Loading [a11y]/accessibility-menu.js
ITK-SNAP: An Intractive Medical Image Segmentation Tool to Meet the Need for Expert-Guided Segmentation of Complex Medical Images | IEEE Journals & Magazine | IEEE Xplore

ITK-SNAP: An Intractive Medical Image Segmentation Tool to Meet the Need for Expert-Guided Segmentation of Complex Medical Images


Abstract:

Imaging is a crucial tool in medicine and biomedical research. Magnetic resonance imaging (MRI), computational tomography (CT), proton emission tomography (PET), and ultr...Show More

Abstract:

Imaging is a crucial tool in medicine and biomedical research. Magnetic resonance imaging (MRI), computational tomography (CT), proton emission tomography (PET), and ultrasound are routinely used not only to diagnose disease but also to plan and guide surgical interventions, track disease progression, measure the response of the body to treatment, and understand how genetic and environmental factors relate to anatomical and functional phenotypes.
Published in: IEEE Pulse ( Volume: 8, Issue: 4, July-Aug. 2017)
Page(s): 54 - 57
Date of Publication: 13 July 2017

ISSN Information:

PubMed ID: 28715317

ITK-SNAP Vision

To meet this need across a range of application domains and imaging modalities, we have developed the interactive software tool ITK-SNAP for medical image segmentation. ITK-SNAP provides an environment for visualizing complex 3-D and four-dimensional (4-D) imaging data sets, coupled with a set of tools for the efficient, reliable labeling of structures of interest. The visualization environment offers linked visualization of two-dimensional (2-D) image cross sections and 3-D surface representations to aid spatial navigation of 3-D images. In addition to basic manual tools for outlining anatomy, ITK-SNAP provides computer-assisted tools that allow the expert to focus on the big-picture anatomical decisions during segmentation, while delegating the repetitive aspects of segmentation to the computer (Figure 1).

The main ITK-SNAP window in the course of brain tumor segmentation from a multimodality MRI data set. The user interface includes 2-D and 3-D visualization of the image data set and segmentation. The segmentation of the tumor and edema illustrated in this figure was generated using the semiautomatic segmentation mode in about 15 min. (Tumor image data are from the Medical Image Computing and Computer Assisted Intervention 2013 Multimodal Brain Tumor Segmentation (BRATS) challenge [11]. Tumor image data publicly available from the ITK-SNAP web page, http://www.itksnap.org.)

Contact IEEE to Subscribe

References

References is not available for this document.