyu Chen, Liu Wei, and 13 more authors
    arXiv, 2025
  3. RT-HAM: Real-Time Holographic Agent Microscopy – An End-to-End Foundation Model for Quantitative Phase Imaging in Haematology Diagnostics Towards Automated Discovery of Predictive Biomarkers in Acute Care
    Kerem Delikoyun, Qianyu Chen, Liu Wei, and 9 more authors
    Computer Vision for Science Workshop, CVPR 2025, 2025
  4. OAH-Net: A Deep Neural Network for Efficient and Robust Hologram Reconstruction for Off-Axis Digital Holographic Microscopy
    Wei Liu*, Kerem Delikoyun*, Qianyu Chen, and 7 more authors
    Biomedical Optics Express, 2025