NUDF - Neural Unsigned Distance Fields for High Resolution 3D Medical Image Segmentation

Kristine Sorensen, Oscar Camara, Ole De Backer, Klaus F. Kofoed, Rasmus R. Paulsen

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review


Medical image segmentation is often considered as the task of labelling each pixel or voxel as being inside or outside a given anatomy. Processing the images at their original size and resolution often result in insuperable memory requirements, but downsampling the images leads to a loss of important details. Instead of aiming to represent a smooth and continuous surface in a binary voxel-grid, we propose to learn a Neural Unsigned Distance Field (NUDF) directly from the image. The small memory requirements of NUDF allow for high resolution processing, while the continuous nature of the distance field allows us to create high resolution 3D mesh models of shapes of any topology (i.e. open surfaces). We evaluate our method on the task of left atrial appendage (LAA) segmentation from Computed Tomography (CT) images. The LAA is a complex and highly variable shape, being thus difficult to represent with traditional segmentation methods using discrete labelmaps. With our proposed method, we are able to predict 3D mesh models that capture the details of the LAA and achieve accuracy in the order of the voxel spacing in the CT images.

Original languageEnglish
Title of host publicationProceedings of 19th IEEE International Symposium on Biomedical Imaging
Number of pages5
Publication date2022
ISBN (Electronic)9781665429238
Publication statusPublished - 2022
Event19th IEEE International Symposium on Biomedical Imaging - ITC Royal Bengal, Kolkata, India
Duration: 28 Mar 202231 Mar 2022


Conference19th IEEE International Symposium on Biomedical Imaging
LocationITC Royal Bengal
Internet address


  • Computed tomography
  • Image segmentation
  • Left atrial appendage
  • Mesh modelling
  • Unsigned distance fields


Dive into the research topics of 'NUDF - Neural Unsigned Distance Fields for High Resolution 3D Medical Image Segmentation'. Together they form a unique fingerprint.

Cite this