Lennart Bastian
I am a Research Fellow at Imperial College London, supported through the UK Royal Society Newton International Fellowship. I work with the Circle Group of Tolga Birdal, where we pursue topics at the intersection of geometry, topology, and machine learning.
I completed my PhD with highest distinction (summa cum laude) at the Chair for Computer Aided Medical Procedures (CAMP) in Munich, supervised by Prof. Nassir Navab, where I also coordinated research of the Surgical Data Science team. Prior to this, I conducted my M.Sc. in Applied Mathematics at the Technical University of Munich with an emphasis on optimization and statistics, and Bachelor's in Mathematics and Computer Science from New York University's Courant Institute.
I also had previous stints in software and data science at H2oMetrics, a cloud water management startup, and as a research assistant at the MSKCC Levine Lab, developing computational tools for the genetic analysis of myeloid leukemias.
Recent Updates
- Recognized as an Outstanding Reviewer for CVPR 2026 (Top 5%) and a Gold Reviewer for ICML 2026 (Top 25%).
- Two papers — DisPOSE and Collapsed Effective Operators — accepted to ICML 2026.
- TopoOR accepted to MICCAI 2026 (Early Accept, Top 9%).
- Co-leading TopoFound, a £1.3m AIRR-funded grant.
- Started as a research fellow to Imperial College London!
- Successfully defended my PhD thesis (summa cum laude)!
- Two papers recognized at the MICCAI COLAS Workshop in Daejeon — TrackOR received the Best Paper Award and Mitigating Biases in Surgical Operating Rooms was selected as an Oral.
- Copresheaf Topological Neural Networks accepted to NeurIPS 2025.
- Awarded a UK Royal Society Newton International Fellowship (3% Funded).
- Forecasting Rotational Dynamics in SO(3) accepted to ICCV 2025 as an Oral (Top 2.6%).
Selected Work
My research applies ideas from geometry and topology in machine learning and computer vision, with application areas spanning medicine, biology, and shape analysis.
DisPOSE: Projected Polystochastic Diffusion for Self-Supervised 3D Multi-View Human Pose Estimation
International Conference on Machine Learning (ICML), 2026
Collapsed Effective Operators for Higher-order Structures
International Conference on Machine Learning (ICML), 2026
TopoOR: A Unified Topological Scene Representation for the Operating Room
Medical Image Computing and Computer Assisted Intervention (MICCAI) - Early Accept (Top 9%), 2026
Non-Rigid 3D Shape Correspondences: From Foundations to Open Challenges and Opportunities
Computer Graphics Forum (Eurographics STAR) 2026, 2026
Copresheaf Topological Neural Networks: A Generalized Deep Learning Framework
Advances in Neural Information Processing Systems (NeurIPS), 2025
Forecasting Continuous Non-Conservative Dynamical Systems in SO(3)
IEEE International Conference on Computer Vision (ICCV) - Oral (Top 2.6%) , 2025
Beyond Role-Based Surgical Domain Modeling
Medical Image Analysis (MedIA) (Journal IF 10.7, rank among CV venues), 2025