Fabian Isensee

Fabian Isensee is a medical AI researcher who develops deep-learning methods for biomedical image analysis. His work includes nnU-Net, a self-adapting segmentation framework, and HD-BET, a tool for brain extraction from MRI. At DKFZ, he works with Klaus Maier-Hein's team on nnU-Net and extends it for CT and MRI segmentation challenges; in 2024, he coauthors a benchmark of 3D methods. He receives the Leopoldina Prize for Young Scientists in 2025 and leads Helmholtz Imaging's Applied Computer Vision Lab at DKFZ. As of October 2026, he is a board member of DKFZ's Medical Image Computing division and continues his work there after stepping down from Helmholtz Imaging's Support Unit leadership in March 2026.

Now at: Helmholtz Association of German Research Centres and Heidelberg University
Born: 1990
Works: 224
Citations: 19,783
h-index: 43
Badges: Bridge

Years link to their sources. Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

2018

Introducing nnU-Net

In 2018, he and his coauthors introduce nnU-Net, a self-adapting framework built on 2D and 3D U-Nets, and evaluate it on the Medical Segmentation Decathlon.
2019

The HD-BET tool

In 2019, he coauthors HD-BET, a neural-network tool for extracting brains from MRI. It handles varied MRI sequences and pathological or treatment-altered tissue.
2021

Training nnU-Net at DKFZ

In 2021, he works with Klaus Maier-Hein's team at DKFZ on nnU-Net, which identifies organs, tumors and other structures by labeling image pixels.
2022

nnU-Net at AMOS2022

In 2022, he and his coauthors adapt nnU-Net for the AMOS challenge's CT and combined CT/MRI tasks. Their ensemble achieves Dice scores of 90.13 and 89.06, respectively.
2024

Revisiting 3D segmentation

In 2024, he coauthors nnU-Net Revisited, a benchmark of CNN, Transformer and Mamba approaches to 3D medical image segmentation. The study calls for rigorous validation when comparing methods.
2025

A prize and lab leadership

In 2025, he receives the Leopoldina Prize for Young Scientists for work in AI-supported medical image analysis. At DKFZ, he leads Helmholtz Imaging's Applied Computer Vision Lab, which advances AI-based imaging and develops segmentation methods.
As of October 2026

A DKFZ board role

In March 2026, he steps down from leadership of Helmholtz Imaging's Support Unit at DKFZ. As of October 2026, he is a board member of DKFZ's Medical Image Computing division and continues his work there.