Closed bc closed 4 years ago
full_retina.zip source files:
got flatmoutn plot to represent the same orientation as in ImageJ.
good overlap with datapoints.csv and outline.roi. should be no data outwith the outline
flipped Y and the axes match perfectly. updated tear_markup_plot to match with some refactoring
now the sampling points work
outline and grid do not align
points on red are being excluded. somehow the units of those points need to be converted to match the coordinate frame of the outline.roi, as it is entered in retistruct
grid point align correctly with output dtt dataframe
raw coordinates from outline.ROI
roi | retistruct |
---|---|
x y
[1,] 866 480
[2,] 867 461
[3,] 830 324
[4,] 711 218
[5,] 620 229
[6,] 657 83
[7,] 562 29
[8,] 484 7
[9,] 402 0
[10,] 233 27
[11,] 150 69
[12,] 172 201
[13,] 221 281
[14,] 287 375
[15,] 138 408
[16,] 28 438
[17,] 19 529
[18,] 20 601
[19,] 41 663
[20,] 91 734
[21,] 140 787
[22,] 179 838
[23,] 249 925
[24,] 353 851
[25,] 406 841
[26,] 419 916
[27,] 426 990
[28,] 560 940
[29,] 667 875
[30,] 767 772
[31,] 831 701
[32,] 793 665
[33,] 737 628
[34,] 697 574
[35,] 865 495
``` |
r$> load_roi(roi_path)$coords x y [1,] 866 516 [2,] 867 535 [3,] 830 672 [4,] 711 778 [5,] 620 767 [6,] 657 913 [7,] 562 967 [8,] 484 989 [9,] 402 996 [10,] 233 969 [11,] 150 927 [12,] 172 795 [13,] 221 715 [14,] 287 621 [15,] 138 588 [16,] 28 558 [17,] 19 467 [18,] 20 395 [19,] 41 333 [20,] 91 262 [21,] 140 209 [22,] 179 158 [23,] 249 71 [24,] 353 145 [25,] 406 155 [26,] 419 80 [27,] 426 6 [28,] 560 56 [29,] 667 121 [30,] 767 224 [31,] 831 295 [32,] 793 331 [33,] 737 368 [34,] 697 422 [35,] 865 501
when using the retistruct-style outline approach, no items land outside the outline!
x y phi lambda phi_deg lambda_deg
1: 866 480 0.0000000 1.6927944 0.00000 96.98998
2: 867 461 0.0000000 1.6286314 0.00000 93.31371
3: 830 324 0.0000000 1.1685015 0.00000 66.95021
4: 711 218 0.0000000 0.6643378 0.00000 38.06375
5: 620 229 -0.4002356 0.6113949 -22.93181 35.03035
6: 657 83 0.0000000 0.6643378 0.00000 38.06375
7: 562 29 0.0000000 0.2725197 0.00000 15.61423
8: 484 7 0.0000000 0.0000000 0.00000 0.00000
9: 402 0 0.0000000 -0.2740023 0.00000 -15.69918
10: 233 27 0.0000000 -0.8313582 0.00000 -47.63331
11: 150 69 0.0000000 -1.1627718 0.00000 -66.62192
12: 172 201 -0.4266316 -1.1430187 -24.44419 -65.49015
13: 221 281 -0.6982187 -1.0705789 -40.00499 -61.33965
14: 287 375 -1.0169022 -0.9527973 -58.26421 -54.59126
15: 138 408 -0.4623417 -1.1268234 -26.49023 -64.56222
16: 28 438 0.0000000 -1.1627718 0.00000 -66.62192
17: 19 529 0.0000000 -1.4589114 0.00000 -83.58947
18: 20 601 0.0000000 -1.6741254 0.00000 -95.92032
19: 41 663 0.0000000 -1.8640987 0.00000 -106.80499
20: 91 734 0.0000000 -2.1198834 0.00000 -121.46037
21: 140 787 0.0000000 -2.3413462 0.00000 -134.14926
22: 179 838 0.0000000 -2.5415106 0.00000 -145.61783
23: 249 925 0.0000000 -2.8942786 0.00000 -165.82995
24: 353 851 -0.3497687 -2.9756204 -20.04027 -170.49049
25: 406 841 -0.4705284 -3.0322503 -26.95929 -173.73515
26: 419 916 -0.2477825 -2.9523932 -14.19689 -169.15967
27: 426 990 0.0000000 -2.8942786 0.00000 -165.82995
28: 560 940 0.0000000 2.9224854 0.00000 167.44608
29: 667 875 0.0000000 2.5295023 0.00000 144.92981
30: 767 772 0.0000000 2.0670298 0.00000 118.43209
31: 831 701 0.0000000 1.7436827 0.00000 99.90566
32: 793 665 -0.1915390 1.7586750 -10.97437 100.76465
33: 737 628 -0.4175587 1.7924179 -23.92435 102.69798
34: 697 574 -0.6189636 1.7930998 -35.46400 102.73705
35: 865 495 0.0000000 1.7436827 0.00000 99.90566```
x y phi lambda phi_deg lambda_deg successful_mapping
1: 866 480 0.0000000 1.6927944 0.00000 96.98998 TRUE
2: 867 461 0.0000000 1.6286314 0.00000 93.31371 TRUE
3: 830 324 0.0000000 1.1685015 0.00000 66.95021 TRUE
4: 711 218 0.0000000 0.6643378 0.00000 38.06375 TRUE
5: 620 229 -0.4002356 0.6113949 -22.93181 35.03035 TRUE
6: 657 83 0.0000000 0.6643378 0.00000 38.06375 TRUE
7: 562 29 0.0000000 0.2725197 0.00000 15.61423 TRUE
8: 484 7 0.0000000 0.0000000 0.00000 0.00000 TRUE
9: 402 0 0.0000000 -0.2740023 0.00000 -15.69918 TRUE
10: 233 27 0.0000000 -0.8313582 0.00000 -47.63331 TRUE
11: 150 69 0.0000000 -1.1627718 0.00000 -66.62192 TRUE
12: 172 201 -0.4266316 -1.1430187 -24.44419 -65.49015 TRUE
13: 221 281 -0.6982187 -1.0705789 -40.00499 -61.33965 TRUE
14: 287 375 -1.0169022 -0.9527973 -58.26421 -54.59126 TRUE
15: 138 408 -0.4623417 -1.1268234 -26.49023 -64.56222 TRUE
16: 28 438 0.0000000 -1.1627718 0.00000 -66.62192 TRUE
17: 19 529 0.0000000 -1.4589114 0.00000 -83.58947 TRUE
18: 20 601 0.0000000 -1.6741254 0.00000 -95.92032 TRUE
19: 41 663 0.0000000 -1.8640987 0.00000 -106.80499 TRUE
20: 91 734 0.0000000 -2.1198834 0.00000 -121.46037 TRUE
21: 140 787 0.0000000 -2.3413462 0.00000 -134.14926 TRUE
22: 179 838 0.0000000 -2.5415106 0.00000 -145.61783 TRUE
23: 249 925 0.0000000 -2.8942786 0.00000 -165.82995 TRUE
24: 353 851 -0.3497687 -2.9756204 -20.04027 -170.49049 TRUE
25: 406 841 -0.4705284 -3.0322503 -26.95929 -173.73515 TRUE
26: 419 916 -0.2477825 -2.9523932 -14.19689 -169.15967 TRUE
27: 426 990 0.0000000 -2.8942786 0.00000 -165.82995 TRUE
28: 560 940 0.0000000 2.9224854 0.00000 167.44608 TRUE
29: 667 875 0.0000000 2.5295023 0.00000 144.92981 TRUE
30: 767 772 0.0000000 2.0670298 0.00000 118.43209 TRUE
31: 831 701 0.0000000 1.7436827 0.00000 99.90566 TRUE
32: 793 665 -0.1915390 1.7586750 -10.97437 100.76465 TRUE
33: 737 628 -0.4175587 1.7924179 -23.92435 102.69798 TRUE
34: 697 574 -0.6189636 1.7930998 -35.46400 102.73705 TRUE
35: 865 495 0.0000000 1.7436827 0.00000 99.90566 TRUE```
with successful mapping plotting true for all outline points
a perfect glorious match between expected and observed outline location and success parameters:
misalignment of measured datapoints.csv
, appears as though it's only a vertical bias offset
target
issue identified: both the points and the outline, when run through retistruct's imageJ transformation have 0 as the minimum point. But we know from the raw data that the minY val for the outline was 6:
and the min Y val for the datapoints was 24
aligned ✅
aligns well after refactoring
added outline path
The azimuthal plot is aligned correctly
completed flatplot, nonsmoothed/predicted
looks like a 180 flip after projection
my_retina_figure.pdf reimplemented falciform
new tear markup plotter now shows where the datapoints are:
plus if you zoom in on the PDF, it will show the data measurement indices (row indices from the measurements)
deformation plots
added notes on flatmount for how the retina was reconstructed