Aktivitäten pro Jahr
Abstract
In many medical image analysis applications, only a limited amount of training data is available due to the costs of image acquisition and the large manual annotation effort required from experts. Training recent state-of-the-art machine learning methods like convolutional neural networks (CNNs) from small datasets is a challenging task. 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 overall need for large training datasets. Our fully convolutional SpatialConfiguration-Net (SCN) learns this simplification due to multiplying the heatmap predictions of its two components and by training the network in an end-to-end manner. Thus, the SCN dedicates one component to locally accurate but ambiguous candidate predictions, while the other component improves robustness to ambiguities by incorporating the …
Originalsprache | englisch |
---|---|
Seiten (von - bis) | 207-219 |
Fachzeitschrift | Medical Image Analysis |
Jahrgang | 54 |
DOIs | |
Publikationsstatus | Veröffentlicht - 2019 |
Fingerprint
Untersuchen Sie die Forschungsthemen von „Integrating spatial configuration into heatmap regression based CNNs for landmark localization“. Zusammen bilden sie einen einzigartigen Fingerprint.-
Integrating Spatial Configuration into CNNs for Anatomical Landmark Localization and Multi-Label Whole Heart Segmentation
Christian Payer (Redner/in)
2019Aktivität: Vortrag oder Präsentation › Invited talk bei Konferenz oder Fachtagung › Science to science
-
Integrating Spatial Configuration into Heatmap Regression Based CNNs for Landmark Localization
Christian Payer (Redner/in)
2019Aktivität: Vortrag oder Präsentation › Posterpräsentation › Science to science