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

2025-04-09

AI Summary

Landmark detection in 3D medical imaging faces challenges including high annotation cost, scarcity of public benchmarks, inconsistent evaluation protocols, and poor model generalizability across modalities. Method: This paper introduces the first end-to-end heatmap regression framework built upon the nnU-Net auto-configuration paradigm, enabling plug-and-play, modality-agnostic anatomical landmark localization across CT and MRI. It eliminates manual hyperparameter tuning and integrates automated preprocessing, multi-scale feature fusion, and Gaussian heatmap-based supervision. Results: On the MML (dental CT) and AFIDs (brain MRI) benchmarks, the method achieves mean radial errors of 1.5 mm and 1.2 mm, respectively—matching inter-expert variability for the first time and substantially outperforming prior approaches. This work establishes a reproducible, generalizable paradigm for medical keypoint detection.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Multi-modal Vision
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Landmark detection plays a crucial role in medical imaging tasks that rely on precise spatial localization, including specific applications in diagnosis, treatment planning, image registration, and surgical navigation. However, manual annotation is labor-intensive and requires expert knowledge. While deep learning shows promise in automating this task, progress is hindered by limited public datasets, inconsistent benchmarks, and non-standardized baselines, restricting reproducibility, fair comparisons, and model generalizability.This work introduces nnLandmark, a self-configuring deep learning framework for 3D medical landmark detection, adapting nnU-Net to perform heatmap-based regression. By leveraging nnU-Net's automated configuration, nnLandmark eliminates the need for manual parameter tuning, offering out-of-the-box usability. It achieves state-of-the-art accuracy across two public datasets, with a mean radial error (MRE) of 1.5 mm on the Mandibular Molar Landmark (MML) dental CT dataset and 1.2 mm for anatomical fiducials on a brain MRI dataset (AFIDs), where nnLandmark aligns with the inter-rater variability of 1.5 mm. With its strong generalization, reproducibility, and ease of deployment, nnLandmark establishes a reliable baseline for 3D landmark detection, supporting research in anatomical localization and clinical workflows that depend on precise landmark identification. The code will be available soon.

Problem

• Automates 3D medical landmark detection without manual tuning
• Addresses limited datasets and inconsistent benchmarks in landmark detection
• Improves accuracy and reproducibility in anatomical localization tasks

Innovation

• Self-configuring deep learning for 3D landmark detection
• Adapts nnU-Net for heatmap-based regression
• Automated parameter tuning without manual intervention