GCPDFFNet: Small Object Detection for Rice Blast Recognition.

Early detection of rice blast disease is pivotal to ensure rice yield. We collected in situ images of rice blast and constructed a rice blast dataset based on variations in lesion shape, size, and color. Given that rice blast lesions are small and typically exhibit round, oval, and fusiform shapes, we proposed a small object detection model named GCPDFFNet (global context-based parallel differentiation feature fusion network) for rice blast recognition. The GCPDFFNet model has three global context feature extraction modules and two parallel differentiation feature fusion modules. The global context modules are employed to focus on the lesion areas; the parallel differentiation feature fusion modules are used to enhance the recognition effect of small-sized lesions. In addition, we proposed the SCYLLA normalized Wasserstein distance loss function, specifically designed to accelerate model convergence and improve the detection accuracy of rice blast disease. Comparative experiments were conducted on the rice blast dataset to evaluate the performance of the model. The proposed GCPDFFNet model outperformed the baseline network CenterNet, with a significant increase in mean average precision from 83.6 to 95.4% on the rice blast test set while maintaining a satisfactory frames per second drop from 147.9 to 122.1. Our results suggest that the GCPDFFNet model can accurately detect in situ rice blast disease while ensuring the inference speed meets the real-time requirements.

Tri-Flow-YOLO: Counter helps to improve cross-domain object detection.

The excellence of intelligent detection models has been widely recognized, but in terms of cross-domain scenes, they still face performance degradation and low accuracy. A multi-supervised Tri-Flow-YOLO model is proposed to improve the accuracy of objects with various scales under cross-domain conditions. Based on the full-supervised traditional detection branch of YOLOv5, another two mutually supporting task branches are designed intently. In brief, we add unsupervised adversarial classification training flow to the backend, to realize the feature alignment requirements and improve the cross-domain performance stability of the model. Meanwhile, a weakly-supervised object counting flow is proposed to improve the model's attention to all the objects and the detection ability is efficiently enforced. In addition, I-Mosaic and iCIOU are designed especially for small hard objects, enriching the positive samples during the training process. With the auxiliary of both improved strategies, the imbalance of positive and negative samples in the anchor-based model is relieved accordingly. The experimental results show that the improved Tri-Flow-YOLO model achieves 56.0 mAP in the Cityscapes→Foggy-Cityscapes task, and 49.8 mAP in the VOC→Clipart task.

The verification of the utility of a commercially available phantom combination for quality control in contrast-enhanced mammography.

Contrast-enhanced mammography is being increasingly implemented clinically, providing much improved contrast between tumour and background structures, particularly in dense breasts. Although CEM is similar to conventional mammography it differs via an additional exposure with high energy X-rays (≥ 40 kVp) and subsequent image subtraction. Because of its special operational aspects, the CEM aspect of a CEM unit needs to be uniquely characterised and evaluated. This study aims to verify the utility of a commercially available phantom set (BR3D model 020 and CESM model 022 phantoms (CIRS, Norfolk, Virginia, USA)) in performing key CEM performance tests (linearity of system response with iodine concentration and background subtraction) on two models of CEM units in a clinical setting. The tests were successfully performed, yielding results similar to previously published studies. Further, similarities and differences in the two systems from different vendors were highlighted, knowledge of which may potentially facilitate optimisation of the systems.

Continuous reach-to-grasp motion recognition based on an extreme learning machine algorithm using sEMG signals.

Recognizing user intention in reach-to-grasp motions is a critical challenge in rehabilitation engineering. To address this, a Machine Learning (ML) algorithm based on the Extreme Learning Machine (ELM) was developed for identifying motor actions using surface Electromyography (sEMG) during continuous reach-to-grasp movements, involving multiple Degrees of Freedom (DoFs). This study explores feature extraction methods based on time domain and autoregressive models to evaluate ELM performance under different conditions. The experimental setup encompassed variations in neuron size, time windows, validation with each muscle, increase in the number of features, comparison with five conventional ML-based classifiers, inter-subjects variability, and temporal dynamic response. To evaluate the efficacy of the proposed ELM-based method, an openly available sEMG dataset containing data from 12 participants was used. Results highlight the method's performance, achieving Accuracy above 85%, F-score above 90%, Recall above 85%, Area Under the Curve of approximately 84% and compilation times (computational cost) of less than 1 ms. These metrics significantly outperform standard methods (p < 0.05). Additionally, specific trends were found in increasing and decreasing performance in identifying specific tasks, as well as variations in the continuous transitions in the temporal dynamics response. Thus, the ELM-based method effectively identifies continuous reach-to-grasp motions through myoelectric data. These findings hold promise for practical applications. The method's success prompts future research into implementing it for more reliable and effective Human-Machine Interface (HMI) control. This can revolutionize real-time upper limb rehabilitation, enabling natural and complex Activities of Daily Living (ADLs) like object manipulation. The robust results encourages further research and innovative solutions to improve people's quality of life through more effective interventions.

It's all in the timing: Delayed feedback in autism may weaken predictive mechanisms during contour integration.

Humans rely on predictive mechanisms during visual processing to efficiently resolve incomplete or ambiguous sensory signals. While initial low-level sensory data are conveyed by feedforward connections, feedback connections are believed to shape sensory processing through conveyance of statistical predictions based on prior exposure to stimulus configurations. Individuals with autism spectrum disorder (ASD) show biases in stimulus processing toward parts rather than wholes, suggesting their sensory processing may be less shaped by statistical predictions acquired through prior exposure to global stimulus properties. Investigations of illusory contour (IC) processing in neurotypical (NT) adults have established a well-tested marker of contour integration characterized by a robust modulation of the visually evoked potential (VEP) - the IC-effect - that occurs over lateral occipital scalp during the timeframe of the N1 component. Converging evidence strongly supports the notion that this IC-effect indexes a signal with significant feedback contributions. Using high-density VEPs, we compared the IC-effect in 6-7-year-old children with ASD (n=32) or NT development (n=53). Both groups of children generated an IC-effect that was equivalent in amplitude. However, the IC-effect notably onset 21ms later in ASD, even though initial VEP afference was identical across groups. This suggests that feedforward information predominated during perceptual processing for 15% longer in ASD compared to NT children. This delay in the feedback dependent IC-effect, in the context of known developmental differences between feedforward and feedback fibers, suggests a potential pathophysiological mechanism of visual processing in ASD, whereby ongoing stimulus processing is less shaped by statistical prediction mechanisms.

A streamlined approach for intelligent ship object detection using EL-YOLO algorithm.

Maritime objects frequently exhibit low-quality and insufficient feature information, particularly in complex maritime environments characterized by challenges such as small objects, waves, and reflections. This situation poses significant challenges to the development of reliable object detection including the strategies of loss function and the feature understanding capabilities in common YOLOv8 (You Only Look Once) detectors. Furthermore, the widespread adoption and unmanned operation of intelligent ships have generated increasing demands on the computational efficiency and cost of object detection hardware, necessitating the development of more lightweight network architectures. This study proposes the EL-YOLO (Efficient Lightweight You Only Look Once) algorithm based on YOLOv8, designed specifically for intelligent ship object detection. EL-YOLO incorporates novel features, including adequate wise IoU (AWIoU) for improved bounding box regression, shortcut multi-fuse neck (SMFN) for a comprehensive analysis of features, and greedy-driven filter pruning (GDFP) to achieve a streamlined and lightweight network design. The findings of this study demonstrate notable advancements in both detection accuracy and lightweight characteristics across diverse maritime scenarios. EL-YOLO exhibits superior performance in intelligent ship object detection using RGB cameras, showcasing a significant improvement compared to standard YOLOv8 models.

HD-tDCS to the lateral occipital complex improves haptic object recognition.

High-definition transcranial direct current stimulation (HD-tDCS) is a non-invasive brain stimulation technique that has been shown to be safe and effective in modulating neuronal activity. The present study investigates the effect of anodal HD-tDCS on haptic object perception and memory through stimulation of the lateral occipital complex (LOC), a structure that has been shown to be involved in both visual and haptic object recognition. In this single-blind, sham-controlled, between-subjects study, blindfolded healthy, sighted participants used their right (dominant) hand to perform haptic discrimination and recognition tasks with 3D-printed, novel objects called "Greebles" while receiving 20 min of 2 milliamp (mA) anodal stimulation (or sham) to the left or right LOC. Compared to sham, those who received left LOC stimulation (contralateral to the hand used) showed an improvement in haptic object recognition but not discrimination-a finding that was evident from the start of the behavioral tasks. A second experiment showed that this effect was not observed with right LOC stimulation (ipsilateral to the hand used). These results suggest that HD-tDCS to the left LOC can improve recognition of objects perceived via touch. Overall, this work sheds light on the LOC as a multimodal structure that plays a key role in object recognition in both the visual and haptic modalities.

Reliable object tracking by multimodal hybrid feature extraction and transformer-based fusion.

Visual object tracking, which is primarily based on visible light image sequences, encounters numerous challenges in complicated scenarios, such as low light conditions, high dynamic ranges, and background clutter. To address these challenges, incorporating the advantages of multiple visual modalities is a promising solution for achieving reliable object tracking. However, the existing approaches usually integrate multimodal inputs through adaptive local feature interactions, which cannot leverage the full potential of visual cues, thus resulting in insufficient feature modeling. In this study, we propose a novel multimodal hybrid tracker (MMHT) that utilizes frame-event-based data for reliable single object tracking. The MMHT model employs a hybrid backbone consisting of an artificial neural network (ANN) and a spiking neural network (SNN) to extract dominant features from different visual modalities and then uses a unified encoder to align the features across different domains. Moreover, we propose an enhanced transformer-based module to fuse multimodal features using attention mechanisms. With these methods, the MMHT model can effectively construct a multiscale and multidimensional visual feature space and achieve discriminative feature modeling. Extensive experiments demonstrate that the MMHT model exhibits competitive performance in comparison with that of other state-of-the-art methods. Overall, our results highlight the effectiveness of the MMHT model in terms of addressing the challenges faced in visual object tracking tasks.

Cognitive deficits in human ApoE4 knock-in mice: A systematic review and meta-analysis.

Apolipoprotein-E4 (ApoE4) is an important genetic risk factor for Alzheimer's disease. The development of targeted-replacement human ApoE knock-in mice facilitates research into mechanisms by which ApoE4 affects the brain. We performed meta-analyses and meta-regression analyses to examine differences in cognitive performance between ApoE4 and ApoE3 mice. We included 61 studies in which at least one of the following tests was assessed: Morris Water Maze (MWM), novel object location (NL), novel object recognition (NO) and Fear Conditioning (FC) test. ApoE4 vs. ApoE3 mice performed significantly worse on the MWM (several outcomes, 0.17 ≤ g ≤ 0.60), NO (exploration, g=0.33; index, g=0.44) and FC (contextual, g=0.49). ApoE4 vs. ApoE3 differences were not systematically related to sex or age. We conclude that ApoE4 knock-in mice in a non-AD condition show some, but limited cognitive deficits, regardless of sex and age. These effects suggest an intrinsic vulnerability in ApoE4 mice that may become more pronounced under additional brain load, as seen in neurodegenerative diseases.

Identifying the nature of episodic memory deficits in Major Depressive Disorder using a Real-World What-Where-When task.

ABSTRACTDeficits in episodic memory have been reported in various psychiatric conditions, including Major Depressive Disorder (MDD). Many widely used episodic memory tests do not have the ability to distinguish between impaired memory of separate components of a real-life event (e.g., what happened, where it happened and when), and impaired binding of such real-life features. To address this issue, a naturalistic, real-world What-Where-When memory task was employed to assess the nature of episodic memory impairments in MDD. A validation study established that the task is sensitive to age-related episodic memory changes, and that intentional encoding does not invalidate the task. The main study then compared the performance of patients with depression and control participants on the intentionally encoded WWW task. Patients with MDD presented an overall episodic memory impairment arising from deficits in object memory and the ability to bind objects to temporal context. Taken together, our study confirms the episodic memory impairment in MDD, by providing evidence of deficient object memory and reduced ability to bind temporal context to objects in patients. Our naturalistic WWW task presents a promising approach for thorough identification of the nature of episodic memory impairments, under a real-world environment, in various conditions, including MDD.

The Effect of Experience on the Perception of Affordances for Aperture Crossing in Cycling.

To ride successfully and safely, cyclists must perceive and act on the affordances that are available in a given situation. This study investigated whether experience in perceiving and acting with respect to a person-plus-object system would influence whether and how a person choses to cross an aperture of different widths, especially in relation to the maximal action capabilities. We also explore whether the distribution of action modes reflects this effect. We examined the performance (i.e., the probability of successfully crossing the aperture) and the decision (i.e., the probability of attempting to cross the aperture) of 8 experienced cyclists and 16 occasional cyclists in an aperture crossing task. In term of performance, experienced cyclists demonstrated greater ability to cross narrower apertures than occasional cyclists, but there were no such differences when aperture width was scaled to maximal action capabilities. In term of decision, both experienced and occasional cyclists tended to over-estimate their abilities, but the experienced cyclists did so to a greater extent. Our findings indicate that experience improves the ability to perform more complex tasks due to utilizing a wider repertoire of actions, but not necessarily the ability to perceive and actualize (action-scaled) affordances."

A simple optical flow model explains why certain object viewpoints are special.

A core challenge in perception is recognizing objects across the highly variable retinal input that occurs when objects are viewed from different directions (e.g. front versus side views). It has long been known that certain views are of particular importance, but it remains unclear why. We reasoned that characterizing the computations underlying visual comparisons between objects could explain the privileged status of certain qualitatively special views. We measured pose discrimination for a wide range of objects, finding large variations in performance depending on the object and the viewing angle, with front and back views yielding particularly good discrimination. Strikingly, a simple and biologically plausible computational model based on measuring the projected three-dimensional optical flow between views of objects accurately predicted both successes and failures of discrimination performance. This provides a computational account of why certain views have a privileged status.

Enhanced Quantitative Wavefront Imaging for Nano-Object Characterization.

Quantitative phase imaging enables precise and label-free characterizations of individual nano-objects within a large volume, without a priori knowledge of the sample or imaging system. While emerging common path implementations are simple enough to promise a broad dissemination, their phase sensitivity still falls short of precisely estimating the mass or polarizability of vesicles, viruses, or nanoparticles in single-shot acquisitions. In this paper, we revisit the Zernike filtering concept, originally crafted for intensity-only detectors, with the aim of adapting it to wavefront imaging. We demonstrate, through numerical simulation and experiments based on high-resolution wavefront sensing, that a simple Fourier-plane add-on can significantly enhance phase sensitivity for subdiffraction objects─achieving over an order of magnitude increase (×12)─while allowing the quantitative retrieval of both intensity and phase. This advancement allows for more precise nano-object detection and metrology.

A multiscale 3D network for lung nodule detection using flexible nodule modeling.

BACKGROUND: Lung cancer is the most common type of cancer. Detection of lung cancer at an early stage can reduce mortality rates. Pulmonary nodules may represent early cancer and can be identified through computed tomography (CT) scans. Malignant risk can be estimated based on attributes like size, shape, location, and density. PURPOSE: Deep learning algorithms have achieved remarkable advancements in this domain compared to traditional machine learning methods. Nevertheless, many existing anchor-based deep learning algorithms exhibit sensitivity to predefined anchor-box configurations, necessitating manual adjustments to obtain optimal outcomes. Conversely, current anchor-free deep learning-based nodule detection methods normally adopt fixed-size nodule models like cubes or spheres. METHODS: To address these technical challenges, we propose a multiscale 3D anchor-free deep learning network (M3N) for pulmonary nodule detection, leveraging adjustable nodule modeling (ANM). Within this framework, ANM empowers the representation of target objects in an anisotropic manner, with a novel point selection strategy (PSS) devised to accelerate the learning process of anisotropic representation. We further incorporate a composite loss function that combines the conventional L2 loss and cosine similarity loss, facilitating M3N to learn nodules' intensity distribution in three dimensions. RESULTS: Experiment results show that the M3N achieves 90.6% competitive performance metrics (CPM) with seven predefined false positives per scan on the LUNA 16 dataset. This performance appears to exceed that of other state-of-the-art deep learning-based networks reported in their respective publications. Individual test results also demonstrate that M3N excels in providing more accurate, adaptive bounding boxes surrounding the contours of target nodules. CONCLUSIONS: The newly developed nodule detection system reduces reliance on prior knowledge, such as the general size of objects in the dataset, thus it should enhance overall robustness and versatility. Distinct from traditional nodule modeling techniques, the ANM approach aligns more closely with the morphological characteristics of nodules. Time consumption and detection results demonstrate promising efficiency and accuracy which should be validated in clinical settings.

Factorized visual representations in the primate visual system and deep neural networks.

Object classification has been proposed as a principal objective of the primate ventral visual stream and has been used as an optimization target for deep neural network models (DNNs) of the visual system. However, visual brain areas represent many different types of information, and optimizing for classification of object identity alone does not constrain how other information may be encoded in visual representations. Information about different scene parameters may be discarded altogether ('invariance'), represented in non-interfering subspaces of population activity ('factorization') or encoded in an entangled fashion. In this work, we provide evidence that factorization is a normative principle of biological visual representations. In the monkey ventral visual hierarchy, we found that factorization of object pose and background information from object identity increased in higher-level regions and strongly contributed to improving object identity decoding performance. We then conducted a large-scale analysis of factorization of individual scene parameters - lighting, background, camera viewpoint, and object pose - in a diverse library of DNN models of the visual system. Models which best matched neural, fMRI, and behavioral data from both monkeys and humans across 12 datasets tended to be those which factorized scene parameters most strongly. Notably, invariance to these parameters was not as consistently associated with matches to neural and behavioral data, suggesting that maintaining non-class information in factorized activity subspaces is often preferred to dropping it altogether. Thus, we propose that factorization of visual scene information is a widely used strategy in brains and DNN models thereof.

When looking at a picture, we can quickly identify a recognizable object, such as an apple, applying a single word label to it. Although extensive neuroscience research has focused on how human and monkey brains achieve this recognition, our understanding of how the brain and brain-like computer models interpret other complex aspects of a visual scene ­ such as object position and environmental context ­ remains incomplete. In particular, it was not clear to what extent object recognition comes at the expense of other important scene details. For example, various aspects of the scene might be processed simultaneously. On the other hand, general object recognition may interfere with processing of such details. To investigate this, Lindsey and Issa analyzed 12 monkey and human brain datasets, as well as numerous computer models, to explore how different aspects of a scene are encoded in neurons and how these aspects are represented by computational models. The analysis revealed that preventing effective separation and retention of information about object pose and environmental context worsened object identification in monkey cortex neurons. In addition, the computer models that were the most brain-like could independently preserve the other scene details without interfering with object identification. The findings suggest that human and monkey high level ventral visual processing systems are capable of representing the environment in a more complex way than previously appreciated. In the future, studying more brain activity data could help to identify how rich the encoded information is and how it might support other functions like spatial navigation. This knowledge could help to build computational models that process the information in the same way, potentially improving their understanding of real-world scenes.

MRGazer: decoding eye gaze points from functional magnetic resonance imaging in individual space.

Eye-tracking research has proven valuable in understanding numerous cognitive functions. Recently, Frey et al. provided an exciting deep learning method for learning eye movements from functional magnetic resonance imaging (fMRI) data. It employed the multi-step co-registration of fMRI into the group template to obtain eyeball signal, and thus required additional templates and was time consuming. To resolve this issue, in this paper, we propose a framework named MRGazer for predicting eye gaze points from fMRI in individual space. The MRGazer consists of an eyeball extraction module and a residual network-based eye gaze prediction module. Compared to the previous method, the proposed framework skips the fMRI co-registration step, simplifies the processing protocol, and achieves end-to-end eye gaze regression. The proposed method achieved superior performance in eye fixation regression (Euclidean error, EE=2.04°) than the co-registration-based method (EE=2.89°)， and delivered objective results within a shorter time (~0.02 second/volume) than prior method (~0.3 second/volume). The code is available at https://github.com/ustc-bmec/MRGazer.

Pharmacological HDAC3 inhibition alters memory updating in young and old male mice.

Long-term memories are not stored in a stable state but must be flexible and dynamic to maintain relevance in response to new information. Existing memories are thought to be updated through the process of reconsolidation, in which memory retrieval initiates destabilization and updating to incorporate new information. Memory updating is impaired in old age, yet little is known about the mechanisms that go awry. One potential mechanism is the repressive histone deacetylase 3 (HDAC3), which is a powerful negative regulator of memory formation that contributes to age-related impairments in memory formation. Here, we tested whether HDAC3 also contributes to age-related impairments in memory updating using the Objects in Updated Locations (OUL) paradigm. We show that blocking HDAC3 immediately after updating with the pharmacological inhibitor RGFP966 ameliorated age-related impairments in memory updating in 18-m.o. male mice. Surprisingly, we found that post-update HDAC3 inhibition in young (3-m.o.) male mice had no effect on memory updating but instead impaired memory for the original information, suggesting that the original and updated information may compete for expression at test and HDAC3 helps regulate which information is expressed. To test this idea, we next assessed whether HDAC3 inhibition would improve memory updating in young male mice given a weak, subthreshold update. Consistent with our hypothesis, we found that HDAC3 blockade strengthened the subthreshold update without impairing memory for the original information, enabling balanced expression of the original and updated information. Together, this research suggests that HDAC3 may contribute to age-related impairments in memory updating and may regulate the strength of a memory update in young mice, shifting the balance between the original and updated information at test.

Night target detection algorithm based on improved YOLOv7.

Aiming at the problems of error detection and missing detection in night target detection, this paper proposes a night target detection algorithm based on YOLOv7(You Only Look Once v7). The algorithm proposed in this paper preprocesses images by means of square equalization and Gamma transform. The GSConv(Group Separable Convolution) module is introduced to reduce the number of parameters and the amount of calculation to improve the detection effect. ShuffleNetv2_×1.5 is introduced as the feature extraction Network to reduce the number of Network parameters while maintaining high tracking accuracy. The hard-swish activation function is adopted to greatly reduce the delay cost. At last, Scylla Intersection over Union function is used instead of Efficient Intersection over Union function to optimize the loss function and improve the robustness. Experimental results demonstrate that the average detection accuracy of the proposed improved YOLOv7 model is 88.1%. It can effectively improve the detection accuracy and accuracy of night target detection.

The real-time hand and object recognition for virtual interaction.

Recognizing hand-object interactions presents a significant challenge in computer vision. It arises due to the varying nature of hand-object interactions. Moreover, estimating the 3D position of a hand from a single frame can be problematic, especially when the hand obstructs the view of the object from the observer's perspective. In this article, we present a novel approach to recognizing objects and facilitating virtual interactions, using a steering wheel as an illustrative example. We propose a real-time solution for identifying hand-object interactions in eXtended reality (XR) environments. Our approach relies on data captured by a single RGB camera during a manipulation scenario involving a steering wheel. Our model pipeline consists of three key components: (a) a hand landmark detector based on the MediaPipe cross-platform hand tracking solution; (b) a three-spoke steering wheel model tracker implemented using the faster region-based convolutional neural network (Faster R-CNN) architecture; and (c) a gesture recognition module designed to analyze interactions between the hand and the steering wheel. This approach not only offers a realistic experience of interacting with steering-based mechanisms but also contributes to reducing emissions in the real-world environment. Our experimental results demonstrate the natural interaction between physical objects in virtual environments, showcasing precision and stability in our system.

Two fundamentally different mechanisms by which unconscious information impairs behavioral performance: Evidence from fMRI and computational modeling.

It is increasingly clear that unconscious information impairs the performance of the corresponding action when the instruction to act is delayed. However, whether this impairment occurs at the response level or at the perceptual level remains controversial. This study used fMRI and a computational model with a pre-post design to address this elusive issue. The fMRI results showed that when the unconscious information containing strong stimulus-response associations was irrelevant to subsequent stimuli, the precuneus in the parietal lobe, which is thought to be involved in sensorimotor processing, was activated. In contrast, when the unconscious information was relevant to subsequent stimuli, regardless of the strength of the stimulus-response associations, some regions in the occipital and temporal cortices, which are thought to be involved in visual perceptual processing, were activated. In addition, the percent signal change in the regions of interest associated with motor inhibition was modulated by compatibility in the irrelevant but not in the relevant stimuli conditions. Modeling of behavioral data further supported that the irrelevant and relevant stimuli conditions involved fundamentally different mechanisms. Our finding reconciles the debate about the mechanism by which unconscious information impairs action performance and has important implications for understanding of unconscious cognition.