cv

Curriculum vitae of Saikat Roy — education, experience, publications, and awards.

General Information

Full Name Saikat Roy
Location Heidelberg, Germany
Email contact@saikatroy.me

Education

  • 2023 - 2026
    Ph.D., Computer Science
    Heidelberg University, Heidelberg, Germany
    • Thesis: Transformers and Self-Attention for Medical Image Segmentation
  • 2017 - 2021
    M.Sc., Computer Science
    University of Bonn, Bonn, Germany
    • Thesis: Fully-3D Deep CNNs for Segmentation of Neuroanatomy
  • 2013 - 2015
    M.E., Software Engineering
    Jadavpur University, Kolkata, India
    • Thesis: Supervised-Layerwise Training of Deep CNNs for Classification

Experience

  • May 2026 - Present
    Senior Machine Learning Engineer
    mediaire GmbH, Berlin, Germany
    • Leading AI development for MRI-based medical image analysis products.
  • Oct. 2025 - Present
    Visiting Researcher
    Diagnostic Image Analysis Group (DIAG), Radboud University Medical Center, Nijmegen, Netherlands
    • AI for prostate cancer diagnosis.
  • Mar. 2021 - Feb. 2026
    Doctoral Researcher (Full-Time)
    German Cancer Research Center (DKFZ), Heidelberg, Germany
    • Developed one of the highest-cited CNN architectures of the last 2 years, using Transformer-based insights, for 3D medical image segmentation.
    • Research topics: semantic segmentation, Transformers, representation learning, transfer learning, self-supervised training, foundation models.
  • Jul. 2024 - Nov. 2024
    Research Intern (Full-Time)
    Siemens Healthineers, Princeton, USA
    • Developed patch-based diffusion models for medical image denoising and super-resolution at 25% less memory with no performance loss.
    • Research topics: diffusion models, denoising, super-resolution.
  • Nov. 2018 - Feb. 2021
    Research Assistant (Half-Time)
    German Center for Neurodegenerative Diseases (DZNE), Bonn, Germany
    • Developed spatially ensembled 3D deep convolutional neural networks for state-of-the-art segmentation of 95 structures in brain MRIs in 30 seconds.
    • Research topics: semantic segmentation, deep CNNs, neuroimaging.

Publications (Selected)

  • Primus: Enforcing Attention Usage for 3D Medical Image Segmentation. TMLR, 2026. Details
  • Investigating the Feasibility of Patch-Based Inference for Generalized Diffusion Priors in Inverse Problems for Medical Images. IEEE ISBI, 2025. Oral Presentation. Details
  • Skeleton Recall Loss for Connectivity Conserving and Resource Efficient Segmentation of Thin Tubular Structures. ECCV, 2024. 100+ citations. Details
  • Touchstone Benchmark: Are We on the Right Way for Evaluating AI Algorithms for Medical Segmentation? NeurIPS, 2024. Benchmark Win: MedNeXt. Details
  • nnU-Net Revisited: A Call for Rigorous Validation in 3D Medical Image Segmentation. MICCAI, 2024. Benchmark Win: MedNeXt, 700+ citations. Details
  • MedNeXt: Transformer-driven Scaling of ConvNets for Medical Image Segmentation. MICCAI, 2023. Early Accept: Top 14%, 700+ citations. Details
  • See the full publications page for the complete list.

Open Source Projects

Talks and Research Visits

  • 2026
    Speaker on MedNeXt-v2 at Raysearch Laboratories, Sweden
  • 2026
    Speaker at Sprint AI Training for African Medical Imaging Knowledge Translation (SPARK) Academy
  • 2025
    Research visit to Diagnostic Image Analysis Group (DIAG), Radboud UMC, Netherlands
  • 2025
    Speaker at Machine Learning Galore Talks, Heidelberg University, Germany
  • 2024
    Keynote speaker, MICCAI 2024 Challenge: "Body Maps: Towards a 3D Atlas of the Human Body"
  • 2023
    Research talk and visit to Luxembourg Centre for Systems Biomedicine, University of Luxembourg
  • 2023
    Invited talk at Medical and Environmental Computing Lab, TU Darmstadt, Germany
  • 2022 - 2023
    Tutorial: "Transformers in Medical Image Analysis", BVM, Germany

Honors and Awards

  • 2021
    • Oxford Machine Learning (OxML) Summer School
  • 2015
    • Erasmus Mundus FUSION Scholarship for PhD mobility to University of Evora, Portugal (offer not accepted)
  • 2013 - 2015
    • GATE Scholarship for Master's Degree, Ministry of Human Resources & Development, Government of India

Academic Interests

  • Research
    • Machine Learning
    • Semantic Segmentation
    • Transformers
    • Diffusion Models
    • Representation Learning
    • Transfer Learning
    • Predictive and Generative Problems in Medical Image Analysis

Skills

  • Proficient
    • Python
    • NumPy
    • PyTorch
    • Scikit-Learn
    • Git
    • LaTeX
    • Matplotlib
  • Familiar
    • Shell Scripting
    • Pandas
    • SciPy
    • Docker
    • PyTest
    • C
    • SQL
    • Matlab
    • Keras
    • HPC (LSF, SLURM)

Reviewer Assignments

  • Conference on Neural Information Processing Systems (NeurIPS), 2026
  • International Conference on Learning Representations (ICLR), 2025 - Present
  • IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2025 - Present
  • International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), 2024 - Present
  • IEEE Transactions on Image Processing, 2024 - Present
  • IEEE Transactions on Neural Networks and Learning Systems, 2023 - Present
  • IEEE Transactions on Medical Imaging, 2023 - Present