Morphological Data Sets Research Articles

Woody encroachment affects multiple dimensions of ant diversity in a neotropical savanna

Abstract Although savanna woody encroachment has become a global phenomenon, relatively little is known about its effects on multiple dimensions and levels of savanna biodiversity. Using a combination of field surveys, a species‐level phylogeny, and functional metrics drawn from a morphological dataset, we evaluated how the progressive increase in tree cover in a fire‐suppressed savanna landscape affects the taxonomic, functional, and phylogenetic diversity of neotropical ant communities, at both the alpha and beta levels. Ants were sampled along an extensive tree cover gradient, ranging from open savannas to forests established in former savanna areas. Variation in tree cover had a significant influence on all facets of diversity at the beta level, whereas at the alpha level tree cover variation affected the taxonomic and functional but not the phylogenetic diversity of the ant communities. In general, ant community responses to variation in tree cover were largely non‐linear as differences in taxonomic alpha diversity and in the taxonomic, functional, and phylogenetic composition of the sampled communities were often much stronger at the savanna/forest transition than at any other part of the gradient. This indicates that savanna ant communities switch rapidly to an alternative state once the savanna turns into forest. Ant communities in the newly formed forest areas lacked many of the species typical of the savanna habitats, suggesting that the maintenance of a fire suppression policy is likely to result in a decrease in ant diversity and in the homogenisation of the ant fauna at the landscape scale.

Ontogenetic variability of the intertympanic sinus distinguishes lineages within Crocodylia.

The phylogenetic relationships within crown Crocodylia remain contentious due to conflicts between molecular and morphological hypotheses. However, morphology-based datasets are mostly constructed on external characters, overlooking internal structures. Here, we use 3D geometric morphometrics to study the shape of the intertympanic sinus system in crown crocodylians during ontogeny, in order to assess its significance in a taxonomic context. Intertympanic sinus shape was found to be highly correlated with size and modulated by cranial shape during development. Still, adult sinus morphology distinguishes specimens at the family, genus and species level. We observe a clear distinction between Alligatoridae and Longirostres, a separation of different Crocodylus species and the subfossil Malagasy genus Voay, and a distinction between the Tomistoma and Gavialis lineages. Our approach is independent of molecular methods but concurs with the molecular topologies. Therefore, sinus characters could add significantly to morphological datasets, offering an alternative viewpoint to resolve problems in crocodylian relationships.

The taxonomic status of the kukri snake Oligodon arenarius Vassilieva, 2015 with a redescription of Oligodon macrurus (Angel, 1927) (Squamata, Serpentes, Colubridae)

We investigated the taxonomic status of the recently described kukri snakeOligodon arenariusVassilieva, 2015 and the morphologically similarOligodon macrurus(Angel, 1927), two species endemic to the southern coast of Vietnam. Based on phylogenetic analyses using three mitochondrial genes (12S–16S rRNA, cytochromeb), we recoveredO. arenariusandO. macrurusin a clade within theO. cyclurus-taeniatusspecies group, agreeing with previous intrageneric classifications. Genetic distances betweenO. arenariusandO. macrurusare extremely low (less than 0.5% based on 12S–16S) and renderO. arenariusparaphyletic. All preserved specimens ofO. arenariusandO. macrurusconvey little to no differences in color pattern, hemipenial morphology and osteological features; the latter of which is based on three dimensional micro computer tomography (µCT) scans of one specimen per species. Contrasting these results, univariate and multivariate analyses revealed significant differences in relative tail length, and the number of ventral and subcaudal scales between both species. Although the molecular and morphological datasets present conflicting results, integrating the evidence leads us to synonymizeO. arenariuswithO. macrurus. We provide a formal redescription ofO. macrurus, designate a neotype specimen to avoid future taxonomic confusion, and provide the first detailed osteological description of this species.Oligodon macrurussensu stricto is endemic to coastal dunefields and adjacent forest habitats in southern Vietnam, where ongoing human development, tourism and road mortality pose significant threats to its conservation. Consequently, we suggest thatO. macrurusshould be listed as “Vulnerable” based on the assessment criteria of the International Union for Conservation of Nature (IUCN).

Using Cluster Analysis to Overcome the Limits of Traditional Phenotype-Genotype Correlations: The Example of RYR1-Related Myopathies.

Thanks to advances in gene sequencing, RYR1-related myopathy (RYR1-RM) is now known to manifest itself in vastly heterogeneous forms, whose clinical interpretation is, therefore, highly challenging. We set out to develop a novel unsupervised cluster analysis method in a large patient population. The objective was to analyze the main RYR1-related characteristics to identify distinctive features of RYR1-RM and, thus, offer more precise genotype-phenotype correlations in a group of potentially life-threatening disorders. We studied 600 patients presenting with a suspicion of inherited myopathy, who were investigated using next-generation sequencing. Among them, 73 index cases harbored variants in RYR1. In an attempt to group genetic variants and fully exploit information derived from genetic, morphological, and clinical datasets, we performed unsupervised cluster analysis in 64 probands carrying monoallelic variants. Most of the 73 patients with positive molecular diagnoses were clinically asymptomatic or pauci-symptomatic. Multimodal integration of clinical and histological data, performed using a non-metric multi-dimensional scaling analysis with k-means clustering, grouped the 64 patients into 4 clusters with distinctive patterns of clinical and morphological findings. In addressing the need for more specific genotype-phenotype correlations, we found clustering to overcome the limits of the "single-dimension" paradigm traditionally used to describe genotype-phenotype relationships.

Syngnathoid Evolutionary History and the Conundrum of Fossil Misplacement.

Seahorses, pipefishes, trumpetfishes, shrimpfishes, and allies are a speciose, globally distributed clade of fishes that have evolved a large number of unusual body plans. The clade that includes all these forms, Syngnathoidei, has become a model for the study of life history evolution, population biology, and biogeography. Yet, the timeline of syngnathoid evolution has remained highly contentious. This debate is largely attributable to the nature of the syngnathoid fossil record, which is both poorly described and patchy for several major lineages. Although fossil syngnathoids have been used to calibrate molecular phylogenies, the interrelationships of extinct species and their affinities to major living syngnathoid clades have scarcely been quantitatively tested. Here, I use an expanded morphological dataset to reconstruct the evolutionary relationships and clade ages of fossil and extant syngnathoids. Phylogenies generated using different analytical methodologies are largely congruent with molecular phylogenetic trees of Syngnathoidei but consistently find novel placements for several key taxa used as fossil calibrators in phylogenomic studies. Tip-dating of the syngnathoid phylogeny finds a timeline for their evolution that differs slightly from the one inferred using molecular trees but is generally congruent with a post-Cretaceous diversification event. These results emphasize the importance of quantitatively testing the relationships of fossil species, particularly when they are critical to assessing divergence times.

A Construction of Object Detection Model for Acute Myeloid Leukemia

The evolution of bone marrow morphology is necessary in Acute Myeloid Leukemia (AML) prediction. It takes an enormous number of times to analyze with the standardization and inter-observer variability. Here, we proposed a novel AML detection model using a Deep Convolutional Neural Network (D-CNN). The proposed Faster R-CNN (Faster Region-Based CNN) models are trained with Morphological Dataset. The proposed Faster R-CNN model is trained using the augmented dataset. For overcoming the Imbalanced Data problem, data augmentation techniques are imposed. The Faster R-CNN performance was compared with existing transfer learning techniques. The results show that the Faster R-CNN performance was significant than other techniques. The number of images in each class is different. For example, the Neutrophil (segmented) class consists of 8,486 images, and Lymphocyte (atypical) class consists of eleven images. The dataset is used to train the CNN for single-cell morphology classification. The proposed work implies the high-class performance server called Nvidia Tesla V100 GPU (Graphics processing unit).

The unbearable uncertainty of panarthropod relationships.

Panarthropoda, the clade comprising the phyla Onychophora, Tardigrada and Euarthropoda, encompasses the largest majority of animal biodiversity. The relationships among the phyla are contested and resolution is key to understanding the evolutionary assembly of panarthropod bodyplans. Molecular phylogenetic analyses generally support monophyly of Onychophora and Euarthropoda to the exclusion of Tardigrada (Lobopodia hypothesis), which is also supported by some analyses of morphological data. However, analyses of morphological data have also been interpreted to support monophyly of Tardigrada and Euarthropoda to the exclusion of Onychophora (Tactopoda hypothesis). Support has also been found for a clade of Onychophora and Tardigrada that excludes Euarthropoda (Protarthropoda hypothesis). Here we show, using a diversity of phylogenetic inference methods, that morphological datasets cannot discriminate statistically between the Lobopodia, Tactopoda and Protarthropoda hypotheses. Since the relationships among the living clades of panarthropod phyla cannot be discriminated based on morphological data, we call into question the accuracy of morphology-based phylogenies of Panarthropoda that include fossil species and the evolutionary hypotheses based upon them.

‘Where is my family?’ Molecular and morphological data reveal the phylogenetic position and diversity of the enigmatic handsome fungus beetle genus Anamycetaea Strohecker, 1975 (Coleoptera, Coccinelloidea)

The genus Anamycetaea Strohecker, 1975, established for Anamycetaea keralae, a single species from India, was originally placed in the diverse endomychid subfamily Mycetaeinae and has subsequently been considered a member of the subfamily Anamorphinae based on closed mesocoxal cavities, a postulated synapomorphy of this group. Recent molecular research resulted in raising Anamorphinae to family level and revealed this group to be distantly related to Endomychidae sensu stricto. However, Anamycetaea has been ‘neglected’ since description. Our detailed study of this genus has been possible due to new material collected from Oriental and Australian regions. Striking overall similarity to the endomychine genus Tharina and a tentorium with anterior arms fused medially (separated in almost all Anamorphidae) have raised our doubts and led to further investigation of the phylogenetic placement of this enigmatic genus within Endomychidae sensu lato (handsome fungus beetles). Phylogenetic analyses of molecular and morphological datasets were conducted under Bayesian (BI), maximum likelihood (ML) and parsimony (MP) frameworks. Our results recovered Anamycetaea as belonging to the family Endomychidae, in the subfamily Endomychinae, distant from Anamorphidae. The close affinity to Stenotarsus and allies was strongly supported in all analyses. Based on material studied, A. keralae is described in detail here and includes description of previously unknown male genitalia. Four new species are also described, extending the ragne of the genus to the Australian region: Anamycetaea borneensis sp. nov. (from Borneo), A. novoguineensis sp. nov. and A. papuensis sp. nov. (from Papua New Guinea) and A. queenslandica sp. nov. (from Australia). Illustrations of morphological details and diagnoses are provided for each species. A key to the species of the genus is also presented. ZooBank: urn:lsid:zoobank.org:pub:90BAA954-7849-4FA9-997B-061FE7BB5702

Tooth morphology fusion technique is more accurate than conventional technique in transferring morphology of provisional to definitive screw-retained, implant-supported crown: A preliminary intervention study.

To compare the accuracy of the tooth morphology fusion (TMF) digital technique and customized impression transfer coping (conventional) technique when transferring the morphology of a provisional crown to a definitive screw-retained implant-supported crown. Six cases of partial edentulism (one anterior and five posterior) treated with oral implant placement in our clinic for the loss of three or fewer teeth in the maxilla or mandible between April 2017 and September 2018 were included. After implant placement and re-entry surgery, provisional restorations were made and adjusted to obtain the ideal morphology. Two definitive restorations were constructed by transferring the complete morphology of the provisional restorations, including the subgingival contour, using the TMF digital and conventional techniques. Three sets of surface morphological data were obtained using a desktop scanner. The three-dimensional total discrepancy volume (TDV) between the provisional restoration (reference) and the two definitive restorations was digitally measured by overlapping the surface data of the stone cast using the Boolean operation. Each TDV ratio (%) was calculated by dividing the TDV by the volume of provisional restoration. The median TDV ratios for TMF and conventional techniques were compared using the Wilcoxon signed-rank test. The median TDV ratio between provisional and definitive restorations constructed using the TMF digital technique (8.05%) was significantly lower than that obtained using the conventional technique (13.56%, P < 0.05). In this preliminary intervention study, the TMF digital technique was more accurate than the conventional technique for the transfer of morphology from provisional to definitive prosthesis.

Shifting agriculture and a depleting aquifer: implications of row-crop farming on mule deer population performance

Context Conversion of native vegetation to cropland is one of the most widespread anthropogenic landscape alterations, particularly in the Great Plains region of the United States. Mule deer occur throughout the Great Plains; however, it is the south-eastern edge of their geographical distribution, and few populations coincide with dense cropland. The rapidly depleting Ogallala Aquifer supplies irrigation to row-crops throughout the region, which will likely shift towards dryland agricultural practices in the near future. Aims We sought to understand how cropland use influences morphology, body condition indices, reproductive output, and survival of free-ranging mule deer. Methods We accumulated a multi-year, longitudinal dataset of movement and morphology for 146 mule deer in the Texas Panhandle. We linked seasonal cropland use with observed morphology, body condition metrics, and reproductive output via linear mixed-effect modelling and assessed the influence of cropland on annual survival by using Cox proportional hazard models. Key results Mule deer that did not use cropland at any time during the year exhibited morphological and nutritional indices similar to those that did; except body-fat percentage being greater for mature (≥4-year-old) males that used cropland. Further, cropland use did not predict survival probability. Analyses of cropland use during seasons defined by life-stage showed context-dependent nutritional benefits. Use of cropland during winter following reproduction demonstrated an increase in young (≤3-year-old) male antler size and body mass and summer crop use increased body condition for all males. Female mule deer that utilised cropland before pregnancy had increased probability of successful reproduction, demonstrating a potential capital investment strategy in reproduction. Conclusions Cropland does not limit morphology or survival of mule deer; however, additive use of row-crops can provide a nutritional buffer and enhanced reproductive output for individuals that choose to utilise it. Implications Our study demonstrates important population-level interactions with the environment for a species near the extent of their geographical distribution. Conversion of row-crop farming from aquifer depletion or climate shifts may not diminish mule deer populations, but these changes may alter specific habitat-nutritional health relationships that can influence population performance and future conservation efforts.

Efficient and Lightweight Neuron Morphology Classification Using Gabor Convolutional Networks

Due to the complex and diverse morphology of neurons, achieving accurate classification of neurons is not an easy task. Most existing methods for neuron classification are based on traditional machine learning algorithms, which not only have high design costs but also poor transferability. Gabor Orientation Filters (GoFs), obtained by modulating Gabor filters and traditional convolutional kernels, can better extract geometric features of neurons and enhance the adaptability of deep learning features to GoFs' orientation and scale changes. Therefore, this paper proposes an efficient and lightweight Gabor Convolutional Network (GCN) for neuron morphology classification based on the geometric shape data of neurons and the spatial transformation capabilities of Gabor filters. The network consists of five consecutive Gabor convolutional layers for extracting geometric features of neurons and two linear layers for neuron morphology classification. Experimental results show that this method achieves accuracy rates of 94.26%, 87.16%, and 87.72% for binary classification and 96.52%, 85.96%, and 86.73% for 12-class classification on the Img_raw, Img_resample, and Img_XYalign neuron morphology datasets, respectively. Meanwhile, compared with classical classification models, the experimental results show that this method can capture enough neuron morphology features with fewer convolutional layers, and it has good classification accuracy and fewer model parameters.

Widespread convergent morphological evolution within the largest family of songbirds.

Although convergence is a common evolutionary phenomenon, few studies have quantified its prevalence across a large, densely sampled clade. Large-scale phylogenies and the advent of novel computational methods facilitate more robust identification of convergent events and their statistical significance. The tanagers (Aves: Thraupidae), the largest family of songbirds, offer an excellent opportunity to study the extent of phenotypic convergence in response to similar ecological pressures on a continental scale. To investigate convergence in the group, we used the largest phylogenetic and multivariate morphological dataset to date for the clade. First, we used phylogenetic comparative analyses to show a correlation between diet and aspects of bill shape. We then investigated our dataset for the presence and magnitude of convergent events and assessed significance through simulations and modeling analyses. Overall, we found that around half (45.3%) of species and clades we tested have converged in morphological space more than would be expected by chance alone. Our study shows that across Thraupidae, various bill shapes have evolved convergently to fill multiple distinct sections of ecological niche space, reflecting a signal of ecological opportunity and structural constraints.

An improved Vision Transformer model for the recognition of blood cells

Leukemia is a common, multiple and dangerous blood disease, whose early diagnosis and treatment are very important. At present, the diagnosis of leukemia heavily relies on morphological examination of blood cell images by pathologists, which is tedious and time-consuming. Meanwhile, the diagnostic results are highly subjective, which may lead to misdiagnosis and missed diagnosis. To address the gap above, we proposed an improved Vision Transformer model for blood cell recognition. First, a faster R-CNN network was used to locate and extract individual blood cell slices from original images. Then, we split the single-cell image into multiple image patches and put them into the encoder layer for feature extraction. Based on the self-attention mechanism of the Transformer, we proposed a sparse attention module which could focus on the discriminative parts of blood cell images and improve the fine-grained feature representation ability of the model. Finally, a contrastive loss function was adopted to further increase the inter-class difference and intra-class consistency of the extracted features. Experimental results showed that the proposed module outperformed the other approaches and significantly improved the accuracy to 91.96% on the Munich single-cell morphological dataset of leukocytes, which is expected to provide a reference for physicians' clinical diagnosis.

Forty new specimens of Ichthyornis provide unprecedented insight into the postcranial morphology of crownward stem group birds.

Ichthyornis has long been recognized as a pivotally important fossil taxon for understanding the latest stages of the dinosaur-bird transition, but little significant new postcranial material has been brought to light since initial descriptions of partial skeletons in the 19th Century. Here, we present new information on the postcranial morphology of Ichthyornis from 40 previously undescribed specimens, providing the most complete morphological assessment of the postcranial skeleton of Ichthyornis to date. The new material includes four partially complete skeletons and numerous well-preserved isolated elements, enabling new anatomical observations such as muscle attachments previously undescribed for Mesozoic euornitheans. Among the elements that were previously unknown or poorly represented for Ichthyornis, the new specimens include an almost-complete axial series, a hypocleideum-bearing furcula, radial carpal bones, fibulae, a complete tarsometatarsus bearing a rudimentary hypotarsus, and one of the first-known nearly complete three-dimensional sterna from a Mesozoic avialan. Several pedal phalanges are preserved, revealing a remarkably enlarged pes presumably related to foot-propelled swimming. Although diagnosable as Ichthyornis, the new specimens exhibit a substantial degree of morphological variation, some of which may relate to ontogenetic changes. Phylogenetic analyses incorporating our new data and employing alternative morphological datasets recover Ichthyornis stemward of Hesperornithes and Iaceornis, in line with some recent hypotheses regarding the topology of the crownward-most portion of the avian stem group, and we establish phylogenetically-defined clade names for relevant avialan subclades to help facilitate consistent discourse in future work. The new information provided by these specimens improves our understanding of morphological evolution among the crownward-most non-neornithine avialans immediately preceding the origin of crown group birds.

Integrative taxonomy of the Lauraceae-feeding species of the genus Stephanitis (Hemiptera, Heteroptera, Tingidae) from Japan

Stephanitis Stål, 1873 (Hemiptera, Heteroptera, Tingidae, Tinginae, Tingini) is one of the most diverse lace bug genera in East Asia with over 50 species, many of them feeding on Ericaceae and Lauraceae. The Japanese species feeding on Lauraceae have high intraspecific and poor interspecific variation, so identification of some previous distribution records (mainly S. (Norba) aperta Horváth, 1912 and closely related taxa) is doubtful. In the present study, 5,080 specimens of Lauraceae-feeding species of Stephanitis from Japan were identified to species based on morphology, and DNA barcodes of the mitochondrial cytochrome c oxidase subunit (COI) region for 53 of those individuals were also obtained. The combined morphological and molecular evidence dataset showed that 10 species of Stephanitis feed on lauraceous trees in Japan: S. (Stephanitis) ambigua Horváth, 1912, S. (Norba) aperta, S. (N.) exigua Horváth, 1912, S. (N.) hayashiisp. nov., S. (N.) hiurai Takeya, 1963, S. (N.) ishikawaisp. nov., S. (N.) mendica Horváth, 1912, S. (S.) tabidula Horváth, 1912, S. (S.) takeyai Drake &amp; Maa, 1955 and S. (S.) tomokuniisp. nov. Additionally, a new synonymy, S. (S.) fasciicarina Takeya, 1931, syn. nov. with S. (S.) tabidula, is proposed, a key for identifying the 10 species is provided and their distribution ranges and host plant relationships are presented.

Body size as a magic trait in two plant-feeding insect species.

When gene flow accompanies speciation, recombination can decouple divergently selected loci and loci conferring reproductive isolation. This barrier to sympatric divergence disappears when assortative mating and disruptive selection involve the same "magic" trait. Although magic traits could be widespread, the relative importance of different types of magic traits to speciation remains unclear. Because body size frequently contributes to host adaptation and assortative mating in plant-feeding insects, we evaluated several magic trait predictions for this trait in a pair of sympatric Neodiprion sawfly species adapted to different pine hosts. A large morphological dataset revealed that sawfly adults from populations and species that use thicker-needled pines are consistently larger than those that use thinner-needled pines. Fitness data from recombinant backcross females revealed that egg size is under divergent selection between the preferred pines. Lastly, mating assays revealed strong size-assortative mating within and between species in three different crosses, with the strongest prezygotic isolation between populations that have the greatest interspecific size differences. Together, our data support body size as a magic trait in pine sawflies and possibly many other plant-feeding insects. Our work also demonstrates how intraspecific variation in morphology and ecology can cause geographic variation in the strength of prezygotic isolation.

Evaluating the Impact of Anatomical Partitioning on Summary Topologies Obtained with Bayesian Phylogenetic Analyses of Morphological Data.

Morphological data are a fundamental source of evidence to reconstruct the Tree of Life, and Bayesian phylogenetic methods are increasingly being used for this task. Bayesian phylogenetic analyses require the use of evolutionary models, which have been intensively studied in the past few years, with significant improvements to our knowledge. Notwithstanding, a systematic evaluation of the performance of partitioned models for morphological data has never been performed. Here we evaluate the influence of partitioned models, defined by anatomical criteria, on the precision and accuracy of summary tree topologies considering the effects of model misspecification. We simulated datasets using partitioning schemes, trees, and other properties obtained from two empirical datasets, and conducted Bayesian phylogenetic analyses. Additionally, we reanalyzed 32 empirical datasets for different groups of vertebrates, applying unpartitioned and partitioned models, and, as a focused study case, we reanalyzed a dataset including living and fossil armadillos, testing alternative partitioning hypotheses based on functional and ontogenetic modules. We found that, in general, partitioning by anatomy has little influence on summary topologies analyzed under alternative partitioning schemes with a varying number of partitions. Nevertheless, models with unlinked branch lengths, which account for heterotachy across partitions, improve topological precision at the cost of reducing accuracy. In some instances, more complex partitioning schemes led to topological changes, as tested for armadillos, mostly associated with models with unlinked branch lengths. We compare our results with other empirical evaluations of morphological data and those from empirical and simulation studies of the partitioning of molecular data, considering the adequacy of anatomical partitioning relative to alternative methods of partitioning morphological datasets. [Evolutionary rates; heterogeneity; morphology; Mk model; partition; topology.].

Sperm phylogeny of Characidae (Teleostei, Characiformes)

AbstractThe phylogeny of the very diverse Neotropical fish family Characidae has been the subject of several recent contributions based on morphological characters, molecular data or both in combined analyses. In cases of conflict between these kinds of data, resolution by combined analyses most often tends to agree with the molecular evidence, given the disproportionate number of characters it contains in comparison with morphological datasets. This happens especially after the advent of massive DNA sequencing methods. In this contribution, we present the most comprehensive set of characters from sperm and spermiogenesis of the Characidae. Since these traits are not expected to be functionally correlated with the general morphology or molecular markers, we consider them a third source of data. We provide a phylogenetic analysis from a combined dataset of seven molecular markers (6444 characters), general morphology (520 characters) and reproductive features (94 characters) coded for 165 species of characiform fishes. Parsimony analyses were done under extended implied weighting under 30 different combinations of weighting schemes and strengths. Most parsimonious trees from two different weighting conditions were selected as representative samples of the obtained topologies, in order to evaluate the performance of the reproductive characters. One of these hypotheses is more conservative regarding the currently accepted phylogenies and the other is the most parsimonious tree that we found as the best correlated with the morphological data. Reproductive characters are shown to be more homoplastic than general morphology and DNA, but provided synapomorphies for 23–24 nodes that had no morphological synapomorphies, justifying their use in phylogenetic analyses. Also, in combination with data from general morphology and considering details of the phylogenetic analysis, they showed to have the potential to challenge well‐established hypotheses based on molecular data.

The resurrection of Cerasommatidiidae, an enigmatic group of coccinelloid beetles (Coleoptera: Coccinelloidea) based on molecular and morphological evidence

Abstract The family Cerasommatidiidae was proposed by Brèthes in 1925 for his new genus Cerasommatidia from Brazil, described as an intermediate taxon between Endomychidae and Coccinellidae (Coccinelloidea). This group was neglected for decades until 1994 when the resemblance of Cerasommatidia with the eupsilobiine genus Ibicarella was noticed, and Cerasommatidiidae was synonymized with Eupsilobiinae (Endomychidae). Based on the newly discovered material from the Neotropical and Afrotropical regions with clear resemblance to Cerasommatidia and Ibicarella, the phylogenetic placement of these taxa and new allied genera was reassessed based on phylogenetic analyses of molecular, morphological and combined datasets. The data were analysed under Bayesian (BI), maximum likelihood (ML) and parsimony (MP) frameworks. Our results support the restoration of Cerasommatidiidae as a distinct family in Coccinelloidea. Close affinity of Cerasommatidiidae with Coccinellidae, Eupsilobiidae and Mycetaeidae is recovered and the sister-relationship to Mycetaeidae is highly supported. Ibicarella is found to be a junior synonym of Cerasommatidia. Three new genera and six new species are described and illustrated: Yamuy gen. nov. (type species, Y. marginatus sp. nov.; and Y. constratus sp. nov.) and Karumbe gen. nov. (type species, K. geiseri sp. nov.; K. brethesi sp. nov. and K. pakaluki sp. nov.) from the Neotropics and Mahavelo gen. nov. (type species, M. madagasus sp. nov.) from Madagascar.

Picasso: An Open-Source Machine Learning Schema for Annotating Images in Hematology

Background: Prediction models to support clinical decision-making are an integral part of medicine. Image recognition and diagnosis are essential in multiple diseases, especially malignant hematologic diseases, where histopathological images are reviewed manually, and an opportunity to automate image analysis with diagnosis-supporting tools exists. Machine learning (ML) is a process where an algorithm creates a predictive model by learning from training data and uncovering relationships between input variables (X) and output variables (Y). For this, a training dataset with labeled examples is necessary. After the learning has occurred, a predictive model is generated that can now be used to make predictions on new input that the model has never seen before. To train a machine learning model appropriately, all input should be annotated in a standard manner, and this process is very time-consuming. One way to make it simpler is by using a schema. A schema is a collected set of rules that must be followed each time in describing the relevant features of an image. For every training image in a dataset, the relevant components are labeled manually to fit the schema and reviewed by trained human examiners, making the process tedious, time-consuming, and difficult to standardize. We created a programmatic approach to annotation where a machine learning algorithm can detect significant features in AML images, annotate the images with features that fit a schema, and feed them forward into training data. Moreover, we have created this on an open-source platform that can be widely used in a resource-limited setting across the globe. Methods: A supervised machine learning approach with a convolutional neural network (CNN) was used for image processing based on a dataset that contained 270 pathological images (confirmed AML) and 30 non-pathological images (no AML) from the public Munich AML Morphology Dataset. The pathological images were from peripheral blood smears of patients diagnosed with AML at Munich University Hospital between 2014 and 2017. The non-pathological controls were taken from patients without hematological malignancy. Initially, for training, all annotation features from the Munich AML dataset are included for both pathological and non-pathological images. The neural network was written and trained on open-source software, TensorFlow using the open-source Keras library with parallel threading and on a multicore GPU machine. The resulting predictive model answers two questions: Does a given image have cells that resemble a blast character, or if the image has cells that belong to non-pathological blood smears. Results: The neural network trained on 300 total images, and for external validation, we used another subset of 100 unlabeled images from the Munich AML Morphology Dataset. Our model could predict blast-like features accurately in 89% of the new images, 6% of new images were unclassifiable, the false positive rate was 3%, and the false negative rate was 2%. After initial prediction, the neural network could accurately annotate unlabeled images 86% of the time in pathological samples and 96% in non-pathological samples. Our training model achieved an AUC of 0.82. Conclusion: A CNN trained on a modest dataset with supervised learning, enhanced with ensemble learning and K-fold cross-validation, can be used to recognize features such as blast cells from histopathological images and label images with a high degree of accuracy. In a data-driven machine learning algorithm such as a neural network, classification performance significantly increases with more available training sample images. Therefore, a more extensive training dataset with more robust hardware is necessary to generate a more sophisticated predictive model. Using CNN-enhancing methodologies can allow for model training in a resource-limiting setting. Future work will focus on using a more extensive pre-trained database to evaluate the performance of our network in a real-world setting. The annotation framework can be expanded to include disease-associated features for use in other domains, such as hematological education, patient resources, and patient education. This open-source platform can support several niches in the future that are currently only served by expensive commercial applications. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal