An efficient approach for large deformation diffeomorphic metric mapping (LDDMM) for brain images by utilizing GPU-based parallel computing and a mixture automatic step size estimation method for gradient descent (MAS-GD)
The main dependency External libraries:
- Insight Segmentation and Registration Toolkit (ITK) -- 5.0
- SimpleITK
- CUDA Toolkit 10.0
Ubuntu Xenial 16.04.3 LTS (with 8 GB RAM and NVIDIA GPU with not less than 2 GB global memory).
The lddmmMASGD is recommanded run after the histogram matching and affine registration.
After the configration of system and dependencies, you can run it in the terminal:
xxxxx$./lddmmMASGD
Usage:
./lddmmMASGD --in InputPath --ref ReferencePath --out OutputPath
field OutputDisplacementFieldPath
--scale Scale
--alpha RegistrationSmoothness
--sigma Sigma
--epsilon LearningRate
--fraction MinimumInitialEnergyFraction
--steps NumberOfTimesteps
--iterations MaximumNumberOfIterations
--cost CostFunction
0 = Mean Square Error
1 = Cross Correlation
2 = Mutual Information
--bins Number of bins used with MI
--radius The radius used with CC
--MASGD use MAS-GD optimisation
--verbose ]
You can run the Example.py in your terminal to browse the registration performance.
chmod -R 775 ./Bin/
python Example.py
===================================================
*************** Do the preprocessing **************
===================================================
===================================================
*********** Do the Affine Registration ************
===================================================
Number of input arguments =
6
Parameters =
0 = /mnt/postData/LDDMM_experiments/version3/Bin/
1 = ./result/step0_inite/template.img
2 = ./result/step0_inite/target.img
3 = result/step1_affine/template_to_target/
4 = 1
5 = template_to_target_affined.img
*************************************************
*********************STEP1***********************
*************************************************
0) Template =
template img = ./result/step0_inite/template.img FILE EXISTS
template hdr = ./result/step0_inite/template.hdr FILE EXISTS
1) Target =
target img = ./result/step0_inite/target.img FILE EXISTS
target hdr = ./result/step0_inite/target.hdr FILE EXISTS
2) Output folder =
output path = result/step1_affine/template_to_target/ DIRECTORY EXISTS
3) Do histogram matching before AIR =
YES
4) Output =
output img = template_to_target_affined.img
*************************************************
*********************STEP2***********************
*************************************************
CREATING OUTPUT TEMPORARY DIRECTORY AND COPYING FILES
*************************************************
*********************STEP3***********************
*************************************************
GLOBAL HISTOGRAM MATCHING
INPUTS TO THE PROGRAM
0> /mnt/postData/LDDMM_experiments/version3/Bin//IMG_histmatch4
1> template_to_target.tmp/template.img
2> template_to_target.tmp/target.img
3> template_to_target.tmp/template_h.img
4> template_to_target.tmp/target_h.img
5> 1024
6> 3
7> 0
8> 1
9> template_to_target.tmp/template.hist
10> template_to_target.tmp/target.hist
11> template_to_target.tmp/template_h.hist
12> template_to_target.tmp/target_h.hist
inputfile1 template_to_target.tmp/template.img
inputfile2 template_to_target.tmp/target.img
outputfile1 template_to_target.tmp/template_h.img
outputfile2 template_to_target.tmp/target_h.img
Nbins 1024
sigma 3
matchendpoints 0
matchhistogram 1
inputfile1_hist template_to_target.tmp/template.hist
inputfile2_hist template_to_target.tmp/target.hist
outputfile1_hist template_to_target.tmp/template_h.hist
outputfile2_hist template_to_target.tmp/target_h.hist
*****READING IMAGES AND ALLOCATING ARRAYS*****
Image-1 max-min values: 255 0
Image-2 max-min values: 255 0
Error between histograms= 0.000117029
Image-1 max-min values: 255 0
Image-2 max-min values: 255 0
Error between histograms= 0.000117029
*****MATCHING HISTOGRAMS*****
Image-1 max-min values: 255 0
Image-2 max-min values: 255 0
Error between histograms= 7.49732e-06
*****IMAGES ARE BEING SAVED*****
Inside AnalyzeImage::saveAnalyzeImageData()
File template_to_target.tmp/template_h.img opened successfully
size_of_data_in_bytes = 31464000
Inside AnalyzeImage::saveAnalyzeImageData()
File template_to_target.tmp/target_h.img opened successfully
size_of_data_in_bytes = 31464000
*****TOTAL RUNNING TIME*****
6.80552
*************************************************
*********************STEP4***********************
*************************************************
AIR CALCULATIONS
INPUTS TO THE PROGRAM
0> /mnt/postData/LDDMM_experiments/version3/Bin//IMG_apply_AIR_tform
1> template_to_target.tmp/template.img
2> template_to_target.tmp/template_d.img
3> template_to_target.tmp/final_air.txt
4> 1
5> template_to_target.tmp/template.imgsize
6> 1
190 230 180
190 230 180
0.989353 -0.000350868 0.0194772 -0.450871
0.0158836 1.03171 -0.0821415 -0.230368
0.0249017 0.0896851 0.928384 -0.754377
0 0 0 1
1.01126 0.00217147 -0.0210237 0.440586
-0.0175929 0.961826 0.0854695 0.278118
-0.025425 -0.092974 1.06945 0.773885
0 0 0 1
190 230 180 1 1 1
Inside AnalyzeImage::saveAnalyzeImageData()
File template_to_target.tmp/template_d.img opened successfully
size_of_data_in_bytes = 31464000
*************************************************
*********************STEP5***********************
*************************************************
CLEANING TEMPORARY FILES
FINISHED...
===================================================
*********** Do the Histogram Matching *************
===================================================
result/step1_affine/template_to_target/template_to_target_affined.img
./result/step0_inite/target.img
INPUTS TO THE PROGRAM
0> /mnt/postData/LDDMM_experiments/version3/Bin/IMG_histmatch4
1> result/step1_affine/template_to_target/template_to_target_affined.img
2> ./result/step0_inite/target.img
3> result/step2_hist//template_to_target_affined_to_target/template_to_target_affined_to_target_histMatched.img
4> result/step2_hist//template_to_target_affined_to_target/target_histMatched.img
5> 1024
6> 3
7> 0
8> 1
9> result/step2_hist//template_to_target_affined_to_target/template_to_target_affined_to_target_hist1
10> result/step2_hist//template_to_target_affined_to_target/target_hist1
11> result/step2_hist//template_to_target_affined_to_target/template_to_target_affined_to_target_hist2
12> result/step2_hist//template_to_target_affined_to_target/target_hist2
inputfile1 result/step1_affine/template_to_target/template_to_target_affined.img
inputfile2 ./result/step0_inite/target.img
outputfile1 result/step2_hist//template_to_target_affined_to_target/template_to_target_affined_to_target_histMatched.img
outputfile2 result/step2_hist//template_to_target_affined_to_target/target_histMatched.img
Nbins 1024
sigma 3
matchendpoints 0
matchhistogram 1
inputfile1_hist result/step2_hist//template_to_target_affined_to_target/template_to_target_affined_to_target_hist1
inputfile2_hist result/step2_hist//template_to_target_affined_to_target/target_hist1
outputfile1_hist result/step2_hist//template_to_target_affined_to_target/template_to_target_affined_to_target_hist2
outputfile2_hist result/step2_hist//template_to_target_affined_to_target/target_hist2
*****READING IMAGES AND ALLOCATING ARRAYS*****
Image-1 max-min values: 254.751 0
Image-2 max-min values: 255 0
Error between histograms= 0.00546247
Image-1 max-min values: 254.751 0
Image-2 max-min values: 255 0
Error between histograms= 0.00546247
*****MATCHING HISTOGRAMS*****
Image-1 max-min values: 255 0
Image-2 max-min values: 255 0
Error between histograms= 7.37313e-06
*****IMAGES ARE BEING SAVED*****
Inside AnalyzeImage::saveAnalyzeImageData()
File result/step2_hist//template_to_target_affined_to_target/template_to_target_affined_to_target_histMatched.img opened successfully
size_of_data_in_bytes = 31464000
Inside AnalyzeImage::saveAnalyzeImageData()
File result/step2_hist//template_to_target_affined_to_target/target_histMatched.img opened successfully
size_of_data_in_bytes = 31464000
*****TOTAL RUNNING TIME*****
6.64489
===================================================
*********** Do the lddmmMASGD Registration ********
===================================================
1> Template image: result/step2_hist//template_to_target_affined_to_target/template_to_target_affined_to_target_histMatched.img
2> Target img: result/step2_hist//template_to_target_affined_to_target/target_histMatched.img
3> Metric: Cross Correlation
Step 0: alpha=0.01, scale=0.25, iteration=1000
E, E_velocity, E_image (E_image %), Step_size
0. -3.52181e+12, 0, -3.52181e+12(100%), 5.000000e-02 BLS-GD step
1. -3.56526e+12, 3.55675e+06, -3.56526e+12(92.4586%), 5.500000e-02 BLS-GD step
2. -3.58724e+12, 8.32238e+06, -3.58725e+12(88.6433%), 6.050000e-02 BLS-GD step
3. -3.60612e+12, 1.63938e+07, -3.60614e+12(85.3651%), 6.655000e-02 BLS-GD step
4. -3.62055e+12, 2.51855e+07, -3.62057e+12(82.8603%), 7.320500e-02 BLS-GD step
5. -3.63314e+12, 3.68644e+07, -3.63318e+12(80.6733%), 8.052550e-02 BLS-GD step
6. -3.63997e+12, 4.65256e+07, -3.64002e+12(79.4857%), 8.857805e-02 BLS-GD step
7. -3.64001e+12, 4.65368e+07, -3.64006e+12(79.4794%), 2.435896e-02 BLS-GD step
8. -3.64669e+12, 5.03424e+07, -3.64674e+12(78.3189%), 2.679486e-02 BLS-GD step
9. -3.64971e+12, 5.40116e+07, -3.64976e+12(77.7952%), 2.947435e-02 BLS-GD step
10. -3.6553e+12, 6.56513e+07, -3.65537e+12 (76.822%), 6.092153e-02. MAS-GD step
11. -3.66663e+12, 8.53932e+07, -3.66672e+12 (74.8522%), 7.020104e-02. MAS-GD step
12. -3.68317e+12, 1.25686e+08, -3.68329e+12 (71.9762%), 8.674193e-02. MAS-GD step
13. -3.69852e+12, 1.81172e+08, -3.6987e+12 (69.3021%), 8.739609e-02. MAS-GD step
14. -3.70814e+12, 2.22406e+08, -3.70836e+12 (67.6261%), 5.851865e-02. MAS-GD step
15. -3.71608e+12, 2.60708e+08, -3.71634e+12 (66.2411%), 5.629652e-02. MAS-GD step
16. -3.72219e+12, 2.95709e+08, -3.72249e+12 (65.1743%), 5.798493e-02. MAS-GD step
17. -3.72602e+12, 3.25836e+08, -3.72635e+12 (64.5042%), 6.103032e-02. MAS-GD step
18. -3.72727e+12, 3.52477e+08, -3.72763e+12 (64.2825%), 7.189995e-02. MAS-GD step
19. -3.72624e+12, 3.66736e+08, -3.7266e+12 (64.4602%), 6.175223e-02. MAS-GD step
20. -3.72727e+12, 3.52477e+08, -3.72763e+12(64.2825%), 6.175223e-02 BLS-GD step
21. -3.72698e+12, 3.38724e+08, -3.72732e+12 (64.3363%), 6.015211e-02. MAS-GD step
22. -3.72752e+12, 3.52563e+08, -3.72788e+12(64.2392%), 6.015211e-02 BLS-GD step
E = -3.72755e+12 (64.2337%)
Length = 8355.71
Time = 4.04984s (0.0674973m)
Step 1: alpha=0.005, scale=0.5, iteration=1000
E, E_velocity, E_image (E_image %), Step_size
0. -3.61264e+12, 0, -3.61264e+12(100%), 5.000000e-02 BLS-GD step
1. -3.63091e+12, 1.57022e+06, -3.63092e+12(95.855%), 5.500000e-02 BLS-GD step
2. -3.63799e+12, 4.18534e+06, -3.63799e+12(94.2498%), 6.050000e-02 BLS-GD step
3. -3.638e+12, 4.1863e+06, -3.63801e+12(94.2469%), 1.663750e-02 BLS-GD step
4. -3.64263e+12, 5.06065e+06, -3.64263e+12(93.1975%), 1.830125e-02 BLS-GD step
5. -3.64494e+12, 6.16682e+06, -3.64495e+12(92.672%), 2.013138e-02 BLS-GD step
6. -3.64732e+12, 7.46903e+06, -3.64732e+12(92.1331%), 2.214451e-02 BLS-GD step
7. -3.64972e+12, 8.97355e+06, -3.64973e+12(91.5873%), 2.435896e-02 BLS-GD step
8. -3.65211e+12, 1.07215e+07, -3.65212e+12(91.0457%), 2.679486e-02 BLS-GD step
9. -3.65458e+12, 1.27292e+07, -3.65459e+12(90.4843%), 2.947435e-02 BLS-GD step
10. -3.65879e+12, 1.67405e+07, -3.6588e+12 (89.5293%), 2.146586e-02. MAS-GD step
11. -3.66286e+12, 2.14945e+07, -3.66288e+12 (88.6046%), 2.700389e-02. MAS-GD step
12. -3.66723e+12, 2.74272e+07, -3.66726e+12 (87.6111%), 2.021786e-02. MAS-GD step
13. -3.67669e+12, 4.44222e+07, -3.67674e+12 (85.4611%), 3.694235e-02. MAS-GD step
14. -3.68039e+12, 5.22512e+07, -3.68044e+12 (84.6214%), 1.253263e-02. MAS-GD step
15. -3.6918e+12, 8.90815e+07, -3.69189e+12 (82.0241%), 4.345689e-02. MAS-GD step
16. -3.69354e+12, 9.27946e+07, -3.69364e+12 (81.6283%), 3.724509e-03. MAS-GD step
17. -3.696e+12, 9.77749e+07, -3.69609e+12 (81.0705%), 4.776377e-03. MAS-GD step
18. -3.69262e+12, 1.2211e+08, -3.69274e+12 (81.8314%), 2.060038e-02. MAS-GD step
19. -3.696e+12, 9.77749e+07, -3.69609e+12(81.0705%), 2.060038e-02 BLS-GD step
20. -3.6964e+12, 9.80124e+07, -3.6965e+12 (80.9786%), 1.963629e-02. MAS-GD step
21. -3.69726e+12, 9.89277e+07, -3.69736e+12 (80.7846%), 2.866156e-02. MAS-GD step
22. -3.69796e+12, 9.99866e+07, -3.69806e+12 (80.6258%), 1.614905e-02. MAS-GD step
23. -3.69951e+12, 1.07943e+08, -3.69961e+12 (80.2725%), 6.242965e-02. MAS-GD step
24. -3.70033e+12, 1.08769e+08, -3.70044e+12 (80.085%), 4.173670e-03. MAS-GD step
25. -3.70145e+12, 1.10355e+08, -3.70156e+12 (79.8298%), 6.171378e-03. MAS-GD step
26. -3.69393e+12, 1.19209e+08, -3.69405e+12 (81.5346%), 2.525719e-02. MAS-GD step
27. -3.70145e+12, 1.10355e+08, -3.70156e+12(79.8298%), 2.525719e-02 BLS-GD step
28. -3.70182e+12, 1.11596e+08, -3.70194e+12 (79.7457%), 2.426502e-02. MAS-GD step
29. -3.7027e+12, 1.14355e+08, -3.70282e+12 (79.5457%), 2.690820e-02. MAS-GD step
30. -3.70412e+12, 1.1985e+08, -3.70424e+12 (79.2231%), 3.611455e-02. MAS-GD step
31. -3.70489e+12, 1.22738e+08, -3.70501e+12 (79.0488%), 1.541742e-02. MAS-GD step
32. -3.70646e+12, 1.33737e+08, -3.70659e+12 (78.69%), 4.935321e-02. MAS-GD step
33. -3.70721e+12, 1.3478e+08, -3.70735e+12 (78.5178%), 4.499963e-03. MAS-GD step
34. -3.70819e+12, 1.36381e+08, -3.70833e+12 (78.2958%), 6.725049e-03. MAS-GD step
35. -3.70378e+12, 1.42474e+08, -3.70392e+12 (79.295%), 2.307049e-02. MAS-GD step
36. -3.70819e+12, 1.36381e+08, -3.70833e+12(78.2958%), 2.307049e-02 BLS-GD step
37. -3.70828e+12, 1.36383e+08, -3.70841e+12 (78.2766%), 2.305597e-02. MAS-GD step
38. -3.70852e+12, 1.36516e+08, -3.70866e+12 (78.2201%), 2.165943e-02. MAS-GD step
39. -3.70882e+12, 1.37061e+08, -3.70895e+12 (78.1535%), 2.626242e-02. MAS-GD step
40. -3.70903e+12, 1.37797e+08, -3.70917e+12 (78.1047%), 1.725130e-02. MAS-GD step
41. -3.70915e+12, 1.40013e+08, -3.70929e+12 (78.0785%), 3.140271e-02. MAS-GD step
42. -3.70951e+12, 1.40918e+08, -3.70966e+12 (77.9946%), 8.994750e-03. MAS-GD step
43. -3.70993e+12, 1.43081e+08, -3.71008e+12 (77.8989%), 1.650806e-02. MAS-GD step
44. -3.71026e+12, 1.44409e+08, -3.7104e+12 (77.8251%), 8.459081e-03. MAS-GD step
45. -3.71084e+12, 1.45927e+08, -3.71098e+12 (77.6936%), 8.423223e-03. MAS-GD step
46. -3.70945e+12, 1.52563e+08, -3.70961e+12 (78.0058%), 3.067871e-02. MAS-GD step
47. -3.71084e+12, 1.45927e+08, -3.71098e+12(77.6936%), 3.067871e-02 BLS-GD step
48. -3.71099e+12, 1.46519e+08, -3.71114e+12 (77.6582%), 2.926567e-02. MAS-GD step
49. -3.71131e+12, 1.4729e+08, -3.71146e+12 (77.5863%), 1.951409e-02. MAS-GD step
50. -3.71162e+12, 1.51105e+08, -3.71177e+12 (77.5141%), 6.438126e-02. MAS-GD step
51. -3.71202e+12, 1.5141e+08, -3.71217e+12 (77.4249%), 5.314419e-03. MAS-GD step
52. -3.712e+12, 1.51954e+08, -3.71215e+12 (77.4284%), 9.105189e-03. MAS-GD step
53. -3.71202e+12, 1.5141e+08, -3.71217e+12(77.4249%), 9.105189e-03 BLS-GD step
54. -3.7122e+12, 1.5139e+08, -3.71235e+12 (77.3837%), 8.671860e-03. MAS-GD step
55. -3.70805e+12, 1.51558e+08, -3.70821e+12 (78.3235%), 3.938622e-02. MAS-GD step
56. -3.7122e+12, 1.5139e+08, -3.71235e+12(77.3837%), 3.938622e-02 BLS-GD step
57. -3.71215e+12, 1.51411e+08, -3.7123e+12 (77.3946%), 3.765703e-02. MAS-GD step
58. -3.7122e+12, 1.5139e+08, -3.71235e+12(77.3836%), 3.765703e-02 BLS-GD step
E = -3.7122e+12 (77.3836%)
Length = 4241.67
Time = 28.4629s (0.474381m)
Step 2: alpha=0.002, scale=1.0, iteration=1000
E, E_velocity, E_image (E_image %), Step_size
0. -3.57624e+12, 0, -3.57624e+12(100%), 5.000000e-02 BLS-GD step
1. -3.59205e+12, 2.42832e+06, -3.59206e+12(96.204%), 5.500000e-02 BLS-GD step
2. -3.59208e+12, 2.42841e+06, -3.59208e+12(96.1984%), 1.512500e-02 BLS-GD step
3. -3.6001e+12, 3.10456e+06, -3.6001e+12(94.2725%), 1.663750e-02 BLS-GD step
4. -3.60316e+12, 4.45251e+06, -3.60316e+12(93.5388%), 1.830125e-02 BLS-GD step
5. -3.60469e+12, 6.06745e+06, -3.60469e+12(93.171%), 2.013138e-02 BLS-GD step
6. -3.60469e+12, 6.0675e+06, -3.6047e+12(93.1707%), 5.536128e-03 BLS-GD step
7. -3.61125e+12, 6.43749e+06, -3.61126e+12(91.5955%), 6.089741e-03 BLS-GD step
8. -3.61309e+12, 7.02214e+06, -3.61309e+12(91.1555%), 6.698715e-03 BLS-GD step
9. -3.61455e+12, 7.69289e+06, -3.61456e+12(90.803%), 7.368587e-03 BLS-GD step
10. -3.61812e+12, 9.48886e+06, -3.61813e+12 (89.9479%), 9.673639e-03. MAS-GD step
11. -3.62232e+12, 1.1938e+07, -3.62233e+12 (88.9394%), 1.048350e-02. MAS-GD step
12. -3.62863e+12, 1.64087e+07, -3.62864e+12 (87.4237%), 1.122152e-02. MAS-GD step
13. -3.6371e+12, 2.41766e+07, -3.63712e+12 (85.389%), 1.250904e-02. MAS-GD step
14. -3.64772e+12, 3.73116e+07, -3.64776e+12 (82.8362%), 1.429565e-02. MAS-GD step
15. -3.65853e+12, 5.57165e+07, -3.65858e+12 (80.2388%), 1.439002e-02. MAS-GD step
16. -3.66778e+12, 7.64492e+07, -3.66786e+12 (78.0135%), 1.260684e-02. MAS-GD step
17. -3.67598e+12, 9.90668e+07, -3.67608e+12 (76.0409%), 1.145545e-02. MAS-GD step
18. -3.6834e+12, 1.23002e+08, -3.68352e+12 (74.2543%), 1.060909e-02. MAS-GD step
19. -3.69021e+12, 1.47768e+08, -3.69036e+12 (72.6146%), 9.975802e-03. MAS-GD step
20. -3.69664e+12, 1.73386e+08, -3.69682e+12 (71.0639%), 9.651500e-03. MAS-GD step
21. -3.70281e+12, 2.00068e+08, -3.70301e+12 (69.5775%), 9.610634e-03. MAS-GD step
22. -3.70869e+12, 2.27366e+08, -3.70892e+12 (68.1606%), 9.574747e-03. MAS-GD step
23. -3.71416e+12, 2.54552e+08, -3.71441e+12 (66.8418%), 9.450408e-03. MAS-GD step
24. -3.71918e+12, 2.81021e+08, -3.71946e+12 (65.6306%), 9.279284e-03. MAS-GD step
25. -3.72378e+12, 3.06269e+08, -3.72408e+12 (64.5212%), 9.081024e-03. MAS-GD step
26. -3.72796e+12, 3.29839e+08, -3.72829e+12 (63.5116%), 8.847520e-03. MAS-GD step
27. -3.73178e+12, 3.51815e+08, -3.73213e+12 (62.5902%), 8.753602e-03. MAS-GD step
28. -3.73535e+12, 3.72429e+08, -3.73572e+12 (61.7282%), 8.848743e-03. MAS-GD step
29. -3.73855e+12, 3.91315e+08, -3.73894e+12 (60.955%), 8.874137e-03. MAS-GD step
30. -3.74128e+12, 4.07881e+08, -3.74169e+12 (60.2955%), 8.685333e-03. MAS-GD step
31. -3.74354e+12, 4.22241e+08, -3.74396e+12 (59.7499%), 8.596648e-03. MAS-GD step
32. -3.74536e+12, 4.34699e+08, -3.7458e+12 (59.3103%), 8.735825e-03. MAS-GD step
33. -3.74667e+12, 4.45035e+08, -3.74712e+12 (58.9937%), 8.762552e-03. MAS-GD step
34. -3.74748e+12, 4.52931e+08, -3.74793e+12 (58.7975%), 8.501041e-03. MAS-GD step
35. -3.74785e+12, 4.58587e+08, -3.74831e+12 (58.7071%), 8.373738e-03. MAS-GD step
36. -3.74783e+12, 4.62221e+08, -3.74829e+12 (58.7108%), 8.499865e-03. MAS-GD step
37. -3.74785e+12, 4.58587e+08, -3.74831e+12(58.7071%), 8.499865e-03 BLS-GD step
38. -3.74873e+12, 4.5474e+08, -3.74918e+12 (58.4983%), 9.024573e-03. MAS-GD step
39. -3.74945e+12, 4.45956e+08, -3.7499e+12 (58.3257%), 1.087113e-02. MAS-GD step
40. -3.74994e+12, 4.3076e+08, -3.75037e+12 (58.2124%), 1.333661e-02. MAS-GD step
41. -3.74923e+12, 4.09351e+08, -3.74964e+12 (58.3876%), 1.548482e-02. MAS-GD step
42. -3.74994e+12, 4.3076e+08, -3.75037e+12(58.2124%), 1.548482e-02 BLS-GD step
43. -3.74987e+12, 4.25861e+08, -3.7503e+12 (58.23%), 1.540369e-02. MAS-GD step
44. -3.74994e+12, 4.30973e+08, -3.75037e+12(58.2122%), 1.540369e-02 BLS-GD step
E = -3.74994e+12 (58.2121%)
Length = 5611.48
Time = 139.577s (2.32628m)
===================================================
*********** apply field to image and saving *******
===================================================
1> image: ./data/template.nii.gz
2> save file: result/step3_lddmm/lddmmcc/LDDMM_Images//template_to_target.nii.gz
3> field: ./result/lddmmcc/field/template_to_target_combine.vtk