Refine
Document Type
- Article (2)
Language
- English (2) (remove)
Has Fulltext
- yes (2)
Is part of the Bibliography
- no (2)
Keywords
- 3D reconstruction (1)
- Automated monitoring (1)
- Chelicerata (1)
- Chiroptera (1)
- Digestive system (1)
- Evolution (1)
- Neuroanatomy (1)
- Pantopoda (1)
- Phylogeny (1)
- Ventral nerve cord (1)
- X-ray tomography (1)
- bat conservation (1)
- camera trap (1)
- deep learning (1)
- infrared light barrier (1)
Institute
Publisher
- BioMed Central (BMC) (1)
- Wiley (1)
Background
Pycnogonida (sea spiders) is the sister group of all other extant chelicerates (spiders, scorpions and relatives) and thus represents an important taxon to inform early chelicerate evolution. Notably, phylogenetic analyses have challenged traditional hypotheses on the relationships of the major pycnogonid lineages (families), indicating external morphological traits previously used to deduce inter-familial affinities to be highly homoplastic. This erodes some of the support for phylogenetic information content in external morphology and calls for the study of additional data classes to test and underpin in-group relationships advocated in molecular analyses. In this regard, pycnogonid internal anatomy remains largely unexplored and taxon coverage in the studies available is limited.
Results
Based on micro-computed X-ray tomography and 3D reconstruction, we created a comprehensive atlas of in-situ representations of the central nervous system and midgut layout in all pycnogonid families. Beyond that, immunolabeling for tubulin and synapsin was used to reveal selected details of ganglionic architecture. The ventral nerve cord consistently features an array of separate ganglia, but some lineages exhibit extended composite ganglia, due to neuromere fusion. Further, inter-ganglionic distances and ganglion positions relative to segment borders vary, with an anterior shift in several families. Intersegmental nerves target longitudinal muscles and are lacking if the latter are reduced. Across families, the midgut displays linear leg diverticula. In Pycnogonidae, however, complex multi-branching diverticula occur, which may be evolutionarily correlated with a reduction of the heart.
Conclusions
Several gross neuroanatomical features are linked to external morphology, including intersegmental nerve reduction in concert with trunk segment fusion, or antero-posterior ganglion shifts in partial correlation to trunk elongation/compaction. Mapping on a recent phylogenomic phylogeny shows disjunct distributions of these traits. Other characters show no such dependency and help to underpin closer affinities in sub-branches of the pycnogonid tree, as exemplified by the tripartite subesophageal ganglion of Pycnogonidae and Rhynchothoracidae. Building on this gross anatomical atlas, future studies should now aim to leverage the full potential of neuroanatomy for phylogenetic interrogation by deciphering pycnogonid nervous system architecture in more detail, given that pioneering work on neuron subsets revealed complex character sets with unequivocal homologies across some families.
BatNet: a deep learning-based tool for automated bat species identification from camera trap images
(2023)
Automated monitoring technologies can increase the efficiency of ecological data collection and support data-driven conservation. Camera traps coupled with infrared light barriers can be used to monitor temperate-zone bat assemblages at underground hibernacula, where thousands of individuals of multiple species can aggregate in winter. However, the broad-scale adoption of such photo-monitoring techniques is limited by the time-consuming bottleneck of manual image processing. Here, we present BatNet, an open-source, deep learning-based tool for automated identification of 13 European bat species from camera trap images. BatNet includes a user-friendly graphical interface, where it can be retrained to identify new bat species or to create site-specific models to improve detection accuracy at new sites. Model accuracy was evaluated on images from both trained and untrained sites, and in an ecological context, where community- and species-level metrics (species diversity, relative abundance, and species-level activity patterns) were compared between human experts and BatNet. At trained sites, model performance was high across all species (F1-score: 0.98–1). At untrained sites, overall classification accuracy remained high (96.7–98.2%), when camera placement was comparable to the training images (<3 m from the entrance; <45° angle relative to the opening). For atypical camera placements (>3 m or >45° angle), retraining the detector model with 500 site-specific annotations achieved an accuracy of over 95% at all sites. In the ecological case study, all investigated metrics were nearly identical between human experts and BatNet. Finally, we exemplify the ability to retrain BatNet to identify a new bat species, achieving an F1-score of 0.99 while maintaining high classification accuracy for all original species. BatNet can be implemented directly to scale up the deployment of camera traps in Europe and enhance bat population monitoring. Moreover, the pretrained model can serve as a baseline for transfer learning to automatize the image-based identification of bat species worldwide.