Integrating Spatial Configuration into Heatmap Regression Based CNNs for Landmark Localization
This addresses the problem of accurate landmark localization in medical imaging for practitioners, but it is incremental as it builds on existing heatmap regression methods.
The paper tackles anatomical landmark localization with limited training data by proposing a CNN architecture that splits the task into two sub-problems, reducing data needs and achieving lower localization error compared to related methods on size-limited datasets.
In many medical image analysis applications, often only a limited amount of training data is available, which makes training of convolutional neural networks (CNNs) challenging. In this work on anatomical landmark localization, we propose a CNN architecture that learns to split the localization task into two simpler sub-problems, reducing the need for large training datasets. Our fully convolutional SpatialConfiguration-Net (SCN) dedicates one component to locally accurate but ambiguous candidate predictions, while the other component improves robustness to ambiguities by incorporating the spatial configuration of landmarks. In our experimental evaluation, we show that the proposed SCN outperforms related methods in terms of landmark localization error on size-limited datasets.