I'm Max, a PhD researcher at the German Cancer Research Center (DKFZ), working at the intersection of deep learning and medical imaging. My research centers on segmentation, temporality and text—developing promptable, interactive tools for longitudinal analysis, tracking, and multimodal understanding. I aim to make these technologies intuitive, reliable, and accessible for both research and clinical use.
-
https://github.com/MIC-DKFZ/nnInteractive: A comprehensive 3D interactive segmentation method supporting diverse prompts like points, scribbles, boxes, and a novel lasso prompt integrated into widely used image viewers like Napari, MITK and 3D Slicer.
-
https://github.com/MIC-DKFZ/LesionLocator (CVPR25): A framework for zero-shot universal tumor segmentation and tracking in 3D whole-body imaging combining a large-scale lesion dataset, promptable segmentation, and deep-learning-based image registration.
-
https://github.com/MIC-DKFZ/LongiSeg (MICCAI24 LDTM Oral): An extension of the nnU-Net framework tailored for longitudinal medical image segmentation incorporating temporal information across multiple timepoints.
-
https://github.com/MIC-DKFZ/Skeleton-Recall (ECCV24): A loss function designed to enhance connectivity conservation in thin tubular structure segmentation without incurring massive computational overheads.
-
https://github.com/MIC-DKFZ/BreastDivider (MICCAI25 WOMEN): The first large-scale dataset and model for left/right breast segmentation in MRI, featuring over 17,000 3D scans with partial lesion annotations.
I'm also a maintainer and contributor to the well-known nnU-Net framework - one of the cornerstones of modern medical image segmentation.
Aside from my main projects I regularly participate in a variety of challenges in the CVPR, MICCAI and other communities:
- 🥇 Foundation Models for Interactive 3D Biomedical Image Segmentation @ CVPR 2025 – First place using nnInteractive
- 🥇 1st place PANTHER @ MICCAI 2025 - Pancreatic Tumor Segmentation in Diagnostic & Linac MRIs
- 🥇 1st place ODELIA @ MICCAI 2025 - Breast Cancer Classification in MRI
- 🥇 1st place autoPET @ MICCAI 2024 – PET/CT lesion segmentation across multiple tracers and centers
- 🥇 1st place TopCoW @ MICCAI 2023 – Multiclass brain vasculature segmentation
- 🥇 1st place ToothFairy @ MICCAI 2024 – Part of the winning team in CBCT segmentation for surgical planning
- 🥉 3rd place AortaSeg @ MICCAI 2024 – High-resolution aortic multiclass segmentation
- 💰 Kaggle Vesuvius Challenge 2023 – 7th out of 1514, $6,500 prize for ancient ink detection
- 🎤 Invited Speaker at Voxel51 Redux@ECCV and Voxel51 Visual AI in Healthcare Events
- 🏅 AI Hero Award – Helmholtz Hackathon on Sustainable AI
- 🧠 Co-lead of The Human Radiome Project (Helmholtz Foundation Model Initiative)
Feel free to reach out if you're interested in collaboration or just want to chat!