Guiding 3D U-nets with signed distance fields for creating 3D models from images

Kristine Aavild Juhl, Rasmus Reinhold Paulsen, Anders Bjorholm Dahl, Vedrana Andersen Dahl, Ole De Backer, Klaus Fuglsang Kofoed, Oscar Camara

Research output: Contribution to conferencePaperResearchpeer-review

75 Downloads (Pure)


Morphological analysis of the left atrial appendage is an important tool to assess risk of ischemic stroke. Most deep learning approaches for 3D segmentation is guided by binary labelmaps, which results in voxelized segmentations unsuitable for morphological analysis. We propose to use signed distance fields to guide a deep network towards morphologically consistent 3D models. The proposed strategy is evaluated on a synthetic dataset of simple geometries, as well as a set of cardiac computed tomography images containing the left atrial appendage. The proposed method produces smooth surfaces with a closer resemblance to the true surface in terms of segmentation overlap and surface distance.
Original languageEnglish
Publication date2019
Number of pages4
Publication statusPublished - 2019
EventMedical Imaging with Deep Learning 2019 - Imperial College’s South Kensington Campus, London, United Kingdom
Duration: 8 Jul 201910 Jul 2019


ConferenceMedical Imaging with Deep Learning 2019
LocationImperial College’s South Kensington Campus
CountryUnited Kingdom
Internet address

Bibliographical note


  • Signed distance fields
  • Pixel-wise regression
  • Left atrial appendage

Cite this

Juhl, K. A., Paulsen, R. R., Dahl, A. B., Dahl, V. A., De Backer, O., Kofoed, K. F., & Camara, O. (2019). Guiding 3D U-nets with signed distance fields for creating 3D models from images. Paper presented at Medical Imaging with Deep Learning 2019, London, United Kingdom.