Table 2 Quantitative analysis of segmentation results obtained from volume shown in Figs. 3 and 4. Segmentation results obtained from DeepSynth were quantitatively compared with those obtained from FARSIGHT, Squassh and CellProfiler using either default settings or settings optimized as described in “Methods”.

From: DeepSynth: Three-dimensional nuclear segmentation of biological images using neural networks trained with synthetic data

 

Time (entire volume)

Voxel based

Object based

Type I

Type II

Accuracy

Precision

Recall

F1

DeepSynth

102 sec

2.14%

1.88%

95.98%

89.38%

95.26%

92.23%

FARSIGHT Default

13 min

30.33%

0.00%

69.67%

64.34%

97.53%

77.53%

FARSIGHT Optimized

13 min

28.58%

0.53%

70.90%

37.72%

97.73%

54.43%

Squassh Default

Hours

24.75%

0.00%

75.25%

90.38%

16.61%

28.06%

Squassh Optimized

Hours

19.54%

0.01%

80.45%

85.07%

20.14%

32.57%

CellProfiler Default

15 min

21.67%

0.31%

78.02%

80.82%

20.92%

33.24%

CellProfiler Optimized

15 min

14.94%

0.11%

84.95%

81.93%

72.08%

76.69%

Otsu- 3DWatershed

32 sec

17.88%

0.23%

81.89%

87.28%

53.36%

66.23%

  1. The values for “Time” reflect the times required to obtain segmentations using. Accuracy was measured using both voxel-based metrics (voxel-by-voxel agreement with ground-truth data) and object-based metrics (agreement in the detection of objects with ground-truth data) in a 64 by 64 by 64 voxel sub-volume. For voxel-based accuracy, type-I error (false positive rate) represents the fraction of voxels in the volume wrongly detected as belonging to nuclei and type-II error (false negative rate) represents the fraction of voxels wrongly detected as background. Object-based accuracy is measured using the F1 score, which is the harmonic mean of precision and recall, where precision is the ratio of the number of correctly identified nuclei to the sum of the number of correctly identified nuclei plus the number of objects incorrectly identified as nuclei and recall is the ratio of the number of correctly identified nuclei to the sum of the number of correctly identified nuclei plus the number of nuclei that failed to be detected. Details of the analyses are described in “Methods”.
Back to article page