User-guided 3D active contour segmentation of anatomical structures: Significantly improved efficiency and reliability

Penn Image Computing and Science Laboratory, Department of Radiology, University of Pennsylvania, PA 19104-6274, USA.
NeuroImage (Impact Factor: 6.36). 08/2006; 31(3):1116-28. DOI: 10.1016/j.neuroimage.2006.01.015
Source: PubMed

ABSTRACT Active contour segmentation and its robust implementation using level set methods are well-established theoretical approaches that have been studied thoroughly in the image analysis literature. Despite the existence of these powerful segmentation methods, the needs of clinical research continue to be fulfilled, to a large extent, using slice-by-slice manual tracing. To bridge the gap between methodological advances and clinical routine, we developed an open source application called ITK-SNAP, which is intended to make level set segmentation easily accessible to a wide range of users, including those with little or no mathematical expertise. This paper describes the methods and software engineering philosophy behind this new tool and provides the results of validation experiments performed in the context of an ongoing child autism neuroimaging study. The validation establishes SNAP intrarater and interrater reliability and overlap error statistics for the caudate nucleus and finds that SNAP is a highly reliable and efficient alternative to manual tracing. Analogous results for lateral ventricle segmentation are provided.

Download full-text


Available from: Heather Cody, Sep 26, 2015
1 Follower
110 Reads
    • "3D segmentation of the Fontan geomety was performed manually on the timeaveraged magnitude images using an open source segmentation tool, ITK-SNAP (Yushkvich et al., 2006). The segmentations included inferior vena cava (IVC), superior vena cava (SVC) and left and right pulmonary arteries (LPA and RPA) with the segmental branches excluded. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Viscous dissipation inside Fontan circulation, a parameter associated with the exercise intolerance of Fontan patients, can be derived from computational fluid dynamics (CFD) or 4D flow MRI velocities. However, the impact of spatial resolution and measurement noise on the estimation of viscous dissipation is unclear. Our aim was to evaluate the influence of these parameters on viscous dissipation calculation. Six Fontan patients underwent whole heart 4D flow MRI. Subject-specific CFD simulations were performed. The CFD velocities were down-sampled to isotropic spatial resolutions of 0.5mm, 1mm, 2mm and to MRI resolution. Viscous dissipation was compared between (1) high resolution CFD velocities, (2) CFD velocities down-sampled to MRI resolution, (3) down-sampled CFD velocities with MRI mimicked noise levels, and (4) in-vivo 4D flow MRI velocities. Relative viscous dissipation between subjects was also calculated. 4D flow MRI velocities (15.6±3.8cm/s) were higher, although not significantly different than CFD velocities (13.8±4.7cm/s, p=0.16), down-sampled CFD velocities (12.3±4.4cm/s, p=0.06) and the down-sampled CFD velocities with noise (13.2±4.2cm/s, p=0.06). CFD-based viscous dissipation (0.81±0.55mW) was significantly higher than those based on down-sampled CFD (0.25±0.19mW, p=0.03), down-sampled CFD with noise (0.49±0.26mW, p=0.03) and 4D flow MRI (0.56±0.28mW, p=0.06). Nevertheless, relative viscous dissipation between different subjects was maintained irrespective of resolution and noise, suggesting that comparison of viscous dissipation between patients is still possible. Copyright © 2015 Elsevier Ltd. All rights reserved.
    Journal of Biomechanics 08/2015; DOI:10.1016/j.jbiomech.2015.07.039 · 2.75 Impact Factor
    • "The above-mentioned data set was then imported into a software for image segmentation (; Yushkevich et al. 2006). Image segmentation refers to a process of examining cross sections of a volumetric data set and outlining the structures of interest visible in these cross sections. "
    [Show abstract] [Hide abstract]
    ABSTRACT: This case series investigated by means of CBCT, buccal bone three-dimensional anatomy at delayed, two-stage implants in the maxillary incisal tooth region. Moreover, the relation between buccal bone anatomy and soft tissue aesthetics was assessed. Twelve implants were analysed after on average 8.9 years in function. Baseline and re-evaluation photographs were assessed using the pink aesthetic score (PES). Marginal bone changes were measured from intraoral X-rays. The buccal bone volume associated with the implant and the implant surface not covered by visible buccal bone was computed on CBCT data sets. Buccal bone thickness and level were assessed, as well as the thickness of the crest distally and mesially of the implant. Changes in soft tissue forms and correlation between aesthetics and bone anatomy were calculated by nonparametric statistics. Buccal bone level was located 3.8 mm apical of the implant shoulder, and none of the implants had complete bone coverage. Buccal bone volume was 144.3 mm(3) , and 4.29 mm(3) in the more coronal 2 mm portion. PES did not differ at re-evaluation (9.7) and baseline (9.2). PES was directly correlated with crestal thickness mesially and distally of the implant shoulder. No other significant correlations were observed between bone anatomy and PES or buccal peri-implant health. Marginal bone gain over time was associated with greater coronal bone volume buccally and with greater buccal and marginal bone thickness, while loss was related to less or no bone. Within present limitations, acceptable and stable aesthetics are not jeopardized by a thin or missing buccal bone. © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.
    Clinical Oral Implants Research 07/2015; DOI:10.1111/clr.12664 · 3.89 Impact Factor
  • Source
    • "Our processing pipeline to evaluate the proposed method using the Hi-res 3D LGE-CMR images of patient hearts is shown in Fig. 1. Initially, an expert manually segmented leftventricular (LV) infarct regions in a given 3D image using ITK-SNAP [35]. As mentioned previously, infarct tissue comprises of scar (also known as infarct core) and semi-viable 250 myocardium (or border zone) [6], and it is important to represent the two regions differently in electrophysiological models [10]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: Accurate three-dimensional (3D) reconstruction of myocardial infarct geometry is crucial to patient-specific modeling of the heart aimed at providing therapeutic guidance in ischemic cardiomyopathy. However, myocardial infarct imaging is clinically performed using two-dimensional (2D) late-gadolinium enhanced cardiac magnetic resonance (LGE-CMR) techniques, and a method to build accurate 3D infarct reconstructions from the 2D LGE-CMR images has been lacking. The purpose of this study was to address this need.
    Medical Physics 07/2015; 42(8):4579-4590. DOI:10.1118/1.4926428 · 2.64 Impact Factor
Show more