MeningiomaMRIRegistrationStudy - Revision history

Fedorov: /* Quantitative comparison */

2009-09-03T20:02:24Z

Quantitative comparison

Fedorov: /* RigidRegistration Slicer module */

2009-07-29T21:10:34Z

RigidRegistration Slicer module

Fedorov: /* BRAINSFit */

2009-07-29T21:07:34Z

BRAINSFit

Fedorov: /* BRAINSFit */

2009-07-29T21:05:32Z

BRAINSFit

Fedorov at 20:56, 29 July 2009

2009-07-29T20:56:02Z

Fedorov: /* Limitations of the Slicer registration tools */

2009-07-29T20:49:38Z

Limitations of the Slicer registration tools

Fedorov: /* Quantitative comparison */

2009-07-29T20:47:18Z

Quantitative comparison

Fedorov: /* Measures of success */

2009-07-29T20:42:48Z

Measures of success

Fedorov: /* Limitations of the Slicer registration tools */

2009-07-29T20:42:10Z

Limitations of the Slicer registration tools

Fedorov: /* Quantitative comparison */

2009-07-29T20:33:02Z

Quantitative comparison

@@ Line 248: / Line 248: @@
 </gallery>
-Execution times recap:
+Execution times recap (UPDATE: these times are for the code compiled in debug mode! In the release mode, execution time of BRAINSFit is under 1 minute):
 * FLIRT: ~ 4 min

@@ Line 121: / Line 121: @@
    ${FIXED_IMAGE} ${MOVING_IMAGE}
 </pre>
 * execution time: ~4min

@@ Line 173: / Line 173: @@
 ===BRAINSFit===
-A distinctive feature of BRAINSFit tool is the ability to automatically detect head region of interest, which is used for collecting intensity samples during metric calculation. Below is the overlay of such ROI in light blue overlayed with the image for Case10.
+A distinctive feature of BRAINSFit tool is the ability to automatically detect head region of interest, which is used for collecting intensity samples during metric calculation. Below is the overlay of such ROI in light blue overlayed with the image for Case10. This functionality appears to work consistently for all the cases used in the comparison.
 <gallery perrow="3" heights="300px" widths="300px">

@@ Line 172: / Line 172: @@
 ===BRAINSFit===
 ** Case01
@@ Line 200: / Line 208: @@
 Image:Case10-BF-coronal.jpg
 </gallery>
 ===FSL FLIRT===

← Older revision		Revision as of 20:56, 29 July 2009
Line 264:		Line 264:
	* not possible to specify the similarity metric: apparently, FLIRT is able to get good result using cross correlation, which is less demanding computationally		* not possible to specify the similarity metric: apparently, FLIRT is able to get good result using cross correlation, which is less demanding computationally

−	* RegisterImages and RigidRegistration do not allow to specify 9-dof transforms, which are reportedly used to adjust for voxel size inconsistency in longitudinal studies [http://dx.doi.org/10.1016/j.mri.2004.01.057]	+	* RegisterImages and RigidRegistration do not allow to specify 9-dof transforms, which are reportedly used to adjust for voxel size inconsistency in longitudinal studies [1].
		+
		+	=Acknowledgements=
		+
		+	We acknowledge and appreciate the efforts of Hans Johnson, Jim Miller and Stephen Aylward in developing open source registration tools analyzed here. FSL FLIRT tool was used here solely for academic research purposes. We thank Luke Bloy for providing the code to perform conversion of FLIRT transform matrix into the ITK-readable format.
		+
		+	=References=
		+
		+	# J. L. Whitwell, J. M. Schotta, E. B. Lewis, D. G. Macmanus, and N. C. Fox, "Using nine degrees-of-freedom registration to correct for changes in voxel size in serial mri studies," Magnetic Resonance Imaging, vol. 22, no. 7, pp. 993-999, September 2004. [http://dx.doi.org/10.1016/j.mri.2004.01.057]

← Older revision		Revision as of 20:49, 29 July 2009
Line 263:		Line 263:

	* not possible to specify the similarity metric: apparently, FLIRT is able to get good result using cross correlation, which is less demanding computationally		* not possible to specify the similarity metric: apparently, FLIRT is able to get good result using cross correlation, which is less demanding computationally
		+
		+	* RegisterImages and RigidRegistration do not allow to specify 9-dof transforms, which are reportedly used to adjust for voxel size inconsistency in longitudinal studies [http://dx.doi.org/10.1016/j.mri.2004.01.057]

@@ Line 238: / Line 238: @@
 Image:max_reg_tre_wrt_flirt.jpg
 </gallery>
 =Conclusions=

@@ Line 80: / Line 80: @@
 * qualitative assessment: visually pleasing results
-* quantitative assessment: ''surrogate'' Target Registration Error (TRE) statistics collected per-pixel over the segmented tumor volume. In the absence of ground truth for registration result, and considering the wide acceptance and thorough testing of FSL FLIRT registration tool, we accept its result to be optimal under existing operating conditions. We then evaluate each of the registration tools as follows:
+* quantitative assessment: ''relative'' Target Registration Error (TRE) statistics collected per-pixel over the segmented tumor volume. In the absence of ground truth for registration result, and considering the wide acceptance and thorough testing of FSL FLIRT registration tool, we accept its result to be optimal under existing operating conditions. We then evaluate each of the registration tools as follows:
 *# Let T_{flirt} and T_{reg} be the transforms aligning the images being registered recovered with FSL FLIRT and the compared registration tool, respectively
 *# Let T(p) be the coordinates of a point after applying a transform
-*# We then define the registration error surrogate ''e'' as follows: ''e = ||T_{flirt}(p)-T_{reg}(p)||''
+*# We then define the relative registration error ''e'' as follows: ''e = ||T_{flirt}(p)-T_{reg}(p)||''
-*# Given the binary mask defining the tumor, we evaluate the value of registration error surrogate at each pixel corresponding to the tumor volume, and report the values of min, max, and average error values for this region.
+*# Given the binary mask defining the tumor, we evaluate the value of registration error surrogate at each pixel corresponding to the tumor volume, and report the values of min, max, and average relative error values for this region.
 * minimum execution time to meet the application requirements: interactive quantification of tumor growth
 * minimum tuning of registration parameters (is this possible? to answer this sub-question we compare the results with FLIRT registration using its default parameters)

← Older revision		Revision as of 20:42, 29 July 2009
Line 255:		Line 255:
	* all of the tools implemented in Slicer require to specify the max number of iterations, which essentially requires the user to monitor convergence (metric value while iterating can be found in Slicer error log), and increase the number of iterations if the satisfactory result is not achieved. Not practical for interactive application		* all of the tools implemented in Slicer require to specify the max number of iterations, which essentially requires the user to monitor convergence (metric value while iterating can be found in Slicer error log), and increase the number of iterations if the satisfactory result is not achieved. Not practical for interactive application

−	* not possible to specify the similarity metric: apparently, FLIRT is able to get good result using cross correlation	+	* not possible to specify the similarity metric: apparently, FLIRT is able to get good result using cross correlation, which is less demanding computationally

@@ Line 233: / Line 233: @@
 ==Quantitative comparison==
-<gallery perrow="3" heights="300px" widths="300px">
+<gallery perrow="3" heights="300px" widths="400px">
 Image:min_reg_tre_wrt_flirt.jpg
 Image:ave_reg_tre_wrt_flirt.jpg