Detail publikace

Matlab Extension for 3DSlicer: A Robust MR Image Processing Tool

Originální název

Matlab Extension for 3DSlicer: A Robust MR Image Processing Tool

Anglický název

Matlab Extension for 3DSlicer: A Robust MR Image Processing Tool

Jazyk

en

Originální abstrakt

Using magnetic resonance tomography to scan biological tissues is currently a very dynamic approach Based on various image parameters, the method enables us to analyze tissue properties, recognize healthy and pathological tissues, and diagnose the disease or indicate its progression. However, the acquired data must be correctly interpreted and visualized by means of a suitable software tool, such as 3DSlicer (http://www.slicer.org). This well-designed platform provides an interface between the user and the data available in the popular DICOM format, and it facilitates very simple 3D visualization of the MR-based data. One of the main advantages of the open-source software package is undoubtedly its ability to be extended with supplementary modules, for example the Matlab script. The paper describes the open-source environment with a focus on Slicer3D and introduces possible extension of this platform with a module for MR data processing via the three-dimensional, mutiparametric, SVM trainable segmentation method. The module is freely downloadable. The paper also presents a comparison of the processing results with respect to the cycle time and the necessary interactivity. Moreover, the author proposes multiparametric segmentation of a brain tumor edema from T1 and T2-weighted images, and the advantages of the SVM technique are compared with corresponding features of both other fast segmentation methods and the one-parameter approach.

Anglický abstrakt

Using magnetic resonance tomography to scan biological tissues is currently a very dynamic approach Based on various image parameters, the method enables us to analyze tissue properties, recognize healthy and pathological tissues, and diagnose the disease or indicate its progression. However, the acquired data must be correctly interpreted and visualized by means of a suitable software tool, such as 3DSlicer (http://www.slicer.org). This well-designed platform provides an interface between the user and the data available in the popular DICOM format, and it facilitates very simple 3D visualization of the MR-based data. One of the main advantages of the open-source software package is undoubtedly its ability to be extended with supplementary modules, for example the Matlab script. The paper describes the open-source environment with a focus on Slicer3D and introduces possible extension of this platform with a module for MR data processing via the three-dimensional, mutiparametric, SVM trainable segmentation method. The module is freely downloadable. The paper also presents a comparison of the processing results with respect to the cycle time and the necessary interactivity. Moreover, the author proposes multiparametric segmentation of a brain tumor edema from T1 and T2-weighted images, and the advantages of the SVM technique are compared with corresponding features of both other fast segmentation methods and the one-parameter approach.

BibTex


@inproceedings{BUT109611,
  author="Jan {Mikulka}",
  title="Matlab Extension for 3DSlicer: A Robust MR Image Processing Tool",
  annote="Using magnetic resonance tomography to scan biological tissues is currently a very dynamic approach Based on various image parameters, the method enables us to analyze tissue properties, recognize healthy and pathological tissues, and diagnose the disease or indicate its progression. However, the acquired data must be correctly interpreted and visualized by means of a suitable software tool, such as 3DSlicer (http://www.slicer.org). This well-designed platform provides an interface between the user and the data available in the popular DICOM format, and it facilitates very simple 3D visualization of the MR-based data. One of the main advantages of the open-source software package is undoubtedly its ability to be extended with supplementary modules, for example the Matlab script. The paper describes the open-source environment with a focus on Slicer3D and introduces possible extension of this platform with a module for MR data processing via the three-dimensional, mutiparametric, SVM trainable segmentation method. The module is freely downloadable. The paper also presents a comparison of the processing results with respect to the cycle time and the necessary interactivity. Moreover, the author proposes multiparametric segmentation of a brain tumor edema from T1 and T2-weighted images, and the advantages of the SVM technique are compared with corresponding features of both other fast segmentation methods and the one-parameter approach.",
  booktitle="Proceedings of PIERS 2014 in Guangzhou",
  chapter="109611",
  howpublished="online",
  year="2014",
  month="september",
  pages="1857--1860",
  type="conference paper"
}