WhACC: Whisker Automatic Contact Classifier with Expert Human-Level Performance

This paper proposes a novel hierarchical approach to improve the accuracy of the classification of normal-vs-abnormal frames in white-light colonoscopy videos. The existing approaches label each frame independently, without considering the temporal consistency between adjacent frames. Temporal consistency, however, can improve the classification accuracy in the presence of unclear/uncertain images. We propose to leverage temporal consistency between adjacent frames for colonoscopy video frame classification using a novel hierarchical classifier. Comparative experiments with five challenging full colonoscopy videos show that the proposed approach considerably improves the mean class normal/abnormal classification accuracy compared to the approaches where the frames are classified independently.

A meta-analysis of the effects of aging on motor cortex neurophysiology assessed by transcranial magnetic stimulation.

Inherited retinal degeneration (IRD) comprises a diverse group of monogenic disorders characterized by marked genetic and phenotypic heterogeneity. Although next-generation sequencing (NGS) enables the identification of candidate variants, many remain classified as variants of uncertain significance (VUS). Ancestry-matched population data can strengthen comparative evidence, and the emergence of national biobanks provides new opportunities to operationalize ACMG/AMP criterion PS4 through case-control analyses. We integrated an IRD cohort of 802 probands with whole-genome allele frequency data from 1,492 individuals in the Taiwan Biobank. An allele-based case-control framework was applied, assigning PS4 when the Haldane-Anscombe-corrected odds ratio was ≥ 5 and the 95% confidence interval excluded 1. Post-PS4 triage required variants to: (i) reside in IRD-associated genes, (ii) be rare in East Asian populations in gnomAD v4.1, and (iii) be annotated in RefSeq as exonic, untranslated regions, or splicing (± 20bp). Baseline ACMG/AMP classifications were generated using GeneBe and finalized through expert curation. Incorporation of PS4 substantially refined variant interpretation, upgrading two variants from Likely Pathogenic to Pathogenic and six from VUS to Likely Pathogenic. Homozygous exemplar variants, including CNGB1 (NM_001297.5): c.2921T > G and CFAP410 (NM_004928.3): c.340_351dup, demonstrated strong genotype-phenotype concordance with confirmatory sequencing, illustrating an end-to-end workflow from statistical enrichment to clinical reporting. An ancestry-aware case-control framework enables effective implementation of PS4 and improves the accuracy of IRD variant classification. This reproducible strategy supports the integration of population-specific genomic data into clinical workflows and is applicable to other monogenic disorders.

The integration of sensory inputs in the motor cortex is crucial for dexterous movement. We recently demonstrated that a closed‐loop control based on the feedback provided through intraneural multichannel electrodes implanted in the median and ulnar nerves of a participant with upper limb amputation improved manipulation skills and increased prosthesis embodiment. Here we assessed, in the same participant, whether and how selective intraneural sensory stimulation also elicits a measurable cortical activation and affects sensorimotor cortical circuits. After estimating the activation of the primary somatosensory cortex evoked by intraneural stimulation, sensorimotor integration was investigated by testing the inhibition of primary motor cortex (M1) output to transcranial magnetic stimulation, after both intraneural and perineural stimulation. Selective sensory intraneural stimulation evoked a low‐amplitude, 16 ms‐latency, parietal response in the same area of the earliest component evoked by whole‐nerve stimulation, compatible with fast‐conducting afferent fibre activation. For the first time, we show that the same intraneural stimulation was also capable of decreasing M1 output, at the same time range of the short‐latency afferent inhibition effect of whole‐nerve superficial stimulation. The inhibition generated by the stimulation of channels activating only sensory fibres was stronger than that due to intraneural or perineural stimulation of channels activating mixed fibres. We demonstrate in a human subject that the cortical sensorimotor integration inhibiting M1 output previously described after the experimental whole‐nerve stimulation is present also with a more ecological selective sensory fibre stimulation.Key pointsCortical integration of sensory inputs is crucial for dexterous movement.Short‐latency somatosensory afferent inhibition of motor cortical output is typically produced by peripheral whole‐nerve stimulation.We exploited intraneural multichannel electrodes used to provide sensory feedback for prosthesis control to assess whether and how selective intraneural sensory stimulation affects sensorimotor cortical circuits in humans.Activation of the primary somatosensory cortex (S1) was explored by recording scalp somatosensory evoked potentials. Sensorimotor integration was tested by measuring the inhibitory effect of the afferent stimulation on the output of the primary motor cortex (M1) generated by transcranial magnetic stimulation.We demonstrate in humans that selective intraneural sensory stimulation elicits a measurable activation of S1 and that it inhibits the output of M1 at the same time range of whole‐nerve superficial stimulation.

Previous studies indicated that sensorimotor integration and plasticity of the sensorimotor system are impaired in dystonia patients. We investigated motor evoked potential amplitudes and short latency afferent inhibition to examine corticospinal excitability and cortical sensorimotor integration, before and after inhibitory 1Hz repetitive transcranial magnetic stimulation over primary sensory and primary motor cortex in patients with cervical dystonia (n=12). Motor evoked potentials were recorded from the right first dorsal interosseous muscle after application of unconditioned transcranial magnetic test stimuli and after previous conditioning electrical stimulation of the right index finger at short interstimulus intervals of 25, 30 and 40ms. Results were compared to a group of healthy age-matched controls. At baseline, motor evoked potential amplitudes did not differ between groups. Short latency afferent inhibition was reduced in cervical dystonia patients compared to healthy controls. Inhibitory 1Hz sensory cortex repetitive transcranial magnetic stimulation but not motor cortex repetitive transcranial magnetic stimulation increased motor evoked potential amplitudes in cervical dystonia patients. Additionally, both 1Hz repetitive transcranial magnetic stimulation over primary sensory and primary motor cortex normalized short latency afferent inhibition in these patients. In healthy subjects, sensory repetitive transcranial magnetic stimulation had no influence on motor evoked potential amplitudes and short latency afferent inhibition. Plasticity of sensorimotor circuits is altered in cervical dystonia patients.

We proposed an innovative end-to-end generative adversarial network structure (CRVC-GAN) for video colorization, which is dedicated to solving the temporal consistency and coloring quality problems in the video coloring domain. For the temporal consistency problem, we utilize the network structure of recurrent neural network (RNN) because the model shape of RNN does not change with the input length and the calculation takes into account the history information to ensure the degree of correlation between successive frames. This better ensures the continuity between frames, so the consecutive frames get good temporal consistency in coloring. For the coloring quality problem, we utilize the UNet++ coding method to decode the video frames, integrating different levels of features to improve accuracy, and then enhance the object boundary information in the video frames, thus improving the coloring quality of grayscale video frames. In addition, we innovatively utilize the style transformation loss function in terms of images for model training to ensure temporal consistency and coloring quality by exploiting the slight difference in style variation from frame to frame. The experimental results show that, compared with the existing methods and through its own designed balance index, the colorization and temporary consistency balance index, CRVC-GAN obtained certain advantages on the DAVIS and Videvo public datasets with the results of 0.56176 and 0.74286, respectively, and achieved a better balance between time consistency and color quality.

Human pose estimation from image or video is a basic issue in computer graphics and computer vision. The challenge of human pose estimation in video lies in the temporal coherency issue. The temporal consistency in video is the contents’ similarity shown in the video frames. In video, temporal consistency maintenance of human pose estimation is to obtain better long-term consistency. Great major methods for the long-term consistency are using the whole video optimization method, which makes very large computation and the absence of consistency before and after the articulated limbs. In this paper, a novel method for the maintenance of temporal consistency is proposed. We maintain the temporal consistency of the video by the structured space learning and halfway temporal evaluation methods. We adopt a three-stage multi-feature deep convolution network framework to generate the initial posture joints position data, and a long-term temporal coherence is propagated to the overall video at each stage. The long-term consistency is more appealing since it produces stable results over larger periods of time. Our method can achieve good temporal consistency and get accurate and stable human pose estimation results. Various experimental results demonstrated the superiority of our method.

Signal detection theory (SDT) is a widely used theoretical framework that describes how variable sensory signals are integrated with a decision criterion to support perceptual decision-making. SDT provides two key measurements: sensitivity (d') and bias (c), which reflect the separability of decision variable distributions (signal and noise) and the position of the decision criterion relative to optimal, respectively. Although changes in the subject's decision criterion can be reflected in changes in bias, decision criterion placement is not the sole contributor to measured bias. Indeed, neuronal representations of bias have been observed in sensory areas, suggesting that some changes in bias are because of effects on sensory encoding. To directly test whether the sensory encoding process can influence bias, we optogenetically manipulated neuronal excitability in primary visual cortex (V1) in mice of both sexes during either an orientation discrimination or a contrast detection task. Increasing excitability in V1 significantly decreased behavioral bias, whereas decreasing excitability had the opposite effect. To determine whether this change in bias is consistent with effects on sensory encoding, we made extracellular recordings from V1 neurons in passively viewing mice. Indeed, we found that optogenetic manipulation of excitability shifted the neuronal bias in the same direction as the behavioral bias. Moreover, manipulating the quality of V1 encoding by changing stimulus contrast or interstimulus interval also resulted in consistent changes in both behavioral and neuronal bias. Thus, changes in sensory encoding are sufficient to drive changes in bias measured using SDT.SIGNIFICANCE STATEMENT Perceptual decision-making involves sensory integration followed by application of a cognitive criterion. Using signal detection theory, one can extract features of the underlying decision variables and rule: sensitivity (d') and bias (c). Because bias is measured as the difference between the optimal and actual criterion, it is sensitive to both the sensory encoding processes and the placement of the decision criterion. Here, we use behavioral and electrophysiological approaches to demonstrate that measures of bias depend on sensory processes. Optogenetic manipulations of V1 in mice bidirectionally affect both behavioral and neuronal measures of bias with little effect on sensitivity. Thus, changes in sensory encoding influence bias, and the absence of changes in sensitivity do not preclude changes in sensory encoding.

Combined Statistical and Multiscale View on Ultrasonic Liver Images for Characterization

Video decolorization is the process of transferring three-channel color videos into single-channel grayscale videos, which is essentially the decolorization operation of video frames. Most existing video decolorization algorithms directly apply image decolorization methods to decolorize video frames. However, if we only take the single-frame decolorization result into account, it will inevitably cause temporal inconsistency and flicker phenomenon meaning that the same local content between continuous video frames may display different gray values. In addition, there are often similar local content features between video frames, which indicates redundant information. To solve the preceding problems, this article proposes a novel video decolorization algorithm based on the convolutional neural network and the long short-term memory neural network. First, we design a local semantic content encoder to learn and extract the same local content of continuous video frames, which can better preserve the contrast of video frames. Second, a temporal feature controller based on the bi-directional recurrent neural networks with Long short-term memory units is employed to refine the local semantic features, which can greatly maintain temporal consistency of the video sequence to eliminate the flicker phenomenon. Finally, we take advantages of deconvolution to decode the features to produce the grayscale video sequence. Experiments have indicated that our method can better preserve the local contrast of video frames and the temporal consistency over the state of the-art.

Modulation of sensorimotor circuits during retrieval of negative Autobiographical Memories: Exploring the impact of personality dimensions

The cerebral cortex provides sensorimotor integration and coordination during motor control of daily functional activities. Power spectrum density based on electroencephalography (EEG) has been employed as an approach that allows an investigation of the spatial–temporal characteristics of neuromuscular modulation; however, the biofeedback mechanism associated with cortical activation during motor control remains unclear among elderly individuals. Thirty one community-dwelling elderly participants were divided into low fall-risk potential (LF) and high fall-risk potential (HF) groups based upon the results obtained from a receiver operating characteristic analysis of the ellipse area of the center of pressure. Electroencephalography (EEG) was performed while the participants stood on a 6-degree-of-freedom Stewart platform, which generated continuous perturbations and done either with or without the virtual reality scene. The present study showed that when there was visual stimulation and poor somatosensory coordination, a higher level of cortical response was activated in order to keep postural balance. The elderly participants in the LF group demonstrated a significant and strong correlation between postural-related cortical regions; however, the elderly individuals in the HF group did not show such a relationship. Moreover, we were able to clarify the roles of various brainwave bands functioning in motor control. Specifically, the gamma and beta bands in the parietal–occipital region facilitate the high-level cortical modulation and sensorimotor integration, whereas the theta band in the frontal–central region is responsible for mediating error detection during perceptual motor tasks. Finally, the alpha band is associated with processing visual challenges in the occipital lobe.With a variety of motor control demands, increment in brainwave band coordination is required to maintain postural stability. These investigations shed light on the cortical modulation of motor control among elderly participants with varying fall-risk potentials. The results suggest that, although elderly adults may be without neurological deficits, inefficient central modulation during challenging postural conditions could be an internal factor that contributes to the risk of fall. Furthermore, training that helps to improve coordinated sensorimotor integration may be a useful approach to reduce the risk of fall among elderly populations or when patients suffer from neurological deficits.

Introduction: Musicians are an excellent example of people executing the perfect intended movements. Timing, sequencing, and precision are essential for purposeful actions. While playing an instrument, musicians must make motor adjustments to perform flawless music. These motor adjustments are made through auditory, tactile, and proprioceptive feedback. This study sought to investigate the cortical excitability and sensory-motor interaction in healthy professional musicians utilizing transcranial magnetic stimulation (TMS). Subjects and Methods: Fifteen healthy professional musicians and equal number of age and sex-matched control volunteers participated in this study. Motor thresholds (at both 200 μV and 1 mV), short-interval intracortical inhibition (SICI) (within interstimulus intervals of 1–7 ms), and short-latency afferent inhibition (SAI) (within interstimulus intervals of 16–24 ms) were recorded in the dominant and nondominant flexor carpi ulnaris muscles. Results: Analysis revealed no significant differences in RMT, SICI, or SAI between healthy musicians and controls across both dominant and nondominant sides. Conclusion: Motor cortex excitability and sensorimotor integration were comparable between healthy musicians and nonmusicians. TMS offers insights into motor system function, yet voluntary movement and TMS-induced motor responses engage different neurobiological pathways. Structural and functional alterations in the sensorimotor integration of healthy musicians may not be detectable through the SAI paradigm. Thus, TMS might not effectively detect alterations in cortical excitability or sensorimotor integration resulting from long-term instrument playing.

Human video motion transfer (HVMT) aims to synthesize videos that one person imitates other persons' actions. Although existing GAN-based HVMT methods have achieved great success, they either fail to preserve appearance details due to the loss of spatial consistency between synthesized and exemplary images, or generate incoherent video results due to the lack of temporal consistency among video frames. In this paper, we propose Coarse-to-Fine Flow Warping Network (C2F-FWN) for spatial-temporal consistent HVMT. Particularly, C2F-FWN utilizes coarse-to-fine flow warping and Layout-Constrained Deformable Convolution (LC-DConv) to improve spatial consistency, and employs Flow Temporal Consistency (FTC) Loss to enhance temporal consistency. In addition, provided with multi-source appearance inputs, C2F-FWN can support appearance attribute editing with great flexibility and efficiency. Besides public datasets, we also collected a large-scale HVMT dataset named SoloDance for evaluation. Extensive experiments conducted on our SoloDance dataset and the iPER dataset show that our approach outperforms state-of-art HVMT methods in terms of both spatial and temporal consistency. Source code and the SoloDance dataset are available at https://github.com/wswdx/C2F-FWN.

To examine the effects of aging and physical activity on distribution patterns in subcutaneous and visceral fat. Distributions of subcutaneous rat mass at six segments (face and neck, forearm, upper arm, trunk, thigh, and lower leg) were determined by adipose tissue thickness measurements by B mode ultrasonogram and body surface areas. Visceral fat mass was calculated by subtracting subcutaneous fat mass from the total fat mass determined hydrodensitometrically. Measurements were made on young and middle aged, trained and sedentary women (four groups). Per cent body fat was lower in trained than in sedentary individuals, both in the young and the middle aged subjects. The distribution of subcutaneous fat mass differed between sedentary and trained women. Trained young women had a reduced subcutaneous fat mass compared to sedentary young subjects in all segments except face and neck; the disparity between middle aged sedentary and trained women was limited to upper arm and trunk (P < 0.01 each), with no significant difference in face and neck, forearm, and lower limb segments. Differences in visceral fat mass between sedentary and trained subjects were similar for young and middle aged women (young, 2.5 v 3.7 kg; middle aged, 4.0 v 6.5 kg). Women who exercise regularly appear to accumulate less adipose tissue, especially in upper arm and trunk segments as they get older, with visceral fat mass remaining lower than in sedentary individuals.