.IM (MIDAS) file format load: performed corrections
a) SwapOrder, SignedShort flags were uninitialized when the format is different from ANALYZE.
b) File size defect in the VA_RAW_FileUINT16interval function
Search This Blog
Wednesday, August 31, 2016
Saturday, August 27, 2016
Rat Brain registration: resolved the quality problem by adding and additional (5th) landmark at the back of the brain in the Eyes Plain. Apparently the 4 Landmark configuration was too close to a Coplanar that was causing the Affine registration to fail.
![]() |
| Additional BrainBack landmark |
![]() |
| Affine registration result: Transformed CT overlaid by the MR-template landmarks. |
![]() |
| CT layer (Green) overlaid with transparency on top of the MR layer (Red) |
Thursday, August 25, 2016
BiCal Smooth preprocess: Changed the algorithm and the User Interface item to a simple checkbox (ON\OFF). This finally makes BiCal with all parameter Quasi-constants. I.e. User can run the BiCal on all volumes with very good result without changing parameter. The eliminated SmoothPower parameter was extremely challenging to replace.
Friday, August 19, 2016
Thursday, August 18, 2016
Tuesday, August 16, 2016
Edge-Constrained volume smooth (sometimes called ObjectSmooth or QuasiSegmentation in FireVoxel): implemented a Zero-Edge variant when the thickness of the edges is 0 instead of current 1-voxel. This allows not to use artificially higher resolution for better result.
This is one of the principal components of the BiCal. But also it has a big potential to improve other operations, possibly MagTrace, and as as the pre-processing for more complex segmentations.
![]() |
| Smoothed power=80 |
Saturday, August 6, 2016
BiCal functionality:
"Low signal level" parameter has replaced the "Air Threshold" parameter. The new parameter is quasi-constant ( in [5,10] range) and is independent of the volume content.
Separated the notion of Air Threshold and the Signal-to-Gradient ration. They produce separate masks now. Air-Threshold can be defined very approximately now and completely automatically without User Input, as it is mostly used cosmetically to make the resulting volume to look without Air amplification. Another component, the areas of the low Signal-to-Gradient ratio are excluded and this improves the Non-Uniformity measures substantially.
For the reference image, BPD-7T, with the same parameters the measures for
Build 223: WM\GM\CSM improved from Build 224: 0.14\0.16\0.045 down to 0.07\0.14\0.031
"Low signal level" parameter has replaced the "Air Threshold" parameter. The new parameter is quasi-constant ( in [5,10] range) and is independent of the volume content.
Separated the notion of Air Threshold and the Signal-to-Gradient ration. They produce separate masks now. Air-Threshold can be defined very approximately now and completely automatically without User Input, as it is mostly used cosmetically to make the resulting volume to look without Air amplification. Another component, the areas of the low Signal-to-Gradient ratio are excluded and this improves the Non-Uniformity measures substantially.
For the reference image, BPD-7T, with the same parameters the measures for
Build 223: WM\GM\CSM improved from Build 224: 0.14\0.16\0.045 down to 0.07\0.14\0.031
![]() |
| ORIGINAL |
![]() |
| After BICAL |
Wednesday, August 3, 2016
Tuesday, August 2, 2016
Monday, August 1, 2016
BiCal profiling result (see diagram below):
1. Starts with Edge detection - virtually 100% parallel.
2. Then processing start and separate peaks corresponds to iterations
with the increasing degree of Legendre polynomials.
3. As degree grows, the SVD (= fully parallel) portion grows so the total
core load grows too.
I will be addressing the non-parallelized part discovered with VC++ Profiler.
1. Starts with Edge detection - virtually 100% parallel.
2. Then processing start and separate peaks corresponds to iterations
with the increasing degree of Legendre polynomials.
3. As degree grows, the SVD (= fully parallel) portion grows so the total
core load grows too.
I will be addressing the non-parallelized part discovered with VC++ Profiler.
Subscribe to:
Posts (Atom)











