Automatic Whole Heart Segmentation in Static Magnetic Resonance Image Volumes

Peters, Jörg; Ecabert, Olivier; Meyer, Carsten; Schramm, Hauke; Kneser, Reinhard; Groth, Alexandra; Weese, Jürgen

doi:10.1007/978-3-540-75759-7_49

Cited by 52 publications

(56 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Anatomy recognition relies on the model-based segmentation approach, using a triangulated surface model composed of the seven major parts of the heart and large thoracic vessels (8)(9)(10). After roughly positioning the model in the image center, segmentation is conducted in two steps: first, pose optimization using global similarity transformations and localized affine transformations, and second, energy minimizing freeform deformation using sequence-specifically trained, locally varying boundary descriptors (11).…”

Section: Anatomy Recognition and Automated Cardiac Mr Geometry Planningmentioning

confidence: 99%

Fully automatic geometry planning for cardiac MR imaging and reproducibility of functional cardiac parameters

Frick¹,

Paetsch

Harder

et al. 2011

Magnetic Resonance Imaging

View full text Add to dashboard Cite

Purpose: To establish operator-independent, fully automated planning of standard cardiac geometries and to determine the impact on interstudy reproducibility of cardiac functional parameters.Materials and Methods: Cardiac MR imaging was done in 50 patients referred for left-ventricular function assessment. In all patients, first standard manual planning was performed followed by automatic planning (AUTO1) and repeat automatic planning (AUTO2) after repositioning the patient to investigate interstudy reproducibility. Cardiac functional parameters were assessed and cine scans were visually graded on a 4-point scale from nondiagnostic to excellent.Results: Overall success rate of AUTO was 94% with good to excellent geometry planning in >94% of cine standard views. Comparing manual versus fully automated planning, a high agreement of cardiac functional parameters (Lin's concordance correlation coefficient, 0.91 to 0.99) with minimal percent bias (0.24 to 3.84%) was found. In addition, a high interstudy reproducibility of automatic planning was demonstrated (Lin's concordance correlation coefficient, 0.89 to 0.99; percent bias, 0.38 to 5.04%; precision, 3.46 to 9.09%). Conclusion:Fully automated planning of cardiac geometries could reliably be performed in patients showing a variety of cardiovascular pathologies. Standard cardiac geometries were precisely replicated and functional parameters were highly accurate.

show abstract

Section: Anatomy Recognition and Automated Cardiac Mr Geometry Planningmentioning

confidence: 99%

Fully automatic geometry planning for cardiac MR imaging and reproducibility of functional cardiac parameters

Frick¹,

Paetsch

Harder

et al. 2011

Magnetic Resonance Imaging

View full text Add to dashboard Cite

show abstract

“…Fully automatic whole-heart segmentation (all cardiac chambers, including the left ventricular myocardium, and the great vessels) has been reported using statistical shape models (SSMs, see [20] for a detailed review of SSMs and their use for segmentation) and atlas-based methods. These methods have been applied to both CT [21••] and MR [22][23][24] image data ( Fig. 1) and more recently to 3D echo data [25].…”

Section: Anatomical Imaging and Segmentationmentioning

confidence: 99%

“…These systems are able to reconstruct the geometry Fig. 1 Fully automatic segmentation result using the method of Peters et al [22]. A highresolution whole-heart SSFP MR dataset is shown in multiplanar view (left) with segmented boundaries shown in red.…”

Section: Anatomical Imaging and Segmentationmentioning

confidence: 99%

Modeling and Registration for Electrophysiology Procedures Based on Three-Dimensional Imaging

Rhode

Sermesant

2011

Curr Cardiovasc Imaging Rep

View full text Add to dashboard Cite

Computer models of cardiac electrophysiology (EP) can help to better understand the mechanisms of arrhythmias and to guide interventions. However, model adjustment to patient data (personalization) is a required step that is still challenging from clinical data. The progress in the fusion of multimodal data opens up new possibilities in generating patient-specific models of the heart. In this paper, we present the state-of-the-art in multimodal data fusion for EP procedure guidance and how such integrated data can be used to personalize models and guide interventions.

show abstract

“…There is important literature on cardiac image analysis for the segmentation of the heart from medical images [11][12][13][14][15][16][17][18][19][20][21]. The idea is not to be exhaustive but to present a generic pipeline where most of the approaches can fit.…”

Section: Patient-specific Myocardial Shapementioning

confidence: 99%