Validation of Liver Tumor Segmentation in CT Scans by Relating Manual and Algorithmic Performance - A Preliminary Study

Jan Hendrik Moltz, Jan Rühaak, Christiane Engel, Ulrike Kayser, Heinz-Otto Peitgen

Abstract

The development of segmentation algorithms for liver tumors in CT scans has found growing attention in recent years. The validation of these methods, however, is often treated as a subordinate task. In this article, we review existing approaches and present rst steps towards a new methodology that evaluates the quality of an algorithm in rela- tion to the variability of manual delineations. We obtained three manual segmentations for 50 liver lesions and computed the results of a segmen- tation algorithm. We compared all four masks with each other and with di erent ground truth estimates and calculated scores according to the validation framework from the MICCAI challenge 2008. Our results show some cases where this more elaborate evaluation re ects the segmenta- tion quality in a more adequate way than traditional approaches. The concepts can also be extended to other similar segmentation problems.
Original languageEnglish
Number of pages8
Publication statusPublished - 01.09.2010
EventMICCAI Workshop on Computational Imaging Biomarkers for Tumors 2010: CIBT 2010 - Beijing, China
Duration: 20.09.201024.09.2010

Conference

ConferenceMICCAI Workshop on Computational Imaging Biomarkers for Tumors 2010
Country/TerritoryChina
CityBeijing
Period20.09.1024.09.10

Fingerprint

Dive into the research topics of 'Validation of Liver Tumor Segmentation in CT Scans by Relating Manual and Algorithmic Performance - A Preliminary Study'. Together they form a unique fingerprint.

Cite this