J
Jens N. Kaftan
Researcher at Siemens
Publications - 38
Citations - 513
Jens N. Kaftan is an academic researcher from Siemens. The author has contributed to research in topics: Segmentation & Image segmentation. The author has an hindex of 14, co-authored 38 publications receiving 494 citations. Previous affiliations of Jens N. Kaftan include RWTH Aachen University & Princeton University.
Papers
More filters
Book ChapterDOI
Automatic multi-organ segmentation using learning-based segmentation and level set optimization
Timo Kohlberger,Michal Sofka,Jingdan Zhang,Neil Birkbeck,Jens Wetzl,Jens N. Kaftan,Jerome Declerck,S. Kevin Zhou +7 more
TL;DR: A novel generic segmentation system for the fully automatic multi-organ segmentation from CT medical images that combines the advantages of learning-based approaches on point cloud-based shape representation with those of PDE-optimization-based level set approaches, such as high accuracy and the straightforward prevention of segment overlaps.
Book ChapterDOI
Multi-stage learning for robust lung segmentation in challenging CT volumes
Michal Sofka,Jens Wetzl,Neil Birkbeck,Jingdan Zhang,Timo Kohlberger,Jens N. Kaftan,Jerome Declerck,S. Kevin Zhou +7 more
TL;DR: A multi-stage learning-based approach that combines anatomical information to predict an initialization of a statistical shape model of the lungs, which is then refined through boundary detection to obtain fine-grained segmentation.
Proceedings ArticleDOI
Fuzzy pulmonary vessel segmentation in contrast enhanced CT data
Jens N. Kaftan,Jens N. Kaftan,Atilla Peter Kiraly,Annemarie Bakai,Marco Das,Carol L. Novak,Til Aach +6 more
TL;DR: This work presents a novel approach to pulmonary vessel segmentation based on a fuzzy segmentation concept, combining the strengths of both threshold and seed point based methods, and focuses on contrast enhanced chest CT data.
Patent
System and method for path based tree matching
TL;DR: In this article, a system and a method for tree matching is presented, which includes: acquiring tree-like structures representing a physical object or model, extracting a path from a first tree-based structure and a path of a second tree-shaped structure, and comparing the paths of the first and second structures by computing a similarity measurement for the paths; and determining if the paths match based on the similarity measurement.
Proceedings ArticleDOI
A novel multipurpose tree and path matching algorithm with application to airway trees
TL;DR: A novel path-based tree matching framework independent of graph matching is presented, based on a point-by-point feature comparison of complete paths rather than branch points, and consequently is relatively unaffected by spurious airways and/or missing branches.