Figure 1.
User labels and algorithm predictions.
Top row: the complete set of user annotations for the first training set (20 brush strokes in total), with yellow labels for synapses, red for membranes, green for the rest. Bottom row: raw data and algorithm predictions on two other slices in the first training set. In black circles: some unlabeled synapses and their probability maps. The color intensity corresponds to the certainty in the prediction, predictions for green class are omitted for clarity.
Figure 2.
Precision and recall of the algorithm and the human experts.
Recall was calculated as the (no. of true positives)/(no. of synapses in the ground truth), precision as the (no. of true positives)/(total no. of synapse candidates). A: Precision and recall of the algorithm results for the four different training sets. B: Precision and recall of the algorithm compared to the human experts with and without the time limit. The synapse probability threshold values are annotated next to the corresponding points of the curve.
Figure 3.
3D visualization of the results.
Top: all synapses detected by the algorithm after training on the labels from Fig. 1. Bottom: a close-up view of three differently oriented synapses.
Figure 4.
Synapse detection summary report.
Part of the summary report produced by ilastik. The fourth detection from the top (no. 36) is a false positive, which can easily be filtered out by a human expert by looking at a larger context.
Figure 5.
A, B, C: false negative decisions of the human observers, D, E, F: false positive detections of the human observers, shown as yellow “ball” labels in the image center, G, H, I: false negative decisions of the algorithm, J, K, L false positive decisions of the algorithm.
Table 1.
Voxel features.