# Deep Learning-Based Regression and Classification for Automatic Landmark Localization in Medical Images

arXiv: https://arxiv.org/abs/2007.05295 · HTML: https://mlsanity.com/paper/2007.05295/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Jul 2020.

## Built on by (most cited first, 22 of 22)

- [U-LanD: Uncertainty-Driven Video Landmark Detection](https://mlsanity.com/paper/2102.01586/) (2021)
- [Learn Fine-grained Adaptive Loss for Multiple Anatomical Landmark Detection in Medical Images](https://mlsanity.com/paper/2105.09124/) (2021)
- [Uncertainty Estimation for Heatmap-based Landmark Localization](https://mlsanity.com/paper/2203.02351/) (2022)
- [Heatmap-based 2D Landmark Detection with a Varying Number of Landmarks](https://mlsanity.com/paper/2101.02737/) (2021)
- [PELE scores: Pelvic X-ray Landmark Detection by Pelvis Extraction and Enhancement](https://mlsanity.com/paper/2305.04294/) (2023)
- [Fine-grained Correlation Loss for Regression](https://mlsanity.com/paper/2207.00347/) (2022)
- [Unsupervised Local Discrimination for Medical Images](https://mlsanity.com/paper/2108.09440/) (2021)
- [Automated computed tomography and magnetic resonance imaging segmentation using deep learning: a beginner's guide](https://mlsanity.com/paper/2304.05901/) (2023)
- [Facial Anatomical Landmark Detection using Regularized Transfer Learning with Application to Fetal Alcohol Syndrome Recognition](https://mlsanity.com/paper/2109.05485/) (2021)
- [Uncertainty-aware U-Net for Medical Landmark Detection](https://mlsanity.com/paper/2303.10349/) (2023)
- [Asynchronous Decentralized Federated Lifelong Learning for Landmark Localization in Medical Imaging](https://mlsanity.com/paper/2303.06783/) (2023)
- [IUP: An Intelligent Utility Prediction Scheme for Solid-State Fermentation in 5G IoT](https://mlsanity.com/paper/2103.15073/) (2021)
- [Localizing Anatomical Landmarks in Ocular Images using Zoom-In Attentive Networks](https://mlsanity.com/paper/2210.02445/) (2022)
- [Multi-environment lifelong deep reinforcement learning for medical imaging](https://mlsanity.com/paper/2306.00188/) (2023)
- [DeepATLAS: One-Shot Localization for Biomedical Data](https://mlsanity.com/paper/2402.09587/) (2024)
- [Aortic root landmark localization with optimal transport loss for heatmap regression](https://mlsanity.com/paper/2407.04921/) (2024)
- [Enhancing Quantitative Image Synthesis through Pretraining and Resolution Scaling for Bone Mineral Density Estimation from a Plain X-ray Image](https://mlsanity.com/paper/2407.20495/) (2024)
- [Temporally Consistent Mitral Annulus Measurements from Sparse Annotations in Echocardiographic Videos](https://mlsanity.com/paper/2503.12087/) (2025)
- [BodyGPS: Anatomical Positioning System](https://mlsanity.com/paper/2505.07744/) (2025)
- [Two-Steps Neural Networks for an Automated Cerebrovascular Landmark Detection](https://mlsanity.com/paper/2507.02349/) (2025)
- [Automatic Landmark-Based Segmentation of Human Subcortical Structures in MRI](https://mlsanity.com/paper/2605.14221/) (2026)
- [Learning To Focus: Anatomy-Guided Attention Regularization for Medical Image Classification](https://mlsanity.com/paper/2607.10851/) (2026)

## Abstract (arXiv)

In this study, we propose a fast and accurate method to automatically localize anatomical landmarks in medical images. We employ a global-to-local localization approach using fully convolutional neural networks (FCNNs). First, a global FCNN localizes multiple landmarks through the analysis of image patches, performing regression and classification simultaneously. In regression, displacement vectors pointing from the center of image patches towards landmark locations are determined. In classification, presence of landmarks of interest in the patch is established. Global landmark locations are obtained by averaging the predicted displacement vectors, where the contribution of each displacement vector is weighted by the posterior classification probability of the patch that it is pointing from. Subsequently, for each landmark localized with global localization, local analysis is performed. Specialized FCNNs refine the global landmark locations by analyzing local sub-images in a similar manner, i.e. by performing regression and classification simultaneously and combining the results. Evaluation was performed through localization of 8 anatomical landmarks in CCTA scans, 2 landmarks in olfactory MR scans, and 19 landmarks in cephalometric X-rays. We demonstrate that the method performs similarly to a second observer and is able to localize landmarks in a diverse set of medical images, differing in image modality, image dimensionality, and anatomical coverage.