/
getDataInTime.ts
216 lines (180 loc) · 6.6 KB
/
getDataInTime.ts
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
import { utilities, cache, Types } from '@cornerstonejs/core';
import { getVoxelOverlap } from '../segmentation/utilities';
import pointInShapeCallback from '../pointInShapeCallback';
/**
* Gets the scalar data for a series of time points for either a single
* coordinate or a segmentation mask, it will return the an array of scalar
* data for a single coordinate or an array of arrays for a segmentation.
*
* @param dynamicVolume - 4D volume to compute time point data from
* @param options - frameNumbers: which frames to use as timepoints, if left
* blank, gets data timepoints over all frames
* maskVolumeId: segmentationId to get timepoint data of
* imageCoordinate: world coordinate to get timepoint data of
* @returns
*/
function getDataInTime(
dynamicVolume: Types.IDynamicImageVolume,
options: {
frameNumbers?;
maskVolumeId?;
imageCoordinate?;
}
): number[] | number[][] {
let dataInTime;
// if frameNumbers is not provided, all frames are selected
const frames = options.frameNumbers || [
...Array(dynamicVolume.numTimePoints).keys(),
];
// You only need to provide either maskVolumeId OR imageCoordinate.
// Throws error if neither maskVolumeId or imageCoordinate is given,
// throws error if BOTH maskVolumeId and imageCoordinate is given
if (!options.maskVolumeId && !options.imageCoordinate) {
throw new Error(
'You should provide either maskVolumeId or imageCoordinate'
);
}
if (options.maskVolumeId && options.imageCoordinate) {
throw new Error('You can only use one of maskVolumeId or imageCoordinate');
}
if (options.maskVolumeId) {
const segmentationVolume = cache.getVolume(options.maskVolumeId);
const [dataInTime, ijkCoords] = _getTimePointDataMask(
frames,
dynamicVolume,
segmentationVolume
);
return [dataInTime, ijkCoords];
}
if (options.imageCoordinate) {
const dataInTime = _getTimePointDataCoordinate(
frames,
options.imageCoordinate,
dynamicVolume
);
return dataInTime;
}
return dataInTime;
}
function _getTimePointDataCoordinate(frames, coordinate, volume) {
const { dimensions, imageData } = volume;
const index = imageData.worldToIndex(coordinate);
index[0] = Math.floor(index[0]);
index[1] = Math.floor(index[1]);
index[2] = Math.floor(index[2]);
if (!utilities.indexWithinDimensions(index, dimensions)) {
throw new Error('outside bounds');
}
// calculate offset for index
const yMultiple = dimensions[0];
const zMultiple = dimensions[0] * dimensions[1];
const allScalarData = volume.getScalarDataArrays();
const value = [];
frames.forEach((frame) => {
const activeScalarData = allScalarData[frame];
const scalarIndex = index[2] * zMultiple + index[1] * yMultiple + index[0];
value.push(activeScalarData[scalarIndex]);
});
return value;
}
function _getTimePointDataMask(frames, dynamicVolume, segmentationVolume) {
const { imageData: maskImageData } = segmentationVolume;
const segScalarData = segmentationVolume.getScalarData();
const len = segScalarData.length;
// Pre-allocate memory for array
const nonZeroVoxelIndices = [];
nonZeroVoxelIndices.length = len;
const ijkCoords = [];
const dimensions = segmentationVolume.dimensions;
// Get the index of every non-zero voxel in mask
let actualLen = 0;
for (let i = 0, len = segScalarData.length; i < len; i++) {
if (segScalarData[i] !== 0) {
ijkCoords.push([
i % dimensions[0],
Math.floor((i / dimensions[0]) % dimensions[1]),
Math.floor(i / (dimensions[0] * dimensions[1])),
]);
nonZeroVoxelIndices[actualLen++] = i;
}
}
// Trim the array to actual size
nonZeroVoxelIndices.length = actualLen;
const dynamicVolumeScalarDataArray = dynamicVolume.getScalarDataArrays();
const values = [];
const isSameVolume =
dynamicVolumeScalarDataArray[0].length === len &&
JSON.stringify(dynamicVolume.spacing) ===
JSON.stringify(segmentationVolume.spacing);
// if the segmentation mask is the same size as the dynamic volume (one frame)
// means we can just return the scalar data for the non-zero voxels
if (isSameVolume) {
for (let i = 0; i < nonZeroVoxelIndices.length; i++) {
const indexValues = [];
frames.forEach((frame) => {
const activeScalarData = dynamicVolumeScalarDataArray[frame];
indexValues.push(activeScalarData[nonZeroVoxelIndices[i]]);
});
values.push(indexValues);
}
return [values, ijkCoords];
}
// In case that the segmentation mask is not the same size as the dynamic volume (one frame)
// then we need to consider each voxel in the segmentation mask and check if it
// overlaps with the other volume, and if so we need to average the values of the
// overlapping voxels.
const callback = ({
pointLPS: segPointLPS,
value: segValue,
pointIJK: segPointIJK,
}) => {
// see if the value is non-zero
if (segValue === 0) {
// not interested
return;
}
// Then for each non-zero voxel in the segmentation mask, we should
// again perform the pointInShapeCallback to run the averaging callback
// function to get the average value of the overlapping voxels.
const overlapIJKMinMax = getVoxelOverlap(
dynamicVolume.imageData,
dynamicVolume.dimensions,
dynamicVolume.spacing,
segPointLPS
);
// count represents the number of voxels of the dynamic volume that represents
// one voxel of the segmentation mask
let count = 0;
const perFrameSum = new Map();
// Pre-initialize the Map
frames.forEach((frame) => perFrameSum.set(frame, 0));
const averageCallback = ({ index }) => {
for (let i = 0; i < frames.length; i++) {
const value = dynamicVolumeScalarDataArray[i][index];
const frame = frames[i];
perFrameSum.set(frame, perFrameSum.get(frame) + value);
}
count++;
};
pointInShapeCallback(
dynamicVolume.imageData,
() => true,
averageCallback,
overlapIJKMinMax
);
// average the values
const averageValues = [];
perFrameSum.forEach((sum) => {
averageValues.push(sum / count);
});
ijkCoords.push(segPointIJK);
values.push(averageValues);
};
// Since we have the non-zero voxel indices of the segmentation mask,
// we theoretically can use them, however, we kind of need to compute the
// pointLPS for each of the non-zero voxel indices, which is a bit of a pain.
// Todo: consider using the nonZeroVoxelIndices to compute the pointLPS
pointInShapeCallback(maskImageData, () => true, callback);
return [values, ijkCoords];
}
export default getDataInTime;