Visual Research Articles

Precision improvement of robotic bioprinting via vision-based tool path compensation

Robotic 3D bioprinting is a rapidly advancing technology with applications in organ fabrication, tissue restoration, and pharmaceutical testing. While the stepwise generation of organs characterizes bioprinting, challenges such as non-linear material behavior, layer shifting, and trajectory tracking are common in freeform reversible embedding of suspended hydrogels (FRESH) bioprinting, leading to imperfections in complex organ construction. To overcome these limitations, we propose a computer vision-based strategy to identify discrepancies between printed filaments and the reference robot path. Employing error compensation techniques, we generate an adjusted reference path, enhancing robotic 3D bioprinting by adapting the robot path based on vision system data. Experimental assessments confirm the reliability and agility of our vision-based robotic 3D bioprinting approach, showcasing precision in fabricating human blood vessel segments through case studies. Significantly, it minimizes the printing layer width disparity to just 0.15 mm compared to the 0.6 mm in traditional methods, and it decreases the average error for curved filaments to 7.0 mm2 from the previous 12.7 mm2 in conventional printing. While these results underscore the significant potential of our innovation in creating precise biomimetic constructs, further investigation is necessary to tackle challenges such as accurately distinguishing closely stacked layers using a vision system, especially under varying lighting conditions. These limitations, coupled with issues of computational complexity and scalability in larger-scale bioprinting, emphasize the importance of enhancing the reliability of the vision-based approach across various conditions. Nonetheless, our innovation demonstrates substantial promise in creating precise biomimetic constructs and paves the way for future advancements in vision-guided robotic bioprinting, including the integration of multi-material printing techniques to enhance versatility.

Menstrual cycle-driven hormone concentrations co-fluctuate with white and gray matter architecture changes across the whole brain.

Cyclic fluctuations in hypothalamic-pituitary-gonadal axis (HPG-axis) hormones exert powerful behavioral, structural, and functional effects through actions on the mammalian central nervous system. Yet, very little is known about how these fluctuations alter the structural nodes and information highways of the human brain. In a study of 30 naturally cycling women, we employed multidimensional diffusion and T1-weighted imaging during three estimated menstrual cycle phases (menses, ovulation, and mid-luteal) to investigate whether HPG-axis hormone concentrations co-fluctuate with alterations in white matter (WM) microstructure, cortical thickness (CT), and brain volume. Across the whole brain, 17β-estradiol and luteinizing hormone (LH) concentrations were directly proportional to diffusion anisotropy (μFA; 17β-estradiol: β1 = 0.145, highest density interval (HDI) = [0.211, 0.4]; LH: β1 = 0.111, HDI = [0.157, 0.364]), while follicle-stimulating hormone (FSH) was directly proportional to CT (β1 = 0 .162, HDI = [0.115, 0.678]). Within several individual regions, FSH and progesterone demonstrated opposing relationships with mean diffusivity (Diso) and CT. These regions mainly reside within the temporal and occipital lobes, with functional implications for the limbic and visual systems. Finally, progesterone was associated with increased tissue (β1 = 0.66, HDI = [0.607, 15.845]) and decreased cerebrospinal fluid (CSF; β1 = -0.749, HDI = [-11.604, -0.903]) volumes, with total brain volume remaining unchanged. These results are the first to report simultaneous brain-wide changes in human WM microstructure and CT coinciding with menstrual cycle-driven hormone rhythms. Effects were observed in both classically known HPG-axis receptor-dense regions (medial temporal lobe, prefrontal cortex) and in other regions located across frontal, occipital, temporal, and parietal lobes. Our results suggest that HPG-axis hormone fluctuations may have significant structural impacts across the entire brain.

Sanhua Tang protects against ischemic stroke by preventing blood-brain barrier injury: a network pharmacology and experiments.

To assess the effect and mechanism of Sanhua Tang (, SHT) in treating ischemic stroke (IS) through the "Kaitong Xuanfu" theory by using network pharmacology and animal experiments. The active ingredients and targets of SHT and IS were screened by public databases such as Traditional Chinese Medicine systems pharmacology, GeneCards, and online mendelian inheritance in man. Visual network topographies were constructed using R, Cytoscape 3.6.0, AutoDockTools, a user-sponsored molecular visualization system on an open-source foundation, and other software to analyze the correlation between targets and active ingredients. The middle cerebral artery occlusion (MCAO) model was established by operation. Animals were divided into the Sham group, MCAO group (M group), aloe-emodin (AE) group (MCAO rats treated with aloe-emodin), SHT at low dosage (SL group) (MCAO rats treated with SL), SHT at medium dosage (SM group), and SHT at high dosage (SH group). 2,3,5-triphenyl tetrazolium chloride staining was used to detect the volume of cerebral infarction; Nissl staining was used to observe the morphology of neuronal cells; transmission electron microscopy was used to observe the integrity of the blood-brain barrier (BBB); enzyme-linked immunosorbent assay was used to detect the content of interleukin-6 (IL-6), IL-10, tumor necrosis factor α (TNF-α) in serum. Western blot was used to detect the expression of vascular endothelial growth factor A (VEGFA) protein in the cerebral ischemic penumbra. Using network pharmacology and molecular docking validation, four active ingredients (lignan, naringenin, aloe-rhodopsin, and β-sitosterol), seven target proteins (protein kinase b 1, IL-6, TNF, VEGFA, TP53, jun proto-oncogene, and cysteinyl aspartate specific proteinase 3), and inflammatory signaling pathways were identified. Animal experiments showed that the SH and AE groups had fewer neurological deficits, reduced brain infarct volumes, decreased serum inflammatory factor levels, increased expression of VEGFA protein, and less structural damage to neurons and BBB. The present study found that the therapeutic mechanism of SHT against IS may be related to the inhibition of BBB inflammatory damage, which is also the mechanism of "Kaitong Xuanfu." The high-dose group of SHT was relatively effective in regulating inflammatory factors, improving BBB permeability, and protecting neuronal cells from damage.

Comparison of T1-weighted landmark placement and ROI transfer onto diffusion-weighted EPI sequences for targeted tractography tasks in the optic nerve.

Diffusion-based tractography in the optic nerve requires sampling strategies assisted by anatomical landmark information (regions of interest [ROIs]). We aimed to investigate the feasibility of expert-placed, high-resolution T1-weighted ROI-data transfer onto lower spatial resolution diffusion-weighted images. Slab volumes from 20 volunteers were acquired and preprocessed including distortion bias correction and artifact reduction. Constrained spherical deconvolution was used to generate a directional diffusion information grid (fibre orientation distribution-model [FOD]). Three neuroradiologists marked landmarks on both diffusion imaging variants and structural datasets. Structural ROI information (volumetric interpolated breath-hold sequence [VIBE]) was respectively registered (linear with 6/12 degrees of freedom [DOF]) onto single-shot EPI (ss-EPI) and readout-segmented EPI (rs-EPI) volumes, respectively. All eight ROI/FOD-combinations were compared in a targeted tractography task of the optic nerve pathway. Inter-rater reliability for placed ROIs among experts was highest in VIBE images (lower confidence interval 0.84 to 0.97, mean 0.91) and lower in both ss-EPI (0.61 to 0.95, mean 0.79) and rs-EPI (0.59 to 0.86, mean 0.70). Tractography success rate based on streamline selection performance was highest in VIBE-drawn ROIs registered (6-DOF) onto rs-EPI FOD (70.0% over 5%-threshold, capped to failed ratio 39/16) followed by both 12-DOF-registered (67.5%; 41/16) and nonregistered VIBE (67.5%; 40/23). On ss-EPI FOD, VIBE-ROI-datasets obtained fewer streamlines overall with each at 55.0% above 5%-threshold and with lower capped to failed ratio (6-DOF: 35/36; 12-DOF: 34/34, nonregistered 33/36). The combination of VIBE-placed ROIs (highest inter-rater reliability) with 6-DOF registration onto rs-EPI targets (best streamline selection performance) is most suitable for white matter template generation required in group studies.

Identification of Key Efferocytosis-Related Genes and Mechanisms in Diabetic Retinopathy.

This study aimed to explore the key efferocytosis-related genes in diabetic retinopathy (DR) and their regulatory mechanisms. Public DR-related gene expression datasets, GSE160306 (training) and GSE60436 (validation), were downloaded. Differentially expressed efferocytosis-related genes (DEERGs) were analyzed using differential expression analysis and weighted gene co-expression network analysis. Functional enrichment analysis was conducted. Moreover, efferocytosis-related signature genes were identified using machine learning analysis, and their expression levels and diagnostic value were analyzed. Furthermore, nomograms were constructed; immune cell infiltration was analyzed; and gene set enrichment analysis, transcriptional regulation analysis, and small-molecule drug (SMD) prediction of efferocytosis-related signature genes were performed. In total, 36 DEERGs were identified in DR, and were markedly enriched in multiple functions, such as visual system development. Through further machine learning analysis, two efferocytosis-related signature genes, Ferritin Light Chain (FTL) and Fc Gamma Binding Protein (FCGBP), were identified, and were found to be upregulated in DR samples and showed high diagnostic performance for DR. A nomogram constructed using FTL and FCGBP accurately predicted the risk of DR. Moreover, the level of infiltration of immature B cells was positively correlated with FTL and FCGBP expression levels. Multiple transcription factors (TFs), such as CCCTC-Binding Factor (CTCF) and KLF Transcription Factor 9 (KLF9), were found to interact with both FTL and FCGBP. In addition, FTL can be targeted by miRNAs, such as miR-22-3p, and FCGBP can be targeted by miR-7973. In addition, both FTL and FCGBP can be targeted by SMDs, such as bisphenol A. Key efferocytosis-related genes, such as FTL and FCGBP, may promote DR development. Detecting or targeting FTL and FCGBP may aid in the prevention, diagnosis, and treatment of DR.

Ferroptosis contributes to diabetes-induced visual pathway neuronal damage via iron accumulation and GPX4 inactivation.

The damage of the diabetic visual pathway is one of the main causes of blindness in diabetic patients. Visual pathways include anatomic parts from the retina to the occipital lobe. This study investigated the involvement of ferroptosis, a planned cell death brought on by the buildup of free iron in cells, in the impairment of visual pathways in diabetes mellitus. Streptozotocin (STZ) was used to construct a diabetic rat model. Pathological and ultrastructural changes of the occipital lobe, retina, and optic nerve were observed by Hematoxylin-Eosin (HE) staining and transmission electron microscopy (TEM). The expressions of Neuronal nuclei (NeuN), Glial fibrillary acidic protein (GFAP), and Glutathione Peroxidase 4 (GPX4) in the occipital lobe and retina were detected by immunofluorescence, and Western Blotting was used to identify the NeuN GFAP and GPX4 expressions in the occipital lobe. Iron content in the occipital lobe and retina was detected by Iron Assay Kit. The success rate of the diabetic rat model was 93.3%. In the diabetic group, the cells of the occipital lobe and retina were arranged disorderly, and the boundaries were unclear. The membrane of the occipital lobe, retina, and optic nerve was broken, some vacuoles were observed, mitochondrial morphology was changed, swelling was observed, and the mitochondrial ridge disappeared. There was a large increase in GFAP expression and iron concentration and a significant decrease in the expression of NeuN, and GPX4 in the retina and occipital lobe. Ferroptosis plays an important role in visual pathway damage in diabetes, and GPX4 regulates this process.

Saccadic "inhibition" unveils the late influence of image content on oculomotor programming.

Image content is prioritized in the visual system. Faces are a paradigmatic example, receiving preferential processing along the visual pathway compared to other visual stimuli. Moreover, face prioritization manifests also in behavior. People tend to look at faces more frequently and for longer periods, and saccadic reaction times can be faster when targeting a face as opposed to a phase-scrambled control. However, it is currently not clear at which stage image content affects oculomotor planning and execution. It can be hypothesized that image content directly influences oculomotor signal generation. Alternatively, the image content could exert its influence on oculomotor planning and execution at a later stage, after the image has been processed. Here we aim to disentangle these two alternative hypotheses by measuring the frequency of saccades toward a visual target when the latter is followed by a visual transient in the central visual field. Behaviorally, this paradigm leads to a reduction in saccade frequency that happens about 90 ms after any visual transient event, also known as saccadic "inhibition". In two experiments, we measured occurrence of saccades in visually guided saccades as well as microsaccades during fixation, using face and noise-matched visual stimuli. We observed that while the reduction in saccade occurrence was similar for both stimulus types, face stimuli lead to a prolonged reduction in eye movements. Moreover, saccade kinematics were altered by both stimulus types, showing an amplitude reduction without change in peak velocity for the earliest saccades. Taken together, our experiments imply that face stimuli primarily affect the later stages of the behavioral phenomenon of saccadic "inhibition". We propose that while some stimulus features are processed at an early stage and can quickly influence eye movements, a delayed signal conveying image content information is necessary to further inhibit/delay activity in the oculomotor system to trigger eye movements.

Optoelectronic Neuromorphic Logic Memory Device Based on Ga2O3/MoS2 Van der Waals Heterostructure with High Rectification and On/Off Ratios

AbstractIt is crucial to develop advanced optoelectronic devices that incorporate multiple functions, including sensing, storage, and computing, which is considered at the forefront of semiconductor optoelectronics to meet emerging functional diversification. In this study, by stacking the n‐type Ga2O3 with the n‐type MoS2 flakes, a Ga2O3/MoS2 heterostructure optoelectronic device with high rectification ratio of ≈105 and on/off ratio of ≈108 is fabricated, which achieves high detectivity of 1.34 × 109 Jones and high responsivity of 28.92 mA/W. More importantly, the Ga2O3/MoS2 heterostructure device shows potential ability to integrate sensing and memorizing, simultaneously, which can be used as artificial neuromorphic synaptic. The device exhibits excellent photo‐induced synaptic functions including short‐term plasticity, long‐term plasticity, and paired‐pulse facilitation, realizing the ability to couple light and electrical signals by Pavlovian associative learning. At last, the device also demonstrates the information processing ability to act as optoelectronic logic gate AND by synergistically regulating the light on/off states and gate voltage. The research introduces an innovative strategy for the development of next‐generation optoelectronic devices which are highly integrated with sensing, memory, and logic processing functions, demonstrating great application prospects in constructing an efficient artificial neuromorphic visual and logic systems.

Large-scale calcium imaging reveals a systematic V4 map for encoding natural scenes

Biological visual systems have evolved to process natural scenes. A full understanding of visual cortical functions requires a comprehensive characterization of how neuronal populations in each visual area encode natural scenes. Here, we utilized widefield calcium imaging to record V4 cortical response to tens of thousands of natural images in male macaques. Using this large dataset, we developed a deep-learning digital twin of V4 that allowed us to map the natural image preferences of the neural population at 100-µm scale. This detailed map revealed a diverse set of functional domains in V4, each encoding distinct natural image features. We validated these model predictions using additional widefield imaging and single-cell resolution two-photon imaging. Feature attribution analysis revealed that these domains lie along a continuum from preferring spatially localized shape features to preferring spatially dispersed surface features. These results provide insights into the organizing principles that govern natural scene encoding in V4.

BOPVis: Bridge Monitoring Data Visualization for Operational Performance Mining

Bridges are fundamental facilities in the transportation system, and their operational performance is crucial for economic and social development. Many large bridges are now equipped with structural health monitoring (SHM) systems that collect various types of real-time data. However, our user study found that despite the accumulation of massive amounts of monitoring data, current analysis methods cannot efficiently process large-scale, high-dimensional data. To address this, we have developed BOPVis, a visualization system for bridge monitoring data. BOPVis allows users to intuitively locate sensors and extract corresponding data from a 3D digital model of a bridge. It also provides convenient and flexible interactions for examining trends over time and correlations across hundreds of monitoring channels. A real-world long-span suspension bridge in China is used as a case study to demonstrate the advantages of the BOPVis system for operational performance mining. Through BOPVis, the global temperature deformation behaviors of the bridge are explored and found to align with the physical mechanism documented in the SHM literature. The BOPVis system, with its interactive visualization analysis capabilities, offers a new method for analyzing bridge monitoring data.

The visual image system of the city of Saratov

The visual image system of the city of Saratov

Integrated Bionic Human Retina Process and In‐Sensor RC System Based on 2D Retinomorphic Memristor Array

AbstractThe rapid development of machine vision has put forward higher requirements for image perception and visual learning systems. However, limited by the von Neumann bottleneck, traditional artificial vision systems rely on complex optical and circuit designs, and the separate distribution of sensors, memory, and computing modules results in high energy consumption and high latency. Memristor‐based reservoir computing (RC) system with integrated sensing and memory functions provides a solution to effectively improve the computational efficiency of artificial vision networks. Here, a complete artificial visual recognition system based on in‐sensor RC is developed by combining the excellent optoelectronic synaptic properties of 2D WS2 memristor array. Through effectively regulating the conductance of memristors by light or electrical stimulation, image information is converted into high‐dimensional information in the reservoir layer, and the classification can be completed by simple training, significantly reducing the training complexity and cost consumption. With this system, an effective recognition of hand‐written numbers with an accuracy of 88.3% is demonstrated, and a 100% recognition of traffic signals of different colors is achieved. The proposed in‐sensor RC system advances the development and application of artificial vision recognition, paving the way for efficient machine learning and neuromorphic vision systems.

Looks like home: numerosity, but not spatial frequency guides preference in zebrafish larvae (Danio rerio)

Despite their young age, zebrafish larvae have a well-developed visual system and can distinguish between different visual stimuli. First, we investigated if the first visual surroundings the larvae experience during the first days after hatching shape their habitat preference. Indeed, these animals seem to “imprint” on the first surroundings they see and select visual stimuli accordingly at 7 days post fertilization (dpf). In particular, if zebrafish larvae experience a bar background just after hatching, they later on prefer bars over white stimuli, and vice versa. We then used this acquired preference for bars to investigate innate numerical abilities. We wanted to specifically test if the zebrafish larvae show real numerical abilities or if they rely on a lower-level mechanism—i.e. spatial frequency—to discriminate between two different numerosities. When we matched the spatial frequency in stimuli with different numbers of bars, the larvae reliably selected the higher numerosity. A previous study has ruled out that 7 dpf zebrafish larvae use convex hull, cumulative surface area and density to choose between two numerosities. Therefore, our results indicate that zebrafish larvae rely on real numerical abilities rather than other cues, including spatial frequency, when spontaneously comparing two sets with different numbers of bars.

A Photoelectrochemical Retinomorphic Synapse.

Reproducing human visual functions with artificial devices is a long-standing goal of the neuromorphic domain. However, emulating the chemical language communication of the visual system in fluids remains a grand challenge. Here, a "multi-color" hydrogel-based photoelectrochemical retinomorphic synapse is reported with unique chemical-ionic-electrical signaling in an aqueous electrolyte that enables, e.g., color perception and biomolecule-mediated synaptic plasticity. Based on the specific enzyme-catalyzed chromogenic reactions, three multifunctional colored hydrogels are developed, which can not only synergize with the Bi2S3 photogate to recognize the primary colors but also synergize with a given polymeric channel to promote the long-term memory of the system. A synaptic array is further constructed for sensing color images and biomolecule-coded information communication. Taking advantage of the versatile biochemistry, the biochemical-driven reversible photoelectric response of the cone cell is further mimicked. This work introduces rich chemical designs into retinomorphic devices, providing a perspective for replicating the human visual system in fluids.

Neural Substrates for Early Data Reduction in Fast Vision: A Psychophysical Investigation

To ensure survival, the visual system must rapidly extract the most important elements from a large stream of information. This necessity clashes with the computational limitations of the human brain, so a strong early data reduction is required to efficiently process information in fast vision. A theoretical early vision model, recently developed to preserve maximum information using minimal computational resources, allows efficient image data reduction by extracting simplified sketches containing only optimally informative, salient features. Here, we investigate the neural substrates of this mechanism for optimal encoding of information, possibly located in early visual structures. We adopted a flicker adaptation paradigm, which has been demonstrated to specifically impair the contrast sensitivity of the magnocellular pathway. We compared flicker-induced contrast threshold changes in three different tasks. The results indicate that, after adapting to a uniform flickering field, thresholds for image discrimination using briefly presented sketches increase. Similar threshold elevations occur for motion discrimination, a task typically targeting the magnocellular system. Instead, contrast thresholds for orientation discrimination, a task typically targeting the parvocellular system, do not change with flicker adaptation. The computation performed by this early data reduction mechanism seems thus consistent with magnocellular processing.

Spatial resolution and optical sensitivity in the compound eyes of two common European wasps, Vespula germanica and Vespula vulgaris.

Vespula germanica and Vespula vulgaris are two common European wasps that have ecological and economic importance due to their artificial introduction to many different countries and environments. Their success has undoubtedly been aided by their capacity for visually guided hunting, foraging, learning, and using visual cues in the context of homing and navigation. However, the visual systems of V. germanica and V. vulgaris have not received any deep attention. We used electrophysiology, together with optical and anatomical techniques, to measure the spatial resolution and optical sensitivity of the compound eyes of both species. We found that both wasps have high anatomical spatial resolution with narrow interommatidial angles (Δφ between 1.0 and 1.5°) and a distinct acute zone in the fronto-ventral part of the eye. These narrow interommatidial angles are matched to photoreceptors having narrow angular sensitivities (acute zone acceptance angles Δρ below 1.3°), indicating eyes of high spatial resolution that are well suited to their ecological needs. Additionally, we found that both species possess an optical sensitivity that is typical of other day-flying hymenopterans.

Anchor objects drive realism while diagnostic objects drive categorization in GAN generated scenes

Our visual surroundings are highly complex. Despite this, we understand and navigate them effortlessly. This requires transforming incoming sensory information into representations that not only span low- to high-level visual features (e.g., edges, object parts, objects), but likely also reflect co-occurrence statistics of objects in real-world scenes. Here, so-called anchor objects are defined as being highly predictive of the location and identity of frequently co-occuring (usually smaller) objects, derived from object clustering statistics in real-world scenes, while so-called diagnostic objects are predictive of the larger semantic context (i.e., scene category). Across two studies (N1 = 50, N2 = 44), we investigate which of these properties underlie scene understanding across two dimensions – realism and categorisation – using scenes generated from Generative Adversarial Networks (GANs) which naturally vary along these dimensions. We show that anchor objects and mainly high-level features extracted from a range of pre-trained deep neural networks (DNNs) drove realism both at first glance and after initial processing. Categorisation performance was mainly determined by diagnostic objects, regardless of realism, at first glance and after initial processing. Our results are testament to the visual system’s ability to pick up on reliable, category specific sources of information that are flexible towards disturbances across the visual feature-hierarchy.

Optic neuropathy from hypovitaminosis A in a series of children with severe dietary restrictions.

Hypovitaminosis A is a leading cause of preventable childhood blindness, especially in developing nations. Vitamin A is a fat-soluble essential micronutrient that serves vital functions in the visual system and in regulating bone resorption. We report on a series of four children with mixed nutritional and compressive optic neuropathy and provide a review of the literature. A retrospective observational study of four males (ages 9-12), three with autism spectrum disorder who presented with loss of vision and multiple vitamin deficiencies including hypovitaminosis A. Patients presented with unexplained visual loss or a change in visual behaviour. All patients had severely restricted diet comprising of predominantly carbohydrates. Two of the four cases demonstrated optic nerve pallor at initial presentation with marked optic atrophy developing in all patients over time. Electrophysiology available in two patients demonstrated optic nerve dysfunction with preserved retinal function. Extensive investigations revealed profound deficiency in multiple vitamins including vitamin A (<0.1-0.2 μmol/L, normal = 0.9-1.7 μmol/L). Three patients also had low vitamin B12 (90-111 pmol/L, normal = 170-800 pmol/L) with normal folate. All four cases had radiological evidence of skull base thickening indicative of low vitamin A. Genetic testing did not find any relevant pathogenic variants. Hypovitaminosis A is a crucial form of nutritional deprivation that results in significant visual loss with potential hyperostosis and optic nerve compression exacerbating nutritional optic neuropathy. Additional micronutrient deficiencies usually co-exist and may contribute. Extra vigilance in vitamin replacement is required of clinicians with patients with autism who have restricted diets.

Jellyfish-Inspired Visual and Sensory Bubbling Robots with Automatic 3D Morphable Films for Underwater Environmental Interactions.

Coelenterates, such as Atolla jellyfish, are capable of integrating color, communication, and motion in a sophisticated manner, thereby enabling them to function as intelligent biological systems that can adapt to the challenges of the underwater environment. Extensive efforts have been dedicated to exploiting underwater visual, sensory, actuating, or combined systems. However, current biomimetic soft systems are still limited by the lack of comprehensive, integrated functional skins that can automatically deform, dynamically sense, and further send color signals when diving into underwater conditions. Here, we propose the synthetic soft skins composed of assembled entangled carbon nanotube networks and fluorescent unit-embedded elastomers which can be applied in a suspended form to allow autonomic 3D deformation, real-time perception, and dynamic fluorescence color transformation. The capabilities of the sensory and color display thresholds were controlled through the entanglement density of carbon nanotubes and the suspended area. As a demonstration, the soft thin skin was integrated into an artificial jellyfish robot, enabling the realization of a closed-loop feedback system for dynamic sensory processing, signal processing, and further 3D morphing-induced fluorescent color change, demonstrating significant potentials in underwater visual display, danger warning, and environmental exploration.

An insect vision-inspired neuromorphic vision systems in low-light obstacle avoidance for intelligent vehicles

An insect vision-inspired neuromorphic vision systems in low-light obstacle avoidance for intelligent vehicles