Drosophila Larvae-Inspired Soft Crawling Robot with Multimodal Locomotion and Versatile Applications.

Soft crawling robots have been widely studied and applied because of their excellent environmental adaptability and flexible movement. However, most existing soft crawling robots typically exhibit a single-motion mode and lack diverse capabilities. Inspired by Drosophila larvae, this paper proposes a compact soft crawling robot (weight, 13 g; length, 165 mm; diameter, 35 mm) with multimodal locomotion (forward, turning, rolling, and twisting). Each robot module uses 4 sets of high-power-density shape memory alloy actuators, endowing it with 4 degrees of motion freedom. We analyze the mechanical characteristics of the robot modules through experiments and simulation analysis. The plug-and-play modules can be quickly assembled to meet different motion and task requirements. The soft crawling robot can be remotely operated with an external controller, showcasing multimodal motion on various material surfaces. In a narrow maze, the robot demonstrates agile movement and effective maneuvering around obstacles. In addition, leveraging the inherent bistable characteristics of the robot modules, we used the robot modules as anchoring units and installed a microcamera on the robot's head for pipeline detection. The robot completed the inspection in horizontal, vertical, curved, and branched pipelines, adjusted the camera view, and twisted a valve in the pipeline for the first time. Our research highlights the robot's superior locomotion and application capabilities, providing an innovative strategy for the development of lightweight, compact, and multifunctional soft crawling robots.

High-Speed Automated Reconstruction of Drosophila Larval Brain from Volumetric EM Data.

To uncover the relationship between neural activity and behavior, it is essential to reconstruct neural circuits. However, methods typically used for neuron reconstruction from volumetric electron microscopy (EM) dataset are often time-consuming and require extensive manual proofreading, making it difficult to reproduce in a typical laboratory setting. To address this challenge, we have developed a set of acceleration techniques that build upon the Flood Filling Network (FFN), significantly reducing the time required for this task. These techniques can be easily adapted to other similar datasets and laboratory settings. To validate our approach, we tested our pipeline on a dataset of Drosophila larval brain serial section EM images at synaptic-resolution level. Our results demonstrate that our pipeline significantly reduces the inference time compared to the FFN baseline method and greatly reduces the time required for reconstructing the 3D morphology of neurons.

Discriminative Radial Domain Adaptation.

Domain adaptation methods reduce domain shift typically by learning domain-invariant features. Most existing methods are built on distribution matching, e.g., adversarial domain adaptation, which tends to corrupt feature discriminability. In this paper, we propose Discriminative Radial Domain Adaptation (DRDR) which bridges source and target domains via a shared radial structure. It's motivated by the observation that as the model is trained to be progressively discriminative, features of different categories expand outwards in different directions, forming a radial structure. We show that transferring such an inherently discriminative structure would enable to enhance feature transferability and discriminability simultaneously. Specifically, we represent each domain with a global anchor and each category a local anchor to form a radial structure and reduce domain shift via structure matching. It consists of two parts, namely isometric transformation to align the structure globally and local refinement to match each category. To enhance the discriminability of the structure, we further encourage samples to cluster close to the corresponding local anchors based on optimal-transport assignment. Extensively experimenting on multiple benchmarks, our method is shown to consistently outperforms state-of-the-art approaches on varied tasks, including the typical unsupervised domain adaptation, multi-source domain adaptation, domain-agnostic learning, and domain generalization.

Scalable Electrophysiology of Millimeter-Scale Animals with Electrode Devices.

Millimeter-scale animals such as Caenorhabditis elegans, Drosophila larvae, zebrafish, and bees serve as powerful model organisms in the fields of neurobiology and neuroethology. Various methods exist for recording large-scale electrophysiological signals from these animals. Existing approaches often lack, however, real-time, uninterrupted investigations due to their rigid constructs, geometric constraints, and mechanical mismatch in integration with soft organisms. The recent research establishes the foundations for 3-dimensional flexible bioelectronic interfaces that incorporate microfabricated components and nanoelectronic function with adjustable mechanical properties and multidimensional variability, offering unique capabilities for chronic, stable interrogation and stimulation of millimeter-scale animals and miniature tissue constructs. This review summarizes the most advanced technologies for electrophysiological studies, based on methods of 3-dimensional flexible bioelectronics. A concluding section addresses the challenges of these devices in achieving freestanding, robust, and multifunctional biointerfaces.

Parameter-Efficient Person Re-Identification in the 3D Space.

People live in a 3D world. However, existing works on person re-identification (re-id) mostly consider the semantic representation learning in a 2D space, intrinsically limiting the understanding of people. In this work, we address this limitation by exploring the prior knowledge of the 3D body structure. Specifically, we project 2D images to a 3D space and introduce a novel parameter-efficient omni-scale graph network (OG-Net) to learn the pedestrian representation directly from 3D point clouds. OG-Net effectively exploits the local information provided by sparse 3D points and takes advantage of the structure and appearance information in a coherent manner. With the help of 3D geometry information, we can learn a new type of deep re-id feature free from noisy variants, such as scale and viewpoint. To our knowledge, we are among the first attempts to conduct person re-id in the 3D space. We demonstrate through extensive experiments that the proposed method: (1) eases the matching difficulty in the traditional 2D space; 2) exploits the complementary information of 2D appearance and 3D structure; 3) achieves competitive results with limited parameters on four large-scale person re-id datasets; and 4) has good scalability to unseen datasets. Our code, models, and generated 3D human data are publicly available at https://github.com/layumi/person-reid-3d.

An Ultra-Low-Noise, Low Power and Miniaturized Dual-Channel Wireless Neural Recording Microsystem.

As the basic tools for neuroscience research, invasive neural recording devices can obtain high-resolution neuronal activity signals through electrodes connected to the subject's brain. Existing wireless neural recording devices are large in size or need external large-scale equipment for wireless power supply, which limits their application. Here, we developed an ultra-low-noise, low power and miniaturized dual-channel wireless neural recording microsystem. With the full-differential front-end structure of the dual operational amplifiers (op-amps), the noise level and power consumption are notably reduced. The hierarchical microassembly technology, which integrates wafer-level packaged op-amps and the miniaturized Bluetooth module, dramatically reduces the size of the wireless neural recording microsystem. The microsystem shows a less than 100 nV/Hz ultra-low noise level, about 10 mW low power consumption, and 9 × 7 × 5 mm3 small size. The neural recording ability was then demonstrated in saline and a chronic rat model. Because of its miniaturization, it can be applied to freely behaving small animals, such as rats. Its features of ultra-low noise and high bandwidth are conducive to low-amplitude neural signal recording, which may help advance neuroscientific discovery.

Using Simulated Training Data of Voxel-Level Generative Models to Improve 3D Neuron Reconstruction.

Reconstructing neuron morphologies from fluorescence microscope images plays a critical role in neuroscience studies. It relies on image segmentation to produce initial masks either for further processing or final results to represent neuronal morphologies. This has been a challenging step due to the variation and complexity of noisy intensity patterns in neuron images acquired from microscopes. Whereas progresses in deep learning have brought the goal of accurate segmentation much closer to reality, creating training data for producing powerful neural networks is often laborious. To overcome the difficulty of obtaining a vast number of annotated data, we propose a novel strategy of using two-stage generative models to simulate training data with voxel-level labels. Trained upon unlabeled data by optimizing a novel objective function of preserving predefined labels, the models are able to synthesize realistic 3D images with underlying voxel labels. We showed that these synthetic images could train segmentation networks to obtain even better performance than manually labeled data. To demonstrate an immediate impact of our work, we further showed that segmentation results produced by networks trained upon synthetic data could be used to improve existing neuron reconstruction methods.

Generative Adversarial Networks in Medical Image augmentation: A review.

OBJECT: With the development of deep learning, the number of training samples for medical image-based diagnosis and treatment models is increasing. Generative Adversarial Networks (GANs) have attracted attention in medical image processing due to their excellent image generation capabilities and have been widely used in data augmentation. In this paper, a comprehensive and systematic review and analysis of medical image augmentation work are carried out, and its research status and development prospects are reviewed. METHOD: This paper reviews 105 medical image augmentation related papers, which mainly collected by ELSEVIER, IEEE Xplore, and Springer from 2018 to 2021. We counted these papers according to the parts of the organs corresponding to the images, and sorted out the medical image datasets that appeared in them, the loss function in model training, and the quantitative evaluation metrics of image augmentation. At the same time, we briefly introduce the literature collected in three journals and three conferences that have received attention in medical image processing. RESULT: First, we summarize the advantages of various augmentation models, loss functions, and evaluation metrics. Researchers can use this information as a reference when designing augmentation tasks. Second, we explore the relationship between augmented models and the amount of the training set, and tease out the role that augmented models may play when the quality of the training set is limited. Third, the statistical number of papers shows that the development momentum of this research field remains strong. Furthermore, we discuss the existing limitations of this type of model and suggest possible research directions. CONCLUSION: We discuss GAN-based medical image augmentation work in detail. This method effectively alleviates the challenge of limited training samples for medical image diagnosis and treatment models. It is hoped that this review will benefit researchers interested in this field.

HTsort: Enabling Fast and Accurate Spike Sorting on Multi-Electrode Arrays.

Spike sorting is used to classify the spikes (action potentials acquired by physiological electrodes), aiming to identify their respective firing units. Now it has been developed to classify the spikes recorded by multi-electrode arrays (MEAs), with the improvement of micro-electrode technology. However, how to improve classification accuracy and maintain low time complexity simultaneously becomes a difficulty. A fast and accurate spike sorting approach named HTsort is proposed for high-density multi-electrode arrays in this paper. Several improvements have been introduced to the traditional pipeline that is composed of threshold detection and clustering method. First, the divide-and-conquer method is employed to utilize electrode spatial information to achieve pre-clustering. Second, the clustering method HDBSCAN (hierarchical density-based spatial clustering of applications with noise) is used to classify spikes and detect overlapping events (multiple spikes firing simultaneously). Third, the template merging method is used to merge redundant exported templates according to the template similarity and the spatial distribution of electrodes. Finally, the template matching method is used to resolve overlapping events. Our approach is validated on simulation data constructed by ourselves and publicly available data and compared to other state-of-the-art spike sorters. We found that the proposed HTsort has a more favorable trade-off between accuracy and time consumption. Compared with MountainSort and SpykingCircus, the time consumption is reduced by at least 40% when the number of electrodes is 64 and below. Compared with HerdingSpikes, the classification accuracy can typically improve by more than 10%. Meanwhile, HTsort exhibits stronger robustness against background noise than other sorters. Our more sophisticated spike sorter would facilitate neurophysiologists to complete spike sorting more quickly and accurately.

An ultrasound image-based deep multi-scale texture network for liver fibrosis grading in patients with chronic HBV infection.

BACKGROUND & AIMS: The evaluation of the stage of liver fibrosis is essential in patients with chronic liver disease. However, due to the low quality of ultrasound images, the non-invasive diagnosis of liver fibrosis based on ultrasound images is still an outstanding question. This study aimed to investigate the diagnostic accuracy of a deep learning-based method in ultrasound images for liver fibrosis staging in multicentre patients. METHODS: In this study, we proposed a novel deep learning-based approach, named multi-scale texture network (MSTNet), to assess liver fibrosis, which extracted multi-scale texture features from constructed image pyramid patches. Its diagnostic accuracy was investigated by comparing it with APRI, FIB-4, Forns and sonographers. Data of 508 patients who underwent liver biopsy were included from 4 hospitals. The area-under-the ROC curve (AUC) was determined by receiver operating characteristics (ROC) curves for significant fibrosis (≥F2) and cirrhosis (F4). RESULTS: The AUCs (95% confidence interval) of MSTNet were 0.92 (0.87-0.96) for ≥F2 and 0.89 (0.83-0.95) for F4 on the validation group, which significantly outperformed APRI, FIB-4 and Forns. The sensitivity and specificity of MSTNet (85.1% (74.5%-92.0%) and 87.6% (78.0%-93.6%)) were better than those of three sonographers in assessing ≥F2. CONCLUSIONS: The proposed MSTNet is a promising ultrasound image-based method for the non-invasive grading of liver fibrosis in patients with chronic HBV infection.

Spatially-Aware Context Neural Networks.

A variety of computer vision tasks benefit significantly from increasingly powerful deep convolutional neural networks. However, the inherently local property of convolution operations prevents most existing models from capturing long-range feature interactions for improved performances. In this paper, we propose a novel module, called Spatially-Aware Context (SAC) block, to learn spatially-aware contexts by capturing multi-mode global contextual semantics for sophisticated long-range dependencies modeling. We enable customized non-local feature interactions for each spatial position through re-weighted global context fusion in a non-normalized way. SAC is very lightweight and can be easily plugged into popular backbone models. Extensive experiments on COCO, ImageNet, and HICO-DET benchmarks show that our SAC block achieves significant performance improvements over existing baseline architectures while with a negligible computational burden increase. The results also demonstrate the exceptional effectiveness and scalability of the proposed approach on capturing long-range dependencies for object detection, segmentation, and image classification, outperforming a bank of state-of-the-art attention blocks.

PyNeval: A Python Toolbox for Evaluating Neuron Reconstruction Performance.

Quality assessment of tree-like structures obtained from a neuron reconstruction algorithm is necessary for evaluating the performance of the algorithm. The lack of user-friendly software for calculating common metrics motivated us to develop a Python toolbox called PyNeval, which is the first open-source toolbox designed to evaluate reconstruction results conveniently as far as we know. The toolbox supports popular metrics in two major categories, geometrical metrics and topological metrics, with an easy way to configure custom parameters for each metric. We tested the toolbox on both synthetic data and real data to show its reliability and robustness. As a demonstration of the toolbox in real applications, we used the toolbox to improve the performance of a tracing algorithm successfully by integrating it into an optimization procedure.

Brain-Machine Interface-Based Rat-Robot Behavior Control.

Brain-machine interface (BMI) provides a bidirectional pathway between the brain and external facilities. The machine-to-brain pathway makes it possible to send artificial information back into the biological brain, interfering neural activities and generating sensations. The idea of the BMI-assisted bio-robotic animal system is accomplished by stimulations on specific sites of the nervous system. With the technology of BMI, animals' locomotion behavior can be precisely controlled as robots, which made the animal turning into bio-robot. In this chapter, we reviewed our lab works focused on rat-robot navigation. The principles of rat-robot system have been briefly described first, including the target brain sites chosen for locomotion control and the design of remote control system. Some methodological advances made by optogenetic technologies for better modulation control have then been introduced. Besides, we also introduced our implementation of "mind-controlled" rat navigation system. Moreover, we have presented our efforts made on combining biological intelligence with artificial intelligence, with developments of automatic control and training system assisted with images or voices inputs. We concluded this chapter by discussing further developments to acquire environmental information as well as promising applications with write-in BMIs.

A Neuronal Pathway that Commands Deceleration in Drosophila Larval Light-Avoidance.

When facing a sudden danger or aversive condition while engaged in on-going forward motion, animals transiently slow down and make a turn to escape. The neural mechanisms underlying stimulation-induced deceleration in avoidance behavior are largely unknown. Here, we report that in Drosophila larvae, light-induced deceleration was commanded by a continuous neural pathway that included prothoracicotropic hormone neurons, eclosion hormone neurons, and tyrosine decarboxylase 2 motor neurons (the PET pathway). Inhibiting neurons in the PET pathway led to defects in light-avoidance due to insufficient deceleration and head casting. On the other hand, activation of PET pathway neurons specifically caused immediate deceleration in larval locomotion. Our findings reveal a neural substrate for the emergent deceleration response and provide a new understanding of the relationship between behavioral modules in animal avoidance responses.

Author Correction: A disinhibitory mechanism biases Drosophila innate light preference.

Affiliation 2 incorrectly read 'Department of Neurology of the Second Affiliated Hospital, Department of Neurobiology, Key Laboratory of Medical Neurobiology of the Ministry of Health of China, Key Laboratory of Neurobiology, Zhejiang University School of Medicine, Hangzhou, Zhejiang 310007, China' and affiliation 3 incorrectly read 'Qiushi Academy for Advanced Studies, Zhejiang University, Hangzhou, Zhejiang 310058, China.'

A disinhibitory mechanism biases Drosophila innate light preference.

Innate preference toward environmental conditions is crucial for animal survival. Although much is known about the neural processing of sensory information, how the aversive or attractive sensory stimulus is transformed through central brain neurons into avoidance or approaching behavior is largely unclear. Here we show that Drosophila larval light preference behavior is regulated by a disinhibitory mechanism. In the disinhibitory circuit, a pair of GABAergic neurons exerts tonic inhibition on one pair of contralateral projecting neurons that control larval reorientation behavior. When a larva enters the light area, the reorientation-controlling neurons are disinhibited to allow reorientation to occur as the upstream inhibitory neurons are repressed by light. When the larva exits the light area, the inhibition on the downstream neurons is restored to repress further reorientation and thus prevents the larva from re-entering the light area. We suggest that disinhibition may serve as a common neural mechanism for animal innate preference behavior.

Abdominal-Waving Control of Tethered Bumblebees Based on Sarsa With Transformed Reward.

Cyborg insects have attracted great attention as the flight performance they have is incomparable by micro aerial vehicles and play a critical role in supporting extensive applications. Approaches to construct cyborg insects consist of two major issues: 1) the stimulating paradigm and 2) the control policy. At present, most cyborg insects are constructed based on invasive methods, requiring the implantation of electrodes into neural or muscle systems, which would harm the insects. As the control policy is basically manual control, the shortcomings of which lie in the requirement of excessive amount of experiments and focused attention. This paper presents the design and implementation of a noninvasive and much safer cyborg insect system based on visual stimulation. The tethered paradigm is adopted here and we look at controlling the flight behavior of bumblebees, especially the abdominal-waving behavior, in the context of a model-free reinforcement learning problem. The problem is formulated as a finite and deterministic Markov decision process, where the agent is designed to change the abdominal-waving behavior from the initial state to the target state. Sarsa with transformed reward function which can speed up the learning process is employed to learn the optimal control policy. Learned policies are compared to the stochastic one by evaluating the results of ten bumblebees, demonstrating that abdominal-waving state can be modulated to approximate the target state quickly with small deviation.

Locally Linear Approximation Approach for Incomplete Data.

The matrix completion problem is restoring a given matrix with missing entries when handling incomplete data. In many existing researches, rank minimization plays a central role in matrix completion. In this paper, noticing that the locally linear reconstruction can be used to approximate the missing entries, we view the problem from a new perspective and propose an algorithm called locally linear approximation (LLA). The LLA method tries to keep the local structure of the data space while restoring the missing entries from row angle and column angle simultaneously. The experimental results have demonstrated the effectiveness of the proposed method.

Identification of genes related to high royal jelly production in the honey bee (Apis mellifera) using microarray analysis

Abstract China is the largest royal jelly producer and exporter in the world, and high royal jelly-yielding strains have been bred in the country for approximately three decades. However, information on the molecular mechanism underlying high royal jelly production is scarce. Here, a cDNA microarray was used to screen and identify differentially expressed genes (DEGs) to obtain an overview on the changes in gene expression levels between high and low royal jelly producing bees. We developed a honey bee gene chip that covered 11,689 genes, and this chip was hybridised with cDNA generated from RNA isolated from heads of nursing bees. A total of 369 DEGs were identified between high and low royal jelly producing bees. Amongst these DEGs, 201 (54.47%) genes were up-regulated, whereas 168 (45.53%) were down-regulated in high royal jelly-yielding bees. Gene ontology (GO) analyses showed that they are mainly involved in four key biological processes, and pathway analyses revealed that they belong to a total of 46 biological pathways. These results provide a genetic basis for further studies on the molecular mechanisms involved in high royal jelly production.

Identification of genes related to high royal jelly production in the honey bee (Apis mellifera) using microarray analysis.

China is the largest royal jelly producer and exporter in the world, and high royal jelly-yielding strains have been bred in the country for approximately three decades. However, information on the molecular mechanism underlying high royal jelly production is scarce. Here, a cDNA microarray was used to screen and identify differentially expressed genes (DEGs) to obtain an overview on the changes in gene expression levels between high and low royal jelly producing bees. We developed a honey bee gene chip that covered 11,689 genes, and this chip was hybridised with cDNA generated from RNA isolated from heads of nursing bees. A total of 369 DEGs were identified between high and low royal jelly producing bees. Amongst these DEGs, 201 (54.47%) genes were up-regulated, whereas 168 (45.53%) were down-regulated in high royal jelly-yielding bees. Gene ontology (GO) analyses showed that they are mainly involved in four key biological processes, and pathway analyses revealed that they belong to a total of 46 biological pathways. These results provide a genetic basis for further studies on the molecular mechanisms involved in high royal jelly production.