Language-Guided 3-D Action Feature Learning Without Ground-Truth Sample Class Label.

This work pays the first research effort to leverage point cloud sequence-based Self-supervised 3-D Action Feature Learning (S3AFL), under text's cross-modality weak supervision. We intend to fill the huge performance gap between point cloud sequence and 3-D skeleton-based manners. The key intuition derives from the observation that skeleton-based manners actually hold the human pose's high-level knowledge that leads to attention on the body's joint-aware local parts.

Misclassification of causes of death among a small all-autopsied group of former nuclear workers: Death certificates vs. autopsy reports.

The U.S. Transuranium and Uranium Registries performs autopsies on each of its deceased Registrants as a part of its mission to follow up occupationally-exposed individuals.

Nanotoxicity of two-dimensional nanomaterials on human skin and the structural evolution of keratin protein.

Two-dimensional (2D) materials have been increasingly widely used in biomedical and cosmetical products nowadays, yet their safe usage in human body and environment necessitates a comprehensive understanding of their nanotoxicity. In this work, the effect of pristine graphene and graphene oxide (GO) on the adsorption and conformational changes of skin keratin using molecular dynamics simulations. It is found that skin keratin can be absorbed through various noncovalent driving forces, such as van der Waals (vdW) and electrostatics.

TaiChiNet: Negative-Positive Cross-Attention Network for Breast Lesion Segmentation in Ultrasound Images.

Breast lesion segmentation in ultrasound images is essential for computer-aided breast-cancer diagnosis. To improve the segmentation performance, most approaches design sophisticated deep-learning models by mining the patterns of foreground lesions and normal backgrounds simultaneously or by unilaterally enhancing foreground lesions via various focal losses. However, the potential of normal backgrounds is underutilized, which could reduce false positives by compacting the feature representation of all normal backgrounds.

Beyond Pattern Variance: Unsupervised 3-D Action Representation Learning With Point Cloud Sequence.

This work pays the first research effort to address unsupervised 3-D action representation learning with point cloud sequence, which is different from existing unsupervised methods that rely on 3-D skeleton information. Our proposition is built on the state-of-the-art 3-D action descriptor 3-D dynamic voxel (3DV) with contrastive learning (CL). The 3DV can compress the point cloud sequence into a compact point cloud of 3-D motion information.

DifFormer: Multi-Resolutional Differencing Transformer With Dynamic Ranging for Time Series Analysis.

Time series analysis is essential to many far-reaching applications of data science and statistics including economic and financial forecasting, surveillance, and automated business processing. Though being greatly successful of Transformer in computer vision and natural language processing, the potential of employing it as the general backbone in analyzing the ubiquitous times series data has not been fully released yet. Prior Transformer variants on time series highly rely on task-dependent designs and pre-assumed "pattern biases", revealing its insufficiency in representing nuanced seasonal, cyclic, and outlier patterns which are highly prevalent in time series.

Latent bone modelling for estimation of plutonium concentration in skeleton of former nuclear workers.

The skeleton is a major plutonium retention site in the human body. Estimation of the total plutonium activity in the skeleton is a challenging problem. For most tissue donors at the United States Transuranium and Uranium Registries, a limited number of bone samples is available.

Temporal Sentence Grounding in Videos: A Survey and Future Directions.

Temporal sentence grounding in videos (TSGV), a.k.a.

Data and Tools Integration in the Canadian Open Neuroscience Platform.

We present the Canadian Open Neuroscience Platform (CONP) portal to answer the research community's need for flexible data sharing resources and provide advanced tools for search and processing infrastructure capacity. This portal differs from previous data sharing projects as it integrates datasets originating from a number of already existing platforms or databases through DataLad, a file level data integrity and access layer. The portal is also an entry point for searching and accessing a large number of standardized and containerized software and links to a computing infrastructure.

GREnet: Gradually REcurrent Network With Curriculum Learning for 2-D Medical Image Segmentation.

Medical image segmentation is a vital stage in medical image analysis. Numerous deep-learning methods are booming to improve the performance of 2-D medical image segmentation, owing to the fast growth of the convolutional neural network. Generally, the manually defined ground truth is utilized directly to supervise models in the training phase.

Contrastive domain adaptation with consistency match for automated pneumonia diagnosis.

Pneumonia can be difficult to diagnose since its symptoms are too variable, and the radiographic signs are often very similar to those seen in other illnesses such as a cold or influenza. Deep neural networks have shown promising performance in automated pneumonia diagnosis using chest X-ray radiography, allowing mass screening and early intervention to reduce the severe cases and death toll. However, they usually require many well-labelled chest X-ray images for training to achieve high diagnostic accuracy.

RCT: Resource Constrained Training for Edge AI.

Efficient neural network training is essential for in situ training of edge artificial intelligence (AI) and carbon footprint reduction in general. Train neural network on the edge is challenging because there is a large gap between limited resources on edge and the resource requirement of current training methods. Existing training methods are based on the assumption that the underlying computing infrastructure has sufficient memory and energy supplies.

Efficient Spiking Neural Networks With Radix Encoding.

Spiking neural networks (SNNs) have advantages in latency and energy efficiency over traditional artificial neural networks (ANNs) due to their event-driven computation mechanism and the replacement of energy-consuming weight multiplication with addition. However, to achieve high accuracy, it usually requires long spike trains to ensure accuracy, usually more than 1000 time steps. This offsets the computation efficiency brought by SNNs because a longer spike train means a larger number of operations and larger latency.

Adversarial multimodal fusion with attention mechanism for skin lesion classification using clinical and dermoscopic images.

Accurate skin lesion diagnosis requires a great effort from experts to identify the characteristics from clinical and dermoscopic images. Deep multimodal learning-based methods can reduce intra- and inter-reader variability and improve diagnostic accuracy compared to the single modality-based methods. This study develops a novel method, named adversarial multimodal fusion with attention mechanism (AMFAM), to perform multimodal skin lesion classification.

A framework for assessing the impact of accelerated approval.

The FDA's Accelerated Approval program (AA) is a regulatory program to expedite availability of products to treat serious or life-threatening illnesses that lack effective treatment alternatives. Ideally, all of the many stakeholders such as patients, physicians, regulators, and health technology assessment [HTA] agencies that are affected by AA should benefit from it. In practice, however, there is intense debate over whether evidence supporting AA is sufficient to meet the needs of the stakeholders who collectively bring an approved product into routine clinical care.

Word2Pix: Word to Pixel Cross-Attention Transformer in Visual Grounding.

Current one-stage methods for visual grounding encode the language query as one holistic sentence embedding before fusion with visual features for target localization. Such a formulation provides insufficient ability to model query at the word level, and therefore is prone to neglect words that may not be the most important ones for a sentence but are critical for the referred object. In this article, we propose Word2Pix: a one-stage visual grounding network based on the encoder-decoder transformer architecture that enables learning for textual to visual feature correspondence via word to pixel attention.

EDCompress: Energy-Aware Model Compression for Dataflows.

Edge devices demand low energy consumption, cost, and small form factor. To efficiently deploy convolutional neural network (CNN) models on the edge device, energy-aware model compression becomes extremely important. However, existing work did not study this problem well because of the lack of considering the diversity of dataflow types in hardware architectures.

Trusted Multi-View Classification With Dynamic Evidential Fusion.

Existing multi-view classification algorithms focus on promoting accuracy by exploiting different views, typically integrating them into common representations for follow-up tasks. Although effective, it is also crucial to ensure the reliability of both the multi-view integration and the final decision, especially for noisy, corrupted and out-of-distribution data. Dynamically assessing the trustworthiness of each view for different samples could provide reliable integration.

Enhanced Stability and Mechanical Properties of a Graphene-Protein Nanocomposite Film by a Facile Non-Covalent Self-Assembly Approach.

Graphene-based nanocomposite films (NCFs) are in high demand due to their superior photoelectric and thermal properties, but their stability and mechanical properties form a bottleneck. Herein, a facile approach was used to prepare nacre-mimetic NCFs through the non-covalent self-assembly of graphene oxide (GO) and biocompatible proteins. Various characterization techniques were employed to characterize the as-prepared NCFs and to track the interactions between GO and proteins.

Plutonium in Manhattan Project workers: Using autopsy data to evaluate organ content and dose estimates based on urine bioassay with implications for radiation epidemiology.

Purpose: Radiation dose estimates in epidemiology typically rely on intake predictions based on urine bioassay measurements. The purpose of this article is to compare the conventional dosimetric estimates for radiation epidemiology with the estimates based on additional post-mortem tissue radiochemical analysis results.

Methods: The comparison was performed on a unique group of 11 former Manhattan Project nuclear workers, who worked with plutonium in the 1940s, and voluntarily donated their bodies to the United States Transuranium and Uranium Registries.

Deep Supervised Domain Adaptation for Pneumonia Diagnosis From Chest X-Ray Images.

Pneumonia is one of the most common treatable causes of death, and early diagnosis allows for early intervention. Automated diagnosis of pneumonia can therefore improve outcomes. However, it is challenging to develop high-performance deep learning models due to the lack of well-annotated data for training.

Point Adversarial Self-Mining: A Simple Method for Facial Expression Recognition.

In this article, we propose a simple yet effective approach, called point adversarial self mining (PASM), to improve the recognition accuracy in facial expression recognition (FER). Unlike previous works focusing on designing specific architectures or loss functions to solve this problem, PASM boosts the network capability by simulating human learning processes: providing updated learning materials and guidance from more capable teachers. Specifically, to generate new learning materials, PASM leverages a point adversarial attack method and a trained teacher network to locate the most informative position related to the target task, generating harder learning samples to refine the network.

Triply Complementary Priors for Image Restoration.

Recent works that utilized deep models have achieved superior results in various image restoration (IR) applications. Such approach is typically supervised, which requires a corpus of training images with distributions similar to the images to be recovered. On the other hand, the shallow methods, which are usually unsupervised remain promising performance in many inverse problems, e.

Deep Spectral Representation Learning From Multi-View Data.

Multi-view representation learning (MvRL) aims to learn a consensus representation from diverse sources or domains to facilitate downstream tasks such as clustering, retrieval, and classification. Due to the limited representative capacity of the adopted shallow models, most existing MvRL methods may yield unsatisfactory results, especially when the labels of data are unavailable. To enjoy the representative capacity of deep learning, this paper proposes a novel multi-view unsupervised representation learning method, termed as Multi-view Laplacian Network (MvLNet), which could be the first deep version of the multi-view spectral representation learning method.