nnLandmark: A Self-Configuring Method for 3D Medical Landmark Detection

Alexandra Ertl, Stefan Denner, Robin Peretzke, Shuhan Xiao, David Zimmerer, Maximilian Fischer, Markus Bujotzek, Xin Yang, Peter Neher, Fabian Isensee, Klaus H. Maier-Hein
Proceedings of The 9th International Conference on Medical Imaging with Deep Learning, PMLR 315:894-927, 2026.

Abstract

Landmark detection is central to many medical applications, such as identifying critical structures for treatment planning or defining control points for biometric measurements. However, manual annotation is labor-intensive and requires expert anatomical knowledge. While deep learning shows promise in automating this task, fair evaluation and interpretation of methods in a broader context, are hindered by limited public benchmarking, inconsistent baseline implementations, and non-standardized experimentation. To overcome these pitfalls, we present nnLandmark, a self-configuring framework for 3D landmark detection that combines tailored heatmap generation, loss design, inference logic, and a robust set of hyperparameters for heatmap regression, while reusing components from nnU-Net’s underlying self-configuration and training engine. nnLandmark achieves state-of-the-art performance across five public and one private dataset, benchmarked against three recently published methods. Its out-of-the-box usability enables training strong landmark detection models on new datasets without expert knowledge or dataset-specific hyperparameter tuning. Beyond accuracy, nnLandmark provides both a strong, common baseline and a flexible, standardized environment for developing and evaluating new methodological contributions. It further streamlines evaluation across multiple datasets by offering data conversion utilities for current public benchmarks. Together, these properties position nnLandmark as a central tool for advancing 3D medical landmark detection through systematic, transparent benchmarking, enabling to genuinely measure methodological progress.

Cite this Paper

BibTeX
@InProceedings{pmlr-v315-ertl26a, title = {nnLandmark: A Self-Configuring Method for 3D Medical Landmark Detection}, author = {Ertl, Alexandra and Denner, Stefan and Peretzke, Robin and Xiao, Shuhan and Zimmerer, David and Fischer, Maximilian and Bujotzek, Markus and Yang, Xin and Neher, Peter and Isensee, Fabian and Maier-Hein, Klaus H.}, booktitle = {Proceedings of The 9th International Conference on Medical Imaging with Deep Learning}, pages = {894--927}, year = {2026}, editor = {Huo, Yuankai and Gao, Mingchen and Kuo, Chang-Fu and Jin, Yueming and Deng, Ruining}, volume = {315}, series = {Proceedings of Machine Learning Research}, month = {08--10 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v315/main/assets/ertl26a/ertl26a.pdf}, url = {https://proceedings.mlr.press/v315/ertl26a.html}, abstract = {Landmark detection is central to many medical applications, such as identifying critical structures for treatment planning or defining control points for biometric measurements. However, manual annotation is labor-intensive and requires expert anatomical knowledge. While deep learning shows promise in automating this task, fair evaluation and interpretation of methods in a broader context, are hindered by limited public benchmarking, inconsistent baseline implementations, and non-standardized experimentation. To overcome these pitfalls, we present nnLandmark, a self-configuring framework for 3D landmark detection that combines tailored heatmap generation, loss design, inference logic, and a robust set of hyperparameters for heatmap regression, while reusing components from nnU-Net’s underlying self-configuration and training engine. nnLandmark achieves state-of-the-art performance across five public and one private dataset, benchmarked against three recently published methods. Its out-of-the-box usability enables training strong landmark detection models on new datasets without expert knowledge or dataset-specific hyperparameter tuning. Beyond accuracy, nnLandmark provides both a strong, common baseline and a flexible, standardized environment for developing and evaluating new methodological contributions. It further streamlines evaluation across multiple datasets by offering data conversion utilities for current public benchmarks. Together, these properties position nnLandmark as a central tool for advancing 3D medical landmark detection through systematic, transparent benchmarking, enabling to genuinely measure methodological progress.} }
Endnote
%0 Conference Paper %T nnLandmark: A Self-Configuring Method for 3D Medical Landmark Detection %A Alexandra Ertl %A Stefan Denner %A Robin Peretzke %A Shuhan Xiao %A David Zimmerer %A Maximilian Fischer %A Markus Bujotzek %A Xin Yang %A Peter Neher %A Fabian Isensee %A Klaus H. Maier-Hein %B Proceedings of The 9th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2026 %E Yuankai Huo %E Mingchen Gao %E Chang-Fu Kuo %E Yueming Jin %E Ruining Deng %F pmlr-v315-ertl26a %I PMLR %P 894--927 %U https://proceedings.mlr.press/v315/ertl26a.html %V 315 %X Landmark detection is central to many medical applications, such as identifying critical structures for treatment planning or defining control points for biometric measurements. However, manual annotation is labor-intensive and requires expert anatomical knowledge. While deep learning shows promise in automating this task, fair evaluation and interpretation of methods in a broader context, are hindered by limited public benchmarking, inconsistent baseline implementations, and non-standardized experimentation. To overcome these pitfalls, we present nnLandmark, a self-configuring framework for 3D landmark detection that combines tailored heatmap generation, loss design, inference logic, and a robust set of hyperparameters for heatmap regression, while reusing components from nnU-Net’s underlying self-configuration and training engine. nnLandmark achieves state-of-the-art performance across five public and one private dataset, benchmarked against three recently published methods. Its out-of-the-box usability enables training strong landmark detection models on new datasets without expert knowledge or dataset-specific hyperparameter tuning. Beyond accuracy, nnLandmark provides both a strong, common baseline and a flexible, standardized environment for developing and evaluating new methodological contributions. It further streamlines evaluation across multiple datasets by offering data conversion utilities for current public benchmarks. Together, these properties position nnLandmark as a central tool for advancing 3D medical landmark detection through systematic, transparent benchmarking, enabling to genuinely measure methodological progress.
APA
Ertl, A., Denner, S., Peretzke, R., Xiao, S., Zimmerer, D., Fischer, M., Bujotzek, M., Yang, X., Neher, P., Isensee, F. & Maier-Hein, K.H.. (2026). nnLandmark: A Self-Configuring Method for 3D Medical Landmark Detection. Proceedings of The 9th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 315:894-927 Available from https://proceedings.mlr.press/v315/ertl26a.html.

Related Material