Semiautomatic segmentation of the heart from CT images based on intensity and morphological features

Abena B. Redwood, Jon J. Camp, Richard A. Robb

Research output: Contribution to journalConference articlepeer-review

6 Scopus citations

Abstract

The incidence of certain types of cardiac arrhythmias is increasing. Effective, minimally invasive treatment has remained elusive. Pharmacologic treatment has been limited by drug intolerance and recurrence of disease. Catheter based ablation has been moderately successful in treating certain types of cardiac arrhythmias, including typical atrial flutter and fibrillation, but there remains a relatively high rate of recurrence. Additional side effects associated with cardiac ablation procedures include stroke, perivascular lung damage, and skin bums caused by x-ray fluoroscopy. Access to patient specific 3-D cardiac images has potential to significantly improve the process of cardiac ablation by providing the physician with a volume visualization of the heart. This would facilitate more effective guidance of the catheter, increase the accuracy of the ablative process, and eliminate or minimize the damage to surrounding tissue. In this study, a semiautomatic method for faithful cardiac segmentation was investigated using Analyze - a comprehensive processing software package developed at the Biomedical Imaging Resource, Mayo Clinic. This method included use of interactive segmentation based on math morphology and separation of the chambers based on morphological connections. The external surfaces of the hearts were readily segmented, while accurate separation of individual chambers was a challenge. Nonetheless, a skilled operator could manage the task in a few minutes. Useful improvements suggested in this paper would give this method a promising future.

Original languageEnglish (US)
Article number198
Pages (from-to)1713-1719
Number of pages7
JournalProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume5747
Issue numberIII
DOIs
StatePublished - 2005
EventMedical Imaging 2005 - Image Processing - San Diego, CA, United States
Duration: Feb 13 2005Feb 17 2005

Keywords

  • Cardiac arrhythmia
  • Catheter ablation
  • Heart segmentation

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Atomic and Molecular Physics, and Optics
  • Radiology Nuclear Medicine and imaging

Fingerprint

Dive into the research topics of 'Semiautomatic segmentation of the heart from CT images based on intensity and morphological features'. Together they form a unique fingerprint.

Cite this