Modules:LabelDiameterEstimation-Documentation-3.6

From Slicer Wiki
Revision as of 22:07, 18 May 2010 by Fedorov (talk | contribs) (Created page with 'Return to Slicer 3.6 Documentation __NOTOC__ ===Label Diameter Estimation=== LabelDiameterEstimation {| |[[Image:labeldiameterestimation1.png|thumb|280px|C…')
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search
Home < Modules:LabelDiameterEstimation-Documentation-3.6

Return to Slicer 3.6 Documentation

Label Diameter Estimation

LabelDiameterEstimation

Caption 1

General Information

Module Type & Category

Type: CLI

Category: Statistics

Authors, Collaborators & Contact

  • Andriy Fedorov, BWH
  • Ron Kikinis, BWH
  • Contact: Andriy Fedorov, fedorov at bwh

Module Description

Given a binary label, this module estimates the largest diameter of the label, its largest in-plane perpendicular diameter, and the third diameter that is perpendicular to the plane of the first two.

Usage

Examples, Use Cases & Tutorials

This module was originally designed to calculate the dimensions of tumor, following the RECIST [1] and WHO guidelines.

  • Implementation details The module implements the following algorithm for calculating the dimensions of the label. First, the area of the label slice is computed in all the slices of the image that are passing through the label. This is done for the three orthogonal directions of the image. The largest area slice is found. The largest diameter (DA) is estimated by finding the two most distant points on the contour of the label cross-section. The second diameter (DB) is found by finding the two most distant points on the contour that lie on the line perpendicular to the first diameter. The third diameter (DC) is estimated by calculating the two points of intersection between the line perpendicular to the plane formed by DA and DBpassing through the point of their intersection, and the contour of the analyzed label.

Quick Tour of Features and Use

  • Control panel contains the following items
    • Label value specifies which label should be used for diameter computation. When your label image has multiple labels, you can choose which one you want to measure. The default value of 0 will force the module to treat all label values in (1,255) as a single label.
    • Input volume specifies the image with the label to be measured
    • Output label image is a label image, which will have the computed diameters represented by distinctive labels. The diameters from the longest to the shortest are represented by consecutive labels from 1 to 3.
    • File to keep the output fiducials must be specified to store the endpoints of the diameters. After computation is completed, the fiducials can be loaded from the File -> Add data... menu. The fiducials are named as DA-0 and DA-1 for the endpoints of the longest diameter, DB-0/DB-1 and DC-0/DC-1 for the other two.
  • Advanced parameters
    • Orthogonality tolerance You should change this parameter only if the module does not give you a meaningful answer with the default value. The larger the tolerance is, the more non-perpendicular the diameters are allowed to be. This is related to the implementation details of the module. Try increasing the tolerance from 0.01 to 0.1 if the results don't make sense.

The module also outputs numeric values for the lengths of the computed diameters. To see these numbers, open Log Viewer, and select the first from top log record with the name Label Diameter Estimation: standard output. The reported numbers are the lenghts of the three diameters, the volume estimate calculated as A*B*C/2 (a formula suggested in the literature for estimating volume of hematomas from planimetric measurements [2]), and the true label volume in mm^3.

LabelDiameterEstimation control panel
Input label outline and the extracted diameters. Note the largest diameter is in blue (label 1), second largest is pink (label 2).
LabelDiameterEstimation standard output in "Log Viewer" reports lenghts of the computed diameters
Fiducials of the diameter endpoints and the model constructed from the input label

Development

Standing issues under development:

  • would be nice to have fiducials read automatically; currently not possible, since fiducials are not fully integrated with CLI
  • use measurements widgets instead of label images for diameter visualization; measurements are also not integrated with CLI, and are not yet feature-complete
  • update Acks tab to point to documentation and give credit to BSF

Dependencies

Known bugs

Follow this link to the Slicer3 bug tracker.


Usability issues

Follow this link to the Slicer3 bug tracker. Please select the usability issue category when browsing or contributing.

Source code & documentation

Source code can accessed here

Links to documentation generated by doxygen.

More Information

Acknowledgment

Supported by Brain Science Foundation.

References

[1] E. Eisenhauer, P. Therasse, J. Bogaerts, L. Schwartz, D. Sargent, R. Ford, J. Dancey, S. Arbuck, S. Gwyther, and M. Mooney, "New response evaluation criteria in solid tumours: Revised recist guideline (version 1.1)," European Journal of Cancer, vol. 45, no. 2, pp. 228-247, January 2009. DOI

[2] H. B. Huttner, T. Steiner, M. Hartmann, M. Kohrmann, E. Juettler, S. Mueller, J. Wikner, U. Meyding-Lamade, P. Schramm, S. Schwab, and P. D. Schellinger, "Comparison of abc/2 estimation technique to computer-assisted planimetric analysis in warfarin-related intracerebral parenchymal hemorrhage," Stroke, vol. 37, no. 2, pp. 404-408, February 2006. DOI