A validation benchmark for assessment of medial surface quality for medical applications

  • Authors:
  • Agnés Borràs;Debora Gil;Sergio Vera;Miguel Angel González

  • Affiliations:
  • Computer Vision Center, Comp. Science Dep., UAB, Spain,Alma I.T. Systems, Barcelona, Spain;Computer Vision Center, Comp. Science Dep., UAB, Spain,Alma I.T. Systems, Barcelona, Spain;Computer Vision Center, Comp. Science Dep., UAB, Spain,Alma I.T. Systems, Barcelona, Spain;Computer Vision Center, Comp. Science Dep., UAB, Spain,Alma I.T. Systems, Barcelona, Spain

  • Venue:
  • ICVS'13 Proceedings of the 9th international conference on Computer Vision Systems
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

Confident use of medial surfaces in medical decision support systems requires evaluating their quality for detecting pathological deformations and describing anatomical volumes. Validation in the medical imaging field is a challenging task mainly due to the difficulties for getting consensual ground truth. In this paper we propose a validation benchmark for assessing medial surfaces in the context of medical applications. Our benchmark includes a home-made database of synthetic medial surfaces and volumes and specific scores for evaluating surface accuracy, its stability against volume deformations and its capabilities for accurate reconstruction of anatomical volumes.