Efficient liver segmentation in CT images based on graph cuts and bottleneck detection

Highlights • The pixel-, patch-based and inter-slice based features are used for segmentation. • A novel boundary refinement method is proposed to avoid over-segmentations. • Our method outperforms several existing methods on liver segmentation. • Our method can also be applied in some other abdomin...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	Physica medica 2016-11, Vol.32 (11), p.1383-1396
Hauptverfasser:	Liao, Miao, Zhao, Yu-qian, Wang, Wei, Zeng, Ye-zhan, Yang, Qing, Shih, Frank Y, Zou, Bei-ji
Format:	Artikel
Sprache:	eng
Schlagworte:	Algorithms Bottleneck detection Databases, Factual Gaussian fitting Graph cuts Humans Image Processing, Computer-Assisted - methods Liver - anatomy & histology Liver - diagnostic imaging Liver segmentation PCA Radiology Tomography, X-Ray Computed
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Beschreibung
Zusammenfassung:	Highlights • The pixel-, patch-based and inter-slice based features are used for segmentation. • A novel boundary refinement method is proposed to avoid over-segmentations. • Our method outperforms several existing methods on liver segmentation. • Our method can also be applied in some other abdominal organ segmentation.
ISSN:	1120-1797 1724-191X
DOI:	10.1016/j.ejmp.2016.10.002