Automated segmentation of insect anatomy from micro-CT images using deep learning
Data files
Oct 03, 2023 version files 13.94 GB
-
README.md
-
testing.zip
-
training.zip
Abstract
README: Automated segmentation of insect anatomy from micro-CT images using deep learning
Evropi Toulkeridou,
Carlos Enrique Gutierrez, Okinawa Institute of Science and Technology Graduate University
Daniel Baum, Zuse Institute Berlin
Kenji Doya, Okinawa Institute of Science and Technology Graduate University
Evan P. Economo, Okinawa Institute of Science and Technology Graduate University
*evropi.toulkeridou@oist.jp
*doya@oist.jp
*economo@oist.jp
Cite this dataset
Toulkeridou E., Gutierrez C.E., Baum D., Doya K., Economo E. P. (2023). Automated segmentation of insect anatomy from micro-CT images using deep learning [Dataset]. Dryad. https://doi.org/10.5061/dryad.qz612jmgv
Abstract
Three-dimensional (3D) imaging, such as micro-computed tomography (micro-CT), is increasingly being used by organismal biologists for precise and comprehensive anatomical characterization. However, the segmentation of anatomical structures remains a bottleneck in research, often requiring tedious manual work. Here, we propose a pipeline for the fully-automated segmentation of anatomical structures in micro-CT images utilizing state-of-the-art deep learning methods, selecting the ant brain as a test case. We implemented the U-Net architecture for 2D image segmentation for our convolutional neural network (CNN), combined with pixel-island detection. For training and validation of the network, we assembled a dataset of semi-manually segmented brain images of 76 ant species. The trained network predicted the brain area in ant images fast and accurately; its performance tested on validation sets showed good agreement between the prediction and the target, scoring 80% Intersection over Union (IoU) and 90% Dice Coefficient (F1) accuracy. While manual segmentation usually takes many hours for each brain, the trained network takes only a few minutes. Furthermore, our network is generalizable for segmenting the whole neural system in full-body scans, and works in tests on distantly related and morphologically divergent insects (e.g., fruit flies). The latter suggests that methods like the one presented here generally apply across diverse taxa. Our method makes the construction of segmented maps and the morphological quantification of different species more efficient and scalable to large datasets, a step toward a big data approach to organismal anatomy.
Image acquisition
In total, we collected one head scan per species from 76 different ant species. We used 2D cross-sections of planes along all three dimensions of our 3D brain scans.
Dataset includes:
The dataset includes micro-CT brain images of different ant species, named with a reference to the species, and their masks (segmented brain), split in testing (40% of images) and training set (60% of images). All images are saved as tiff files.
The genera and species of used ants as well as the number of slices (images) and voxel sizes of their respective head scans are provided in the table below:
Table 1 Taxon names of all species that were used for this study. The four columns indicate the number of slices per direction that were used from each specimen’s scan and the corresponding voxel size of the scan. | |||||
---|---|---|---|---|---|
Taxon code | xy slices | xz slices | yz slices | voxel size () | |
Acanthomyrmex_glabfemoralis | 300 | 150 | 100 | 1.84105 | |
Acromyrmex_versicolor | 130 | 200 | 330 | 3.21392 | |
Aenictus_paradentatus | 250 | 200 | 140 | 1.09831 | |
Anoplolepis_gracilipes | 370 | 220 | 180 | 1.00562 | |
Atta_texana | 250 | 300 | 200 | 1.51302 | |
Camponotus_nearcticus | 500 | 300 | 500 | 1.62596 | |
Camponotus_hyatti | 300 | 250 | 230 | 1.84968 | |
Camponotus_modoc | 200 | 400 | 300 | 2.53308 | |
Camponotus_vicinus | 350 | 350 | 200 | 2.25045 | |
Carebara_affinis | 300 | 250 | 330 | 0.914802 | |
Carebara_atoma | 300 | 200 | 200 | 0.51953 | |
Carebara_diversa | 270 | 100 | 130 | 0.990045 | |
Cephalotes_atratus | 200 | 450 | 450 | 2.18409 | |
Cephalotes_minutus | 400 | 250 | 400 | 1.38875 | |
Crematogaster_lineolata | 280 | 400 | 350 | 1.06144 | |
Crematogaster_pinicola | 320 | 400 | 430 | 1.01303 | |
Daceton_armigerum | 200 | 250 | 3.52417 | ||
Dolichoderus_pustulatus | 400 | 350 | 420 | 0.93929 | |
Dorylus_kohli | 400 | 400 | 400 | 1.05723 | |
Dorymyrmex_insanus_small | 400 | 300 | 400 | 1.12528 | |
Dorymyrmex_insanus_large | 250 | 250 | 250 | 1.30802 | |
Eciton_burchellii | 300 | 300 | 300 | 2.67454 | |
Formica_pallidefulva | 300 | 400 | 200 | 1.82821 | |
Formica_exsectoides | 300 | 200 | 400 | 2.04899 | |
Formica_gnava | 250 | 350 | 380 | 1.62671 | |
Formica_neogagates | 150 | 50 | 150 | 2.43199 | |
Formica_obscuripes | 320 | 300 | 280 | 2.3236 | |
Formica_polyctena | 450 | 300 | 400 | 1.79283 | |
Formica_rufa | 230 | 450 | 400 | 1.52425 | |
Gesomyrmex_howardi | 430 | 400 | 200 | 1.03468 | |
Gnamptogenys_sp | 300 | 1.94183 | |||
Labidus_praedator | 300 | 300 | 300 | 1.38717 | |
Lasius_fuliginosus | 400 | 200 | 150 | 2.18947 | |
Leptogenys_peuqueti | 250 | 100 | 100 | 3.0069 | |
Leptomyrmex_darlingtoni | 430 | 230 | 300 | 1.55856 | |
Linepithema_humile | 430 | 250 | 300 | 0.892673 | |
Liometopum_apiculatum | 200 | 350 | 330 | 1.66387 | |
Monomorium_floricola | 250 | 100 | 200 | 0.613733 | |
Monomorium_pharaonis | 350 | 350 | 380 | 0.673107 | |
Mystrium_camillae | 220 | 1.79928 | |||
Ooceraea_biroi | 300 | 200 | 200 | 0.870043 | |
Orectognathus_versicolor | 350 | 150 | 200 | 1.35041 | |
Parasyscia_cribrinobis | 310 | 310 | 350 | 1.15305 | |
Patagonomyrmex_angustus | 250 | 350 | 350 | 1.38728 | |
Pheidole_bicarinata | 460 | 0.783321 | |||
Pheidole_rhea | 300 | 250 | 300 | 1.11148 | |
Pogonomyrmex_badius | 250 | 150 | 300 | 2.24449 | |
Pogonomyrmex_brevispinosus | 250 | 250 | 250 | 1.941 | |
Pogonomyrmex_desertorum | 250 | 150 | 250 | 2.19439 | |
Pogonomyrmex_magnacanthus | 150 | 200 | 230 | 1.82453 | |
Pogonomyrmex_pima | 350 | 350 | 200 | 1.36688 | |
Pogonomyrmex_rugosus | 200 | 300 | 350 | 2.30809 | |
Pogonomyrmex_schmitti | 300 | 300 | 300 | 1.34886 | |
Pristomyrmex_profundus | 480 | 0.742557 | |||
Pseudomyrmex_ejectus | 350 | 300 | 330 | 1.11057 | |
Pseudomyrmex_ferrugineus | 400 | 1.29843 | |||
Pseudomyrmex_gracilis | 300 | 1.82826 | |||
Pseudomyrmex_triplarinus | 300 | 300 | 1.61201 | ||
Solenopsis_geminata | 250 | 250 | 270 | 1.34886 | |
Stenamma_heathi | 390 | 250 | 350 | 1.27715 | |
Stigmatomma_sp | 230 | 2.44965 | |||
Syscia_augustae | 420 | 0.787728 | |||
Temnothorax_curvispinosus | 450 | 350 | 400 | 0.656251 | |
Tetramorium_hispidum | 250 | 200 | 150 | 1.31344 | |
Tetramorium_pacificum | 430 | 1.37859 | |||
Tetraponera_sp | 310 | 1.12522 | |||
Trachymyrmex_carinatus | 100 | 100 | 220 | 2.04899 | |
Veromessor_andrei | 200 | 200 | 200 | 2.25027 | |
Veromessor_lariversi | 250 | 250 | 100 | 1.70412 | |
Veromessor_pergandei | 100 | 1.67449 | |||
Veromessor_smithi | 200 | 250 | 200 | 1.99029 | |
Brachymyrmex_depilis | 380 | 380 | 405 | 0.621502 | |
Cyphomyrmex_flavidus | 350 | 300 | 300 | 1.02553 | |
Pseudomyrmex_veneficus | 370 | 1.2192 | |||
Dolichoderus.mariae | 300 | 200 | 300 | 1.20906 | |
Tetramorium.immigrans | 290 | 240 | 200 | 1.35523 | |
Myrmelachista_nodigera | 330 | 0.708583 |