An integrated micro- and macroarchitectural analysis of the Drosophila brain by computer-assisted serial section electron microscopy

Neurons with larval synaptic targets pioneer the later nervous system in the annelid Malacoceros fuliginosus

Comparative studies on the development of nervous systems have a significant impact on understanding animal nervous system evolution. Nevertheless, an important question is to what degree neuronal structures, which play an important role in early stages, become part of the adult nervous system or are relevant for its formation. This is likely in many direct developers, but it is not the case in forms with catastrophic metamorphosis. It is not clear in many forms with gradual metamorphosis. This introduces uncertainty in tracing the evolution of nervous systems and of larval forms. One of the prominent larval characteristics of numerous planktonic marine organisms is the epidermal ciliation used for swimming and steering, which disappears during metamorphosis. Therefore, the neuronal elements controlling the ciliary beating are often assumed to vanish with the cilia and regarded as purely larval specializations. With volume EM, we followed the neuronal targets of the very first pioneer neurons at the apical and posterior ends of the larva of the annelid Malacoceros fuliginosus. We observed that all of these pioneers appear to have a dual function. Although they are laying the paths for the later adult nervous system, they also make synaptic contact with the main ciliated ring of the larva. We propose that the formation of the later adult nervous system and the innervation of the larval locomotory organ are indeed closely linked to each other. This has implications for understanding the early nervous system development of marine larvae and for existing hypotheses on nervous system evolution.

State-of-the-Art Deep Learning Methods for Microscopic Image Segmentation: Applications to Cells, Nuclei, and Tissues

Microscopic image segmentation (MIS) is a fundamental task in medical imaging and biological research, essential for precise analysis of cellular structures and tissues. Despite its importance, the segmentation process encounters significant challenges, including variability in imaging conditions, complex biological structures, and artefacts (e.g., noise), which can compromise the accuracy of traditional methods. The emergence of deep learning (DL) has catalyzed substantial advancements in addressing these issues. This systematic literature review (SLR) provides a comprehensive overview of state-of-the-art DL methods developed over the past six years for the segmentation of microscopic images. We critically analyze key contributions, emphasizing how these methods specifically tackle challenges in cell, nucleus, and tissue segmentation. Additionally, we evaluate the datasets and performance metrics employed in these studies. By synthesizing current advancements and identifying gaps in existing approaches, this review not only highlights the transformative potential of DL in enhancing diagnostic accuracy and research efficiency but also suggests directions for future research. The findings of this study have significant implications for improving methodologies in medical and biological applications, ultimately fostering better patient outcomes and advancing scientific understanding.

Turbinal bones are still one of the last frontiers of the tetrapod skull: hypotheses, challenges and perspectives

Turbinals are bony or cartilaginous structures that are present in the nasal cavity of most tetrapods. They are involved in key functions such as olfaction, heat, and moisture conservation, as well as protection of the respiratory tract. Despite recent studies that challenged long-standing hypotheses about their physiological and genomic correlation, turbinals remain largely unexplored, particularly for non-mammalian species. Herein, we review and synthesise the current knowledge of turbinals using an integrative approach that includes comparative anatomy, physiology, histology and genomics. In addition, we provide synonyms and correspondences of tetrapod turbinals from about 80 publications. This work represents a first step towards drawing hypotheses of homology for the whole clade, and provides a strong basis to develop new research avenues.

Channel prior convolutional attention for medical image segmentation

Characteristics such as low contrast and significant organ shape variations are often exhibited in medical images. The improvement of segmentation performance in medical imaging is limited by the generally insufficient adaptive capabilities of existing attention mechanisms. An efficient Channel Prior Convolutional Attention (CPCA) method is proposed in this paper, supporting the dynamic distribution of attention weights in both channel and spatial dimensions. Spatial relationships are effectively extracted while preserving the channel prior by employing a multi-scale depth-wise convolutional module. The ability to focus on informative channels and important regions is possessed by CPCA. A segmentation network called CPCANet for medical image segmentation is proposed based on CPCA. CPCANet is validated on two publicly available datasets. Improved segmentation performance is achieved by CPCANet while requiring fewer computational resources through comparisons with state-of-the-art algorithms. Our code is publicly available at https://github.com/Cuthbert-Huang/CPCANet.

Segment anything model for medical images?

The Segment Anything Model (SAM) is the first foundation model for general image segmentation. It has achieved impressive results on various natural image segmentation tasks. However, medical image segmentation (MIS) is more challenging because of the complex modalities, fine anatomical structures, uncertain and complex object boundaries, and wide-range object scales. To fully validate SAM's performance on medical data, we collected and sorted 53 open-source datasets and built a large medical segmentation dataset with 18 modalities, 84 objects, 125 object-modality paired targets, 1050K 2D images, and 6033K masks. We comprehensively analyzed different models and strategies on the so-called COSMOS 1050K dataset. Our findings mainly include the following: (1) SAM showed remarkable performance in some specific objects but was unstable, imperfect, or even totally failed in other situations. (2) SAM with the large ViT-H showed better overall performance than that with the small ViT-B. (3) SAM performed better with manual hints, especially box, than the Everything mode. (4) SAM could help human annotation with high labeling quality and less time. (5) SAM was sensitive to the randomness in the center point and tight box prompts, and may suffer from a serious performance drop. (6) SAM performed better than interactive methods with one or a few points, but will be outpaced as the number of points increases. (7) SAM's performance correlated to different factors, including boundary complexity, intensity differences, etc. (8) Finetuning the SAM on specific medical tasks could improve its average DICE performance by 4.39% and 6.68% for ViT-B and ViT-H, respectively. Codes and models are available at: https://github.com/yuhoo0302/Segment-Anything-Model-for-Medical-Images. We hope that this comprehensive report can help researchers explore the potential of SAM applications in MIS, and guide how to appropriately use and develop SAM.

FANet: A Feedback Attention Network for Improved Biomedical Image Segmentation

The increase of available large clinical and experimental datasets has contributed to a substantial amount of important contributions in the area of biomedical image analysis. Image segmentation, which is crucial for any quantitative analysis, has especially attracted attention. Recent hardware advancement has led to the success of deep learning approaches. However, although deep learning models are being trained on large datasets, existing methods do not use the information from different learning epochs effectively. In this work, we leverage the information of each training epoch to prune the prediction maps of the subsequent epochs. We propose a novel architecture called feedback attention network (FANet) that unifies the previous epoch mask with the feature map of the current training epoch. The previous epoch mask is then used to provide hard attention to the learned feature maps at different convolutional layers. The network also allows rectifying the predictions in an iterative fashion during the test time. We show that our proposed feedback attention model provides a substantial improvement on most segmentation metrics tested on seven publicly available biomedical imaging datasets demonstrating the effectiveness of FANet. The source code is available at https://github.com/nikhilroxtomar/FANet.

MDF-Net: A Multi-Scale Dynamic Fusion Network for Breast Tumor Segmentation of Ultrasound Images

Breast tumor segmentation of ultrasound images provides valuable information of tumors for early detection and diagnosis. Accurate segmentation is challenging due to low image contrast between areas of interest; speckle noises, and large inter-subject variations in tumor shape and size. This paper proposes a novel Multi-scale Dynamic Fusion Network (MDF-Net) for breast ultrasound tumor segmentation. It employs a two-stage end-to-end architecture with a trunk sub-network for multiscale feature selection and a structurally optimized refinement sub-network for mitigating impairments such as noise and inter-subject variation via better feature exploration and fusion. The trunk network is extended from UNet++ with a simplified skip pathway structure to connect the features between adjacent scales. Moreover, deep supervision at all scales, instead of at the finest scale in UNet++, is proposed to extract more discriminative features and mitigate errors from speckle noise via a hybrid loss function. Unlike previous works, the first stage is linked to a loss function of the second stage so that both the preliminary segmentations and refinement subnetworks can be refined together at training. The refinement sub-network utilizes a structurally optimized MDF mechanism to integrate preliminary segmentation information (capturing general tumor shape and size) at coarse scales and explores inter-subject variation information at finer scales. Experimental results from two public datasets show that the proposed method achieves better Dice and other scores over state-of-the-art methods. Qualitative analysis also indicates that our proposed network is more robust to tumor size/shapes, speckle noise and heavy posterior shadows along tumor boundaries. An optional post-processing step is also proposed to facilitate users in mitigating segmentation artifacts. The efficiency of the proposed network is also illustrated on the "Electron Microscopy neural structures segmentation dataset". It outperforms a state-of-the-art algorithm based on UNet-2022 with simpler settings. This indicates the advantages of our MDF-Nets in other challenging image segmentation tasks with small to medium data sizes.

Learning and memory using Drosophila melanogaster: a focus on advances made in the fifth decade of research

In the last decade, researchers using Drosophila melanogaster have made extraordinary progress in uncovering the mysteries underlying learning and memory. This progress has been propelled by the amazing toolkit available that affords combined behavioral, molecular, electrophysiological, and systems neuroscience approaches. The arduous reconstruction of electron microscopic images resulted in a first-generation connectome of the adult and larval brain, revealing complex structural interconnections between memory-related neurons. This serves as substrate for future investigations on these connections and for building complete circuits from sensory cue detection to changes in motor behavior. Mushroom body output neurons (MBOn) were discovered, which individually forward information from discrete and non-overlapping compartments of the axons of mushroom body neurons (MBn). These neurons mirror the previously discovered tiling of mushroom body axons by inputs from dopamine neurons and have led to a model that ascribes the valence of the learning event, either appetitive or aversive, to the activity of different populations of dopamine neurons and the balance of MBOn activity in promoting avoidance or approach behavior. Studies of the calyx, which houses the MBn dendrites, have revealed a beautiful microglomeruluar organization and structural changes of synapses that occur with long-term memory (LTM) formation. Larval learning has advanced, positioning it to possibly lead in producing new conceptual insights due to its markedly simpler structure over the adult brain. Advances were made in how cAMP response element-binding protein interacts with protein kinases and other transcription factors to promote the formation of LTM. New insights were made on Orb2, a prion-like protein that forms oligomers to enhance synaptic protein synthesis required for LTM formation. Finally, Drosophila research has pioneered our understanding of the mechanisms that mediate permanent and transient active forgetting, an important function of the brain along with acquisition, consolidation, and retrieval. This was catalyzed partly by the identification of memory suppressor genes-genes whose normal function is to limit memory formation.

A novel registration method for long-serial section images of EM with a serial split technique based on unsupervised optical flow network

MOTIVATION

The registration of serial section electron microscope images is a critical step in reconstructing biological tissue volumes, and it aims to eliminate complex nonlinear deformations from sectioning and replicate the correct neurite structure. However, due to the inherent properties of biological structures and the challenges posed by section preparation of biological tissues, achieving an accurate registration of serial sections remains a significant challenge. Conventional nonlinear registration techniques, which are effective in eliminating nonlinear deformation, can also eliminate the natural morphological variation of neurites across sections. Additionally, accumulation of registration errors alters the neurite structure.

RESULTS

This article proposes a novel method for serial section registration that utilizes an unsupervised optical flow network to measure feature similarity rather than pixel similarity to eliminate nonlinear deformation and achieve pairwise registration between sections. The optical flow network is then employed to estimate and compensate for cumulative registration error, thereby allowing for the reconstruction of the structure of biological tissues. Based on the novel serial section registration method, a serial split technique is proposed for long-serial sections. Experimental results demonstrate that the state-of-the-art method proposed here effectively improves the spatial continuity of serial sections, leading to more accurate registration and improved reconstruction of the structure of biological tissues.

AVAILABILITY AND IMPLEMENTATION

The source code and data are available at https://github.com/TongXin-CASIA/EFSR.

Host cell egress of Brucella abortus requires BNIP3L-mediated mitophagy

The facultative intracellular pathogen Brucella abortus interacts with several organelles of the host cell to reach its replicative niche inside the endoplasmic reticulum. However, little is known about the interplay between the intracellular bacteria and the host cell mitochondria. Here, we showed that B. abortus triggers substantive mitochondrial network fragmentation, accompanied by mitophagy and the formation of mitochondrial Brucella-containing vacuoles during the late steps of cellular infection. Brucella-induced expression of the mitophagy receptor BNIP3L is essential for these events and relies on the iron-dependent stabilisation of the hypoxia-inducible factor 1α. Functionally, BNIP3L-mediated mitophagy appears to be advantageous for bacterial exit from the host cell as BNIP3L depletion drastically reduces the number of reinfection events. Altogether, these findings highlight the intricate link between Brucella trafficking and the mitochondria during host cell infection.

Seg2Link: an efficient and versatile solution for semi-automatic cell segmentation in 3D image stacks

Recent advances in microscopy techniques, especially in electron microscopy, are transforming biomedical studies by acquiring large quantities of high-precision 3D cell image stacks. To examine cell morphology and connectivity in organs such as the brain, scientists need to conduct cell segmentation, which extracts individual cell regions of different shapes and sizes from a 3D image. This is challenging due to the indistinct images often encountered in real biomedical research: in many cases, automatic segmentation methods inevitably contain numerous mistakes in the segmentation results, even when using advanced deep learning methods. To analyze 3D cell images effectively, a semi-automated software solution is needed that combines powerful deep learning techniques with the ability to perform post-processing, generate accurate segmentations, and incorporate manual corrections. To address this gap, we developed Seg2Link, which takes deep learning predictions as inputs and use watershed 2D + cross-slice linking to generate more accurate automatic segmentations than previous methods. Additionally, it provides various manual correction tools essential for correcting mistakes in 3D segmentation results. Moreover, our software has been optimized for efficiently processing large 3D images in diverse organisms. Thus, Seg2Link offers an practical solution for scientists to study cell morphology and connectivity in 3D image stacks.

A survey on recent trends in deep learning for nucleus segmentation from histopathology images

Nucleus segmentation is an imperative step in the qualitative study of imaging datasets, considered as an intricate task in histopathology image analysis. Segmenting a nucleus is an important part of diagnosing, staging, and grading cancer, but overlapping regions make it hard to separate and tell apart independent nuclei. Deep Learning is swiftly paving its way in the arena of nucleus segmentation, attracting quite a few researchers with its numerous published research articles indicating its efficacy in the field. This paper presents a systematic survey on nucleus segmentation using deep learning in the last five years (2017-2021), highlighting various segmentation models (U-Net, SCPP-Net, Sharp U-Net, and LiverNet) and exploring their similarities, strengths, datasets utilized, and unfolding research areas.

CFPNet-M: A light-weight encoder-decoder based network for multimodal biomedical image real-time segmentation

-Deep learning techniques are proving instrumental in identifying, classifying, and quantifying patterns in medical images. Segmentation is one of the important applications in medical image analysis. The U-Net has become the predominant deep-learning approach to medical image segmentation tasks. Existing U-Net based models have limitations in several respects, however, including: the requirement for millions of parameters in the U-Net, which consumes considerable computational resources and memory; the lack of global information; and incomplete segmentation in difficult cases. To remove some of those limitations, we built on our previous work and applied two modifications to improve the U-Net model: 1) we designed and added the dilated channel-wise CNN module and 2) we simplified the U-shape network. We then proposed a novel light-weight architecture, the Channel-wise Feature Pyramid Network for Medicine (CFPNet-M). To evaluate our method, we selected five datasets from different imaging modalities: thermography, electron microscopy, endoscopy, dermoscopy, and digital retinal images. We compared its performance with several models having a variety of complexities. We used the Tanimoto similarity instead of the Jaccard index for gray-level image comparisons. The CFPNet-M achieves segmentation results on all five medical datasets that are comparable to existing methods, yet require only 8.8 MB memory, and just 0.65 million parameters, which is about 2% of U-Net. Unlike other deep-learning segmentation methods, this new approach is suitable for real-time application: its inference speed can reach 80 frames per second when implemented on a single RTX 2070Ti GPU with an input image size of 256 × 192 pixels.

E-DU: Deep neural network for multimodal medical image segmentation based on semantic gap compensation

BACKGROUND

U-Net includes encoder, decoder and skip connection structures. It has become the benchmark network in medical image segmentation. However, the direct fusion of low-level and high-level convolution features with semantic gaps by traditional skip connections may lead to problems such as fuzzy generated feature maps and target region segmentation errors.

OBJECTIVE

We use spatial enhancement filtering technology to compensate for the semantic gap and propose an enhanced dense U-Net (E-DU), aiming to apply it to multimodal medical image segmentation to improve the segmentation performance and efficiency.

METHODS

Before combining encoder and decoder features, we replace the traditional skip connection with a multiscale denoise enhancement (MDE) module. The encoder features need to be deeply convolved by the spatial enhancement filter and then combined with the decoder features. We propose a simple and efficient deep full convolution network structure E-DU, which can not only fuse semantically various features but also denoise and enhance the feature map.

RESULTS

We performed experiments on medical image segmentation datasets with seven image modalities and combined MDE with various baseline networks to perform ablation studies. E-DU achieved the best segmentation results on evaluation indicators such as DSC on the U-Net family, with DSC values of 97.78, 97.64, 95.31, 94.42, 94.93, 98.85, and 98.38 (%), respectively. The addition of the MDE module to the attention mechanism network improves segmentation performance and efficiency, reflecting its generalization performance. In comparison to advanced methods, our method is also competitive.

CONCLUSION

Our proposed MDE module has a good segmentation effect and operating efficiency, and it can be easily extended to multiple modal medical segmentation datasets. Our idea and method can achieve clinical multimodal medical image segmentation and make full use of image information to provide clinical decision support. It has great application value and promotion prospects.

Reconstruction of ovine axonal cytoarchitecture enables more accurate models of brain biomechanics

There is an increased need and focus to understand how local brain microstructure affects the transport of drug molecules directly administered to the brain tissue, for example in convection-enhanced delivery procedures. This study reports a systematic attempt to characterize the cytoarchitecture of commissural, long association and projection fibres, namely the corpus callosum, the fornix and the corona radiata, with the specific aim to map different regions of the tissue and provide essential information for the development of accurate models of brain biomechanics. Ovine samples are imaged using scanning electron microscopy combined with focused ion beam milling to generate 3D volume reconstructions of the tissue at subcellular spatial resolution. Focus is placed on the characteristic cytological feature of the white matter: the axons and their alignment in the tissue. For each tract, a 3D reconstruction of relatively large volumes, including a significant number of axons, is performed and outer axonal ellipticity, outer axonal cross-sectional area and their relative perimeter are measured. The study of well-resolved microstructural features provides useful insight into the fibrous organization of the tissue, whose micromechanical behaviour is that of a composite material presenting elliptical tortuous tubular axonal structures embedded in the extra-cellular matrix. Drug flow can be captured through microstructurally-based models using 3D volumes, either reconstructed directly from images or generated in silico using parameters extracted from the database of images, leading to a workflow to enable physically-accurate simulations of drug delivery to the targeted tissue.

Imaging and reconstruction of sheep brain axonal cytoarchitecture provides insight for brain biomechanics models that simulate drug delivery and other biological processes governed by interstitial fluid flow and transport.

An atlas of the developing Tribolium castaneum brain reveals conservation in anatomy and divergence in timing to Drosophila melanogaster

Insect brains are formed by conserved sets of neural lineages whose fibers form cohesive bundles with characteristic projection patterns. Within the brain neuropil, these bundles establish a system of fascicles constituting the macrocircuitry of the brain. The overall architecture of the neuropils and the macrocircuitry appear to be conserved. However, variation is observed, for example, in size, shape, and timing of development. Unfortunately, the developmental and genetic basis of this variation is poorly understood, although the rise of new genetically tractable model organisms such as the red flour beetle Tribolium castaneum allows the possibility to gain mechanistic insights. To facilitate such work, we present an atlas of the developing brain of T. castaneum, covering the first larval instar, the prepupal stage, and the adult, by combining wholemount immunohistochemical labeling of fiber bundles (acetylated tubulin) and neuropils (synapsin) with digital 3D reconstruction using the TrakEM2 software package. Upon comparing this anatomical dataset with the published work in Drosophila melanogaster, we confirm an overall high degree of conservation. Fiber tracts and neuropil fascicles, which can be visualized by global neuronal antibodies like antiacetylated tubulin in all invertebrate brains, create a rich anatomical framework to which individual neurons or other regions of interest can be referred to. The framework of a largely conserved pattern allowed us to describe differences between the two species with respect to parameters such as timing of neuron proliferation and maturation. These features likely reflect adaptive changes in developmental timing that govern the change from larval to adult brain.

R2U++: a multiscale recurrent residual U-Net with dense skip connections for medical image segmentation

U-Net is a widely adopted neural network in the domain of medical image segmentation. Despite its quick embracement by the medical imaging community, its performance suffers on complicated datasets. The problem can be ascribed to its simple feature extracting blocks: encoder/decoder, and the semantic gap between encoder and decoder. Variants of U-Net (such as R2U-Net) have been proposed to address the problem of simple feature extracting blocks by making the network deeper, but it does not deal with the semantic gap problem. On the other hand, another variant UNET++ deals with the semantic gap problem by introducing dense skip connections but has simple feature extraction blocks. To overcome these issues, we propose a new U-Net based medical image segmentation architecture R2U++. In the proposed architecture, the adapted changes from vanilla U-Net are: (1) the plain convolutional backbone is replaced by a deeper recurrent residual convolution block. The increased field of view with these blocks aids in extracting crucial features for segmentation which is proven by improvement in the overall performance of the network. (2) The semantic gap between encoder and decoder is reduced by dense skip pathways. These pathways accumulate features coming from multiple scales and apply concatenation accordingly. The modified architecture has embedded multi-depth models, and an ensemble of outputs taken from varying depths improves the performance on foreground objects appearing at various scales in the images. The performance of R2U++ is evaluated on four distinct medical imaging modalities: electron microscopy, X-rays, fundus, and computed tomography. The average gain achieved in IoU score is 1.5 ± 0.37% and in dice score is 0.9 ± 0.33% over UNET++, whereas, 4.21 ± 2.72 in IoU and 3.47 ± 1.89 in dice score over R2U-Net across different medical imaging segmentation datasets.

Stimulation of medial amygdala GABA neurons with kinetically different channelrhodopsins yields opposite behavioral outcomes

The medial amygdala (MeA) receives pheromone information about conspecifics and has crucial functions in social behaviors. A previous study showed that activation of GABA neurons in the postero-dorsal MeA (MeApd) with channelrhodopsin-2^H134R (ChR2) stimulates inter-male aggression. When performing these experiments using the faster channelrhodopsin^H134R,E123T (ChETA), we find the opposite behavioral outcome. A systematic comparison between the two channelrhodopsin variants reveals that optogenetic activation of MeApd GABA neurons with ChETA suppresses aggression, whereas activation under ChR2 increases aggression. Although the mechanism for this paradoxical difference is not understood, we observe that activation of MeApd GABA neurons with ChR2 causes larger plateau depolarizations, smaller action potentials, and larger local inhibition than with ChETA. Thus, the channelrhodopsin variant used for in vivo optogenetic experiments can radically influence the behavioral outcome. Future work should continue to study the role of specific sub-populations of MeApd GABA neurons in aggression control.

Smart(Sampling)Augment: Optimal and Efficient Data Augmentation for Semantic Segmentation

Data augmentation methods enrich datasets with augmented data to improve the performance of neural networks. Recently, automated data augmentation methods have emerged, which automatically design augmentation strategies. The existing work focuses on image classification and object detection, whereas we provide the first study on semantic image segmentation and introduce two new approaches: SmartAugment and SmartSamplingAugment. SmartAugment uses Bayesian Optimization to search a rich space of augmentation strategies and achieves new state-of-the-art performance in all semantic segmentation tasks we consider. SmartSamplingAugment, a simple parameter-free approach with a fixed augmentation strategy, competes in performance with the existing resource-intensive approaches and outperforms cheap state-of-the-art data augmentation methods. Furthermore, we analyze the impact, interaction, and importance of data augmentation hyperparameters and perform ablation studies, which confirm our design choices behind SmartAugment and SmartSamplingAugment. Lastly, we will provide our source code for reproducibility and to facilitate further research.

Dense Dilated Deep Multiscale Supervised U-Network for biomedical image segmentation

Biomedical image segmentation is essential for computerized medical image analysis. Deep learning algorithms allow us to design state-of-the-art models for solving segmentation problems. The U-Net and its variants have provided positive results across various datasets. However, the existing networks have the same receptive field at each level and the models are supervised only at the shallow level. Considering these two ideas, we have proposed the D3MSU-Net where the field of view in each level is varied depending upon the depth of the resolution layer and the model is supervised at each resolution level. We have evaluated our network in eight benchmark datasets such as Electron Microscopy, Lung segmentation, Montgomery Chest X-ray, Covid-Radiopaedia, Wound, Medetec, Brain MRI, and Covid-19 lung CT dataset. Additionally, we have provided the performance for various ablations. The experimental results show the superiority of the proposed network. The proposed D3MSU-Net and ablation models are available at www.github.com/shirshabose/D3MSUNET.

A novel approach to assess volumetric bone loss at immediate implant sites and comparison to linear measurements: a pilot study and measurement workflow

OBJECTIVES

Studies have evaluated changes in hard tissue following immediate implant placement (IIP) through cone beam computed tomography (CBCT) imaging, typically examining the amount of two-dimensional (2D) linear, rather than three-dimensional (3D) volumetric bone loss. This study compared the 3D volumetric changes of the alveolar bone at immediate implant sites with 2D linear measurement outcomes by using a novel image analysis workflow.

METHODS

Preoperative and 6-month postoperative CBCT images of patients who underwent IIP and bone grafting in the maxillary esthetic area were acquired. Linear and volumetric measurements of buccal bone dimensions were taken using a specially designed workflow. The 2D and 3D measurements were compared, and their correlations were determined.

RESULTS

Images from 13 patients (13 implants) were analyzed. Linear measurements revealed that the general linear buccal bone loss was less than 1mm in all segments. The 3D volumetric bone reduction (reported as median [first quantile, third quantile]) in the vertical, cervical, middle, and apical segments was 14.27 [11.33, 30.66] mm³ (51.30 [42.78, 66.91]%), 16.20 [10.35, 30.52] mm³ (18.20 [9.88, 24.74]%), 17.48 [8.42, 21.17] mm³ (24.05 [12.39, 28.22]%), and 6.87 [3.88, 9.45] mm³ (11.34 [5.14, 22.54]%), respectively. Significant positive correlations between 2D and 3D measurements were consistently identified in the cervical and middle segments, but no significant correlation was noted in the vertical segment.

CONCLUSIONS

The results revealed that linear measurements could not fully represent volumetric bone dimensional changes. Performing volumetric measurements and 3D rendering could be valuable in presenting the actual amount and topography of peri-implant bone remodeling.

CLINICAL SIGNIFICANCE

Linear measurements only partially represent the real-life event of 3D bone changes at immediate implant sites. Factors affecting hard tissue alterations following IIP should be reassessed using 3D volumetric measurement outcomes.

Blood Vessels: The Pathway Used by Schwann Cells to Colonize Nerve Conduits

The repair of severe nerve injuries requires an autograft or conduit to bridge the gap and avoid axon dispersion. Several conduits are used routinely, but their effectiveness is comparable to that of an autograft only for short gaps. Understanding nerve regeneration within short conduits could help improve their efficacy for longer gaps. Since Schwann cells are known to migrate on endothelial cells to colonize the “nerve bridge”, the new tissue spontaneously forming to connect the injured nerve stumps, here we aimed to investigate whether this migratory mechanism drives Schwann cells to also proceed within the nerve conduits used to repair large nerve gaps. Injured median nerves of adult female rats were repaired with 10 mm chitosan conduits and the regenerated nerves within conduits were analyzed at different time points using confocal imaging of sequential thick sections. Our data showed that the endothelial cells formed a dense capillary network used by Schwann cells to migrate from the two nerve stumps into the conduit. We concluded that angiogenesis played a key role in the nerve conduits, not only by supporting cell survival but also by providing a pathway for the migration of newly formed Schwann cells.

Genetic U-Net: Automatically Designed Deep Networks for Retinal Vessel Segmentation Using a Genetic Algorithm

Recently, many methods based on hand-designed convolutional neural networks (CNNs) have achieved promising results in automatic retinal vessel segmentation. However, these CNNs remain constrained in capturing retinal vessels in complex fundus images. To improve their segmentation performance, these CNNs tend to have many parameters, which may lead to overfitting and high computational complexity. Moreover, the manual design of competitive CNNs is time-consuming and requires extensive empirical knowledge. Herein, a novel automated design method, called Genetic U-Net, is proposed to generate a U-shaped CNN that can achieve better retinal vessel segmentation but with fewer architecture-based parameters, thereby addressing the above issues. First, we devised a condensed but flexible search space based on a U-shaped encoder-decoder. Then, we used an improved genetic algorithm to identify better-performing architectures in the search space and investigated the possibility of finding a superior network architecture with fewer parameters. The experimental results show that the architecture obtained using the proposed method offered a superior performance with less than 1% of the number of the original U-Net parameters in particular and with significantly fewer parameters than other state-of-the-art models. Furthermore, through in-depth investigation of the experimental results, several effective operations and patterns of networks to generate superior retinal vessel segmentations were identified. The codes of this work are available at https://github.com/96jhwei/Genetic-U-Net.

Sub-micro scale cell segmentation using deep learning

Automated cell segmentation is key for rapid and accurate investigation of cell responses. As instrumentation resolving power increases, clear delineation of newly revealed cellular features at the sub-micron through nanoscale becomes important. Reliance on manual investigation of myriad small features retards investigation; however, use of deep learning methods has great potential to reveal cell features both at high accuracy and high speed, which may lead to new discoveries in the near term. In this study, semantic cell segmentation systems were investigated by implementing fully convolutional neural networks called U-nets for segmentation of astrocytes cultured on Poly-L-lysine-functionalized planar glass. The network hyperparameters were determined by changing the number of network layers, loss functions, and input image modalities. Atomic force microscopy (AFM) images were selected for investigation as these are inherently nanoscale and are also dimensional. AFM height, deflection, and friction images were used as inputs separately and together, and the segmentation performances were investigated on five-fold cross-validation data. Transfer learning methods, including VGG16, VGG19, and Xception, were used to improve cell segmentation performance. We find that AFM height images inherit more discriminative features than AFM deflection and AFM friction images for cell segmentation. When transfer learning methods are applied, statistically significant segmentation performance improvements are observed. Segmentation performance was compared to classical image processing algorithms and other algorithms in use by considering both AFM and electron microscopy segmentation. An accuracy of 0.9849, Matthews correlation coefficient of 0.9218, and Dice's similarity coefficient of 0.9306 were obtained on the AFM test images. Performance evaluations show that the proposed system can be successfully used for AFM cell segmentation with high precision. This article is protected by copyright. All rights reserved.

MTNeuro: A Benchmark for Evaluating Representations of Brain Structure Across Multiple Levels of Abstraction

There are multiple scales of abstraction from which we can describe the same image, depending on whether we are focusing on fine-grained details or a more global attribute of the image. In brain mapping, learning to automatically parse images to build representations of both small-scale features (e.g., the presence of cells or blood vessels) and global properties of an image (e.g., which brain region the image comes from) is a crucial and open challenge. However, most existing datasets and benchmarks for neuroanatomy consider only a single downstream task at a time. To bridge this gap, we introduce a new dataset, annotations, and multiple downstream tasks that provide diverse ways to readout information about brain structure and architecture from the same image. Our multi-task neuroimaging benchmark (MTNeuro) is built on volumetric, micrometer-resolution X-ray microtomography images spanning a large thalamocortical section of mouse brain, encompassing multiple cortical and subcortical regions. We generated a number of different prediction challenges and evaluated several supervised and self-supervised models for brain-region prediction and pixel-level semantic segmentation of microstructures. Our experiments not only highlight the rich heterogeneity of this dataset, but also provide insights into how self-supervised approaches can be used to learn representations that capture multiple attributes of a single image and perform well on a variety of downstream tasks. Datasets, code, and pre-trained baseline models are provided at: https://mtneuro.github.io/.

The Active Segmentation Platform for Microscopic Image Classification and Segmentation

Image segmentation still represents an active area of research since no universal solution can be identified. Traditional image segmentation algorithms are problem-specific and limited in scope. On the other hand, machine learning offers an alternative paradigm where predefined features are combined into different classifiers, providing pixel-level classification and segmentation. However, machine learning only can not address the question as to which features are appropriate for a certain classification problem. The article presents an automated image segmentation and classification platform, called Active Segmentation, which is based on ImageJ. The platform integrates expert domain knowledge, providing partial ground truth, with geometrical feature extraction based on multi-scale signal processing combined with machine learning. The approach in image segmentation is exemplified on the ISBI 2012 image segmentation challenge data set. As a second application we demonstrate whole image classification functionality based on the same principles. The approach is exemplified using the HeLa and HEp-2 data sets. Obtained results indicate that feature space enrichment properly balanced with feature selection functionality can achieve performance comparable to deep learning approaches. In summary, differential geometry can substantially improve the outcome of machine learning since it can enrich the underlying feature space with new geometrical invariant objects.

MHSU-Net: A more versatile neural network for medical image segmentation

BACKGROUND AND OBJECTIVE

Medical image segmentation plays an important role in clinic. Recently, with the development of deep learning, many convolutional neural network (CNN)-based medical image segmentation algorithms have been proposed. Among them, U-Net is one of the most famous networks. However, the standard convolutional layers used by U-Net limit its capability to capture abundant features. Additionally, the consecutive maximum pooling operations in U-Net cause certain features to be lost. This paper aims to improve the feature extraction capability of U-Net and reduce the feature loss during the segmentation process. Meanwhile, the paper also focuses on improving the versatility of the proposed segmentation model.

METHODS

Firstly, in order to enable the model to capture richer features, we have proposed a novel multiscale convolutional block (MCB). MCB adopts a wider and deeper structure, which can be applied to different types of segmentation tasks. Secondly, a hybrid down-sampling block (HDSB) has been proposed to reduce the feature loss via replacing the maximum pooling layer. Thirdly, we have proposed a context module (CIF) based on atrous convolution and SKNet to extract sufficient context information. Finally, we combined the CIF module with Skip Connection of U-Net, and further proposed the Skip Connection+ structure.

RESULTS

We name the proposed network MHSU-Net. MHSU-Net has been evaluated on three different datasets, including lung, cell contour, and pancreas. Experimental results demonstrate that MHSU-Net outperforms U-Net and other state-of-the-art models under various evaluation metrics, and owns greater potential in clinical applications.

CONCLUSIONS

The proposed modules can greatly improve the feature extraction capability of the segmentation model and effectively reduce the feature loss during the segmentation process. MHSU-Net can also be applied to different types of medical image segmentation tasks.

Synapses from the Motor Cortex and a High-Order Thalamic Nucleus are Spatially Clustered in Proximity to Each Other in the Distal Tuft Dendrites of Mouse Somatosensory Cortex

The posterior medial nucleus of the thalamus (POm) and vibrissal primary motor cortex (vM1) convey essential information to the barrel cortex (S1BF) regarding whisker position and movement. Therefore, understanding the relative spatial relationship of these two inputs is a critical prerequisite for acquiring insights into how S1BF synthesizes information to interpret the location of an object. Using array tomography, we identified the locations of synapses from vM1 and POm on distal tuft dendrites of L5 pyramidal neurons where the two inputs are combined. Synapses from vM1 and POm did not show a significant branchlet preference and impinged on the same set of dendritic branchlets. Within dendritic branches, on the other hand, the two inputs formed robust spatial clusters of their own type. Furthermore, we also observed POm clusters in proximity to vM1 clusters. This work constitutes the first detailed description of the relative distribution of synapses from POm and vM1, which is crucial to elucidate the synaptic integration of whisker-based sensory information.

Sharp U-Net: Depthwise convolutional network for biomedical image segmentation

The U-Net architecture, built upon the fully convolutional network, has proven to be effective in biomedical image segmentation. However, U-Net applies skip connections to merge semantically different low- and high-level convolutional features, resulting in not only blurred feature maps, but also over- and under-segmented target regions. To address these limitations, we propose a simple, yet effective end-to-end depthwise encoder-decoder fully convolutional network architecture, called Sharp U-Net, for binary and multi-class biomedical image segmentation. The key rationale of Sharp U-Net is that instead of applying a plain skip connection, a depthwise convolution of the encoder feature map with a sharpening kernel filter is employed prior to merging the encoder and decoder features, thereby producing a sharpened intermediate feature map of the same size as the encoder map. Using this sharpening filter layer, we are able to not only fuse semantically less dissimilar features, but also to smooth out artifacts throughout the network layers during the early stages of training. Our extensive experiments on six datasets show that the proposed Sharp U-Net model consistently outperforms or matches the recent state-of-the-art baselines in both binary and multi-class segmentation tasks, while adding no extra learnable parameters. Furthermore, Sharp U-Net outperforms baselines that have more than three times the number of learnable parameters.

Multi-level dilated residual network for biomedical image segmentation

We propose a novel multi-level dilated residual neural network, an extension of the classical U-Net architecture, for biomedical image segmentation. U-Net is the most popular deep neural architecture for biomedical image segmentation, however, despite being state-of-the-art, the model has a few limitations. In this study, we suggest replacing convolutional blocks of the classical U-Net with multi-level dilated residual blocks, resulting in enhanced learning capability. We also propose to incorporate a non-linear multi-level residual blocks into skip connections to reduce the semantic gap and to restore the information lost when concatenating features from encoder to decoder units. We evaluate the proposed approach on five publicly available biomedical datasets with different imaging modalities, including electron microscopy, magnetic resonance imaging, histopathology, and dermoscopy, each with its own segmentation challenges. The proposed approach consistently outperforms the classical U-Net by 2%, 3%, 6%, 8%, and 14% relative improvements in dice coefficient, respectively for magnetic resonance imaging, dermoscopy, histopathology, cell nuclei microscopy, and electron microscopy modalities. The visual assessments of the segmentation results further show that the proposed approach is robust against outliers and preserves better continuity in boundaries compared to the classical U-Net and its variant, MultiResUNet.

HyAdamC: A New Adam-Based Hybrid Optimization Algorithm for Convolution Neural Networks

As the performance of devices that conduct large-scale computations has been rapidly improved, various deep learning models have been successfully utilized in various applications. Particularly, convolution neural networks (CNN) have shown remarkable performance in image processing tasks such as image classification and segmentation. Accordingly, more stable and robust optimization methods are required to effectively train them. However, the traditional optimizers used in deep learning still have unsatisfactory training performance for the models with many layers and weights. Accordingly, in this paper, we propose a new Adam-based hybrid optimization method called HyAdamC for training CNNs effectively. HyAdamC uses three new velocity control functions to adjust its search strength carefully in term of initial, short, and long-term velocities. Moreover, HyAdamC utilizes an adaptive coefficient computation method to prevent that a search direction determined by the first momentum is distorted by any outlier gradients. Then, these are combined into one hybrid method. In our experiments, HyAdamC showed not only notable test accuracies but also significantly stable and robust optimization abilities when training various CNN models. Furthermore, we also found that HyAdamC could be applied into not only image classification and image segmentation tasks.

FusionNet: A Deep Fully Residual Convolutional Neural Network for Image Segmentation in Connectomics

Cellular-resolution connectomics is an ambitious research direction with the goal of generating comprehensive brain connectivity maps using high-throughput, nano-scale electron microscopy. One of the main challenges in connectomics research is developing scalable image analysis algorithms that require minimal user intervention. Deep learning has provided exceptional performance in image classification tasks in computer vision, leading to a recent explosion in popularity. Similarly, its application to connectomic analyses holds great promise. Here, we introduce a deep neural network architecture, FusionNet, with a focus on its application to accomplish automatic segmentation of neuronal structures in connectomics data. FusionNet combines recent advances in machine learning, such as semantic segmentation and residual neural networks, with summation-based skip connections. This results in a much deeper network architecture and improves segmentation accuracy. We demonstrate the performance of the proposed method by comparing it with several other popular electron microscopy segmentation methods. We further illustrate its flexibility through segmentation results for two different tasks: cell membrane segmentation and cell nucleus segmentation.

Antennal-lobe neurons in the moth Helicoverpa armigera: Morphological features of projection neurons, local interneurons, and centrifugal neurons

The relatively large primary olfactory center of the insect brain, the antennal lobe (AL), contains several heterogeneous neuronal types. These include projection neurons (PNs), providing olfactory information to higher‐order neuropils via parallel pathways, and local interneurons (LNs), which provide lateral processing within the AL. In addition, various types of centrifugal neurons (CNs) offer top‐down modulation onto the other AL neurons. By performing iontophoretic intracellular staining, we collected a large number of AL neurons in the moth, Helicoverpa armigera, to examine the distinct morphological features of PNs, LNs, and CNs. We characterize 190 AL neurons. These were allocated to 25 distinct neuronal types or sub‐types, which were reconstructed and placed into a reference brain. In addition to six PN types comprising 15 sub‐types, three LN and seven CN types were identified. High‐resolution confocal images allowed us to analyze AL innervations of the various reported neurons, which demonstrated that all PNs innervating ventroposterior glomeruli contact a protocerebral neuropil rarely targeted by other PNs, that is the posteriorlateral protocerebrum. We also discuss the functional roles of the distinct CNs, which included several previously uncharacterized types, likely involved in computations spanning from multisensory processing to olfactory feedback signalization into the AL.

Democratising deep learning for microscopy with ZeroCostDL4Mic

Deep Learning (DL) methods are powerful analytical tools for microscopy and can outperform conventional image processing pipelines. Despite the enthusiasm and innovations fuelled by DL technology, the need to access powerful and compatible resources to train DL networks leads to an accessibility barrier that novice users often find difficult to overcome. Here, we present ZeroCostDL4Mic, an entry-level platform simplifying DL access by leveraging the free, cloud-based computational resources of Google Colab. ZeroCostDL4Mic allows researchers with no coding expertise to train and apply key DL networks to perform tasks including segmentation (using U-Net and StarDist), object detection (using YOLOv2), denoising (using CARE and Noise2Void), super-resolution microscopy (using Deep-STORM), and image-to-image translation (using Label-free prediction - fnet, pix2pix and CycleGAN). Importantly, we provide suitable quantitative tools for each network to evaluate model performance, allowing model optimisation. We demonstrate the application of the platform to study multiple biological processes.

DeepMIB: User-friendly and open-source software for training of deep learning network for biological image segmentation

We present DeepMIB, a new software package that is capable of training convolutional neural networks for segmentation of multidimensional microscopy datasets on any workstation. We demonstrate its successful application for segmentation of 2D and 3D electron and multicolor light microscopy datasets with isotropic and anisotropic voxels. We distribute DeepMIB as both an open-source multi-platform Matlab code and as compiled standalone application for Windows, MacOS and Linux. It comes in a single package that is simple to install and use as it does not require knowledge of programming. DeepMIB is suitable for everyone interested of bringing a power of deep learning into own image segmentation workflows.

MSU-Net: Multi-Scale U-Net for 2D Medical Image Segmentation

Aiming at the limitation of the convolution kernel with a fixed receptive field and unknown prior to optimal network width in U-Net, multi-scale U-Net (MSU-Net) is proposed by us for medical image segmentation. First, multiple convolution sequence is used to extract more semantic features from the images. Second, the convolution kernel with different receptive fields is used to make features more diverse. The problem of unknown network width is alleviated by efficient integration of convolution kernel with different receptive fields. In addition, the multi-scale block is extended to other variants of the original U-Net to verify its universality. Five different medical image segmentation datasets are used to evaluate MSU-Net. A variety of imaging modalities are included in these datasets, such as electron microscopy, dermoscope, ultrasound, etc. Intersection over Union (IoU) of MSU-Net on each dataset are 0.771, 0.867, 0.708, 0.900, and 0.702, respectively. Experimental results show that MSU-Net achieves the best performance on different datasets. Our implementation is available at https://github.com/CN-zdy/MSU_Net.

Reconstruction of motor control circuits in adult Drosophila using automated transmission electron microscopy

To investigate circuit mechanisms underlying locomotor behavior, we used serial-section electron microscopy (EM) to acquire a synapse-resolution dataset containing the ventral nerve cord (VNC) of an adult female Drosophila melanogaster. To generate this dataset, we developed GridTape, a technology that combines automated serial-section collection with automated high-throughput transmission EM. Using this dataset, we studied neuronal networks that control leg and wing movements by reconstructing all 507 motor neurons that control the limbs. We show that a specific class of leg sensory neurons synapses directly onto motor neurons with the largest-caliber axons on both sides of the body, representing a unique pathway for fast limb control. We provide open access to the dataset and reconstructions registered to a standard atlas to permit matching of cells between EM and light microscopy data. We also provide GridTape instrumentation designs and software to make large-scale EM more accessible and affordable to the scientific community.

ADID-UNET-a segmentation model for COVID-19 infection from lung CT scans

Currently, the new coronavirus disease (COVID-19) is one of the biggest health crises threatening the world. Automatic detection from computed tomography (CT) scans is a classic method to detect lung infection, but it faces problems such as high variations in intensity, indistinct edges near lung infected region and noise due to data acquisition process. Therefore, this article proposes a new COVID-19 pulmonary infection segmentation depth network referred as the Attention Gate-Dense Network- Improved Dilation Convolution-UNET (ADID-UNET). The dense network replaces convolution and maximum pooling function to enhance feature propagation and solves gradient disappearance problem. An improved dilation convolution is used to increase the receptive field of the encoder output to further obtain more edge features from the small infected regions. The integration of attention gate into the model suppresses the background and improves prediction accuracy. The experimental results show that the ADID-UNET model can accurately segment COVID-19 lung infected areas, with performance measures greater than 80% for metrics like Accuracy, Specificity and Dice Coefficient (DC). Further when compared to other state-of-the-art architectures, the proposed model showed excellent segmentation effects with a high DC and F1 score of 0.8031 and 0.82 respectively.

Anatomy and Neural Pathways Modulating Distinct Locomotor Behaviors in Drosophila Larva

The control of movements is a fundamental feature shared by all animals. At the most basic level, simple movements are generated by coordinated neural activity and muscle contraction patterns that are controlled by the central nervous system. How behavioral responses to various sensory inputs are processed and integrated by the downstream neural network to produce flexible and adaptive behaviors remains an intense area of investigation in many laboratories. Due to recent advances in experimental techniques, many fundamental neural pathways underlying animal movements have now been elucidated. For example, while the role of motor neurons in locomotion has been studied in great detail, the roles of interneurons in animal movements in both basic and noxious environments have only recently been realized. However, the genetic and transmitter identities of many of these interneurons remains unclear. In this review, we provide an overview of the underlying circuitry and neural pathways required by Drosophila larvae to produce successful movements. By improving our understanding of locomotor circuitry in model systems such as Drosophila, we will have a better understanding of how neural circuits in organisms with different bodies and brains lead to distinct locomotion types at the organism level. The understanding of genetic and physiological components of these movements types also provides directions to understand movements in higher organisms.

TMD-Unet: Triple-Unet with Multi-Scale Input Features and Dense Skip Connection for Medical Image Segmentation

Deep learning is one of the most effective approaches to medical image processing applications. Network models are being studied more and more for medical image segmentation challenges. The encoder-decoder structure is achieving great success, in particular the Unet architecture, which is used as a baseline architecture for the medical image segmentation networks. Traditional Unet and Unet-based networks still have a limitation that is not able to fully exploit the output features of the convolutional units in the node. In this study, we proposed a new network model named TMD-Unet, which had three main enhancements in comparison with Unet: (1) modifying the interconnection of the network node, (2) using dilated convolution instead of the standard convolution, and (3) integrating the multi-scale input features on the input side of the model and applying a dense skip connection instead of a regular skip connection. Our experiments were performed on seven datasets, including many different medical image modalities such as colonoscopy, electron microscopy (EM), dermoscopy, computed tomography (CT), and magnetic resonance imaging (MRI). The segmentation applications implemented in the paper include EM, nuclei, polyp, skin lesion, left atrium, spleen, and liver segmentation. The dice score of our proposed models achieved 96.43% for liver segmentation, 95.51% for spleen segmentation, 92.65% for polyp segmentation, 94.11% for EM segmentation, 92.49% for nuclei segmentation, 91.81% for left atrium segmentation, and 87.27% for skin lesion segmentation. The experimental results showed that the proposed model was superior to the popular models for all seven applications, which demonstrates the high generality of the proposed model.

Dynamics in protein translation sustaining T cell preparedness

In response to pathogenic threats, naive T cells rapidly transition from a quiescent to an activated state, yet the underlying mechanisms are incompletely understood. Using a pulsed SILAC approach, we investigated the dynamics of mRNA translation kinetics and protein turnover in human naive and activated T cells. Our datasets uncovered that transcription factors maintaining T cell quiescence had constitutively high turnover, which facilitated their depletion following activation. Furthermore, naive T cells maintained a surprisingly large number of idling ribosomes as well as 242 repressed mRNA species and a reservoir of glycolytic enzymes. These components were rapidly engaged following stimulation, promoting an immediate translational and glycolytic switch to ramp up the T cell activation program. Our data elucidate new insights into how T cells maintain a prepared state to mount a rapid immune response, and provide a resource of protein turnover, absolute translation kinetics and protein synthesis rates in T cells ( https://www.immunomics.ch ).

UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation

The state-of-the-art models for medical image segmentation are variants of U-Net and fully convolutional networks (FCN). Despite their success, these models have two limitations: (1) their optimal depth is apriori unknown, requiring extensive architecture search or inefficient ensemble of models of varying depths; and (2) their skip connections impose an unnecessarily restrictive fusion scheme, forcing aggregation only at the same-scale feature maps of the encoder and decoder sub-networks. To overcome these two limitations, we propose UNet++, a new neural architecture for semantic and instance segmentation, by (1) alleviating the unknown network depth with an efficient ensemble of U-Nets of varying depths, which partially share an encoder and co-learn simultaneously using deep supervision; (2) redesigning skip connections to aggregate features of varying semantic scales at the decoder sub-networks, leading to a highly flexible feature fusion scheme; and (3) devising a pruning scheme to accelerate the inference speed of UNet++. We have evaluated UNet++ using six different medical image segmentation datasets, covering multiple imaging modalities such as computed tomography (CT), magnetic resonance imaging (MRI), and electron microscopy (EM), and demonstrating that (1) UNet++ consistently outperforms the baseline models for the task of semantic segmentation across different datasets and backbone architectures; (2) UNet++ enhances segmentation quality of varying-size objects-an improvement over the fixed-depth U-Net; (3) Mask RCNN++ (Mask R-CNN with UNet++ design) outperforms the original Mask R-CNN for the task of instance segmentation; and (4) pruned UNet++ models achieve significant speedup while showing only modest performance degradation. Our implementation and pre-trained models are available at https://github.com/MrGiovanni/UNetPlusPlus.

Neuroarchitecture of the central complex in the brain of the honeybee: Neuronal cell types

The central complex (CX) in the insect brain is a higher order integration center that controls a number of behaviors, most prominently goal directed locomotion. The CX comprises the protocerebral bridge (PB), the upper division of the central body (CBU), the lower division of the central body (CBL), and the paired noduli (NO). Although spatial orientation has been extensively studied in honeybees at the behavioral level, most electrophysiological and anatomical analyses have been carried out in other insect species, leaving the morphology and physiology of neurons that constitute the CX in the honeybee mostly enigmatic. The goal of this study was to morphologically identify neuronal cell types of the CX in the honeybee Apis mellifera. By performing iontophoretic dye injections into the CX, we traced 16 subtypes of neuron that connect a subdivision of the CX with other regions in the bee's central brain, and eight subtypes that mainly interconnect different subdivisions of the CX. They establish extensive connections between the CX and the lateral complex, the superior protocerebrum and the posterior protocerebrum. Characterized neuron classes and subtypes are morphologically similar to those described in other insects, suggesting considerable conservation in the neural network relevant for orientation.

SANDI: A compartment-based model for non-invasive apparent soma and neurite imaging by diffusion MRI

This work introduces a compartment-based model for apparent cell body (namely soma) and neurite density imaging (SANDI) using non-invasive diffusion-weighted MRI (DW-MRI). The existing conjecture in brain microstructure imaging through DW-MRI presents water diffusion in white (WM) and gray (GM) matter as restricted diffusion in neurites, modelled by infinite cylinders of null radius embedded in the hindered extra-neurite water. The extra-neurite pool in WM corresponds to water in the extra-axonal space, but in GM it combines water in the extra-cellular space with water in soma. While several studies showed that this microstructure model successfully describe DW-MRI data in WM and GM at b ​≤ ​3,000 ​s/mm² (or 3 ​ms/μm²), it has been also shown to fail in GM at high b values (b≫3,000 ​s/mm² or 3 ​ms/μm²). Here we hypothesise that the unmodelled soma compartment (i.e. cell body of any brain cell type: from neuroglia to neurons) may be responsible for this failure and propose SANDI as a new model of brain microstructure where soma of any brain cell type is explicitly included. We assess the effects of size and density of soma on the direction-averaged DW-MRI signal at high b values and the regime of validity of the model using numerical simulations and comparison with experimental data from mouse (b_max ​= ​40,000 ​s/mm², or 40 ​ms/μm²) and human (b_max ​= ​10,000 ​s/mm², or 10 ​ms/μm²) brain. We show that SANDI defines new contrasts representing complementary information on the brain cyto- and myelo-architecture. Indeed, we show maps from 25 healthy human subjects of MR soma and neurite signal fractions, that remarkably mirror contrasts of histological images of brain cyto- and myelo-architecture. Although still under validation, SANDI might provide new insight into tissue architecture by introducing a new set of biomarkers of potential great value for biomedical applications and pure neuroscience.