HydrogelFinder: A Foundation Model for Efficient Self-Assembling Peptide Discovery Guided by Non-Peptidal Small Molecules.

Self-assembling peptides have numerous applications in medicine, food chemistry, and nanotechnology. However, their discovery has traditionally been serendipitous rather than driven by rational design. Here, HydrogelFinder, a foundation model is developed for the rational design of self-assembling peptides from scratch. This model explores the self-assembly properties by molecular structure, leveraging 1,377 self-assembling non-peptidal small molecules to navigate chemical space and improve structural diversity. Utilizing HydrogelFinder, 111 peptide candidates are generated and synthesized 17 peptides, subsequently experimentally validating the self-assembly and biophysical characteristics of nine peptides ranging from 1-10 amino acids-all achieved within a 19-day workflow. Notably, the two de novo-designed self-assembling peptides demonstrated low cytotoxicity and biocompatibility, as confirmed by live/dead assays. This work highlights the capacity of HydrogelFinder to diversify the design of self-assembling peptides through non-peptidal small molecules, offering a powerful toolkit and paradigm for future peptide discovery endeavors.

Neural Network Compression Based on Tensor Ring Decomposition.

Deep neural networks (DNNs) have made great breakthroughs and seen applications in many domains. However, the incomparable accuracy of DNNs is achieved with the cost of considerable memory consumption and high computational complexity, which restricts their deployment on conventional desktops and portable devices. To address this issue, low-rank factorization, which decomposes the neural network parameters into smaller sized matrices or tensors, has emerged as a promising technique for network compression. In this article, we propose leveraging the emerging tensor ring (TR) factorization to compress the neural network. We investigate the impact of both parameter tensor reshaping and TR decomposition (TRD) on the total number of compressed parameters. To achieve the maximal parameter compression, we propose an algorithm based on prime factorization that simultaneously identifies the optimal tensor reshaping and TRD. In addition, we discover that different execution orders of the core tensors result in varying computational complexities. To identify the optimal execution order, we construct a novel tree structure. Based on this structure, we propose a top-to-bottom splitting algorithm to schedule the execution of core tensors, thereby minimizing computational complexity. We have performed extensive experiments using three kinds of neural networks with three different datasets. The experimental results demonstrate that, compared with the three state-of-the-art algorithms for low-rank factorization, our algorithm can achieve better performance with much lower memory consumption and lower computational complexity.

FARN: Fetal Anatomy Reasoning Network for Detection with Global Context Semantic and Local Topology Relationship.

Accurate recognition of fetal anatomical structure is a pivotal task in ultrasound (US) image analysis. Sonographers naturally apply anatomical knowledge and clinical expertise to recognizing key anatomical structures in complex US images. However, mainstream object detection approaches usually treat each structure recognition separately, overlooking anatomical correlations between different structures in fetal US planes. In this work, we propose a Fetal Anatomy Reasoning Network (FARN) that incorporates two kinds of relationship forms: a global context semantic block summarized with visual similarity and a local topology relationship block depicting structural pair constraints. Specifically, by designing the Adaptive Relation Graph Reasoning (ARGR) module, anatomical structures are treated as nodes, with two kinds of relationships between nodes modeled as edges. The flexibility of the model is enhanced by constructing the adaptive relationship graph in a data-driven way, enabling adaptation to various data samples without the need for predefined additional constraints. The feature representation is further enhanced by aggregating the outputs of the ARGR module. Comprehensive experimental results demonstrate that FARN achieves promising performance in detecting 37 anatomical structures across key US planes in tertiary obstetric screening. FARN effectively utilizes key relationships to improve detection performance, demonstrates robustness to small-scale, similar, and indistinct structures, and avoids some detection errors that deviate from anatomical norms. Overall, our study serves as a resource for developing efficient and concise approaches to model inter-anatomy relationships.

Toward Interpreting the Thermally Activated ß Dynamics in Metallic Glass Using the Structural Constraint Neural Network.

It is crucial to unravel the structural factors influencing the dynamics of the amorphous solids. Deep learning aids in navigating these complexities, while transparency issues persist. Drawing inspiration from the successful application of prototype neural networks in image analysis, this study introduces a novel machine learning approach to address interpretability challenges in glassy research. Distinguishing from traditional machine learning models, the proposed neural network tries to learn distant structural motifs for solid-like atoms and liquid-like atoms. Such learned structural motifs constrain the underlying structural space and thus can serve as a breakthrough in explaining how structural differences impact dynamics. We further used the proposed model to explore the correlation between the local structure and activation energy in the CuZr alloys. Building upon this interpretable model, we demonstrated significant structural differences among atoms with different activation energies. Our interpretable model is a data-driven solution that provides a pathway to reveal the origin of structural heterogeneity in amorphous alloys.

HFSCCD: A Hybrid Neural Network for Fetal Standard Cardiac Cycle Detection in Ultrasound Videos.

In the fetal cardiac ultrasound examination, standard cardiac cycle (SCC) recognition is the essential foundation for diagnosing congenital heart disease. Previous studies have mostly focused on the detection of adult CCs, which may not be applicable to the fetus. In clinical practice, localization of SCCs needs to recognize end-systole (ES) and end-diastole (ED) frames accurately, ensuring that every frame in the cycle is a standard view. Most existing methods are not based on the detection of key anatomical structures, which may not recognize irrelevant views and background frames, results containing non-standard frames, or even it does not work in clinical practice. We propose an end-to-end hybrid neural network based on an object detector to detect SCCs from fetal ultrasound videos efficiently, which consists of 3 modules, namely Anatomical Structure Detection (ASD), Cardiac Cycle Localization (CCL), and Standard Plane Recognition (SPR). Specifically, ASD uses an object detector to identify 9 key anatomical structures, 3 cardiac motion phases, and the corresponding confidence scores from fetal ultrasound videos. On this basis, we propose a joint probability method in the CCL to learn the cardiac motion cycle based on the 3 cardiac motion phases. In SPR, to reduce the impact of structure detection errors on the accuracy of the standard plane recognition, we use XGBoost algorithm to learn the relation knowledge of the detected anatomical structures. We evaluate our method on the test fetal ultrasound video datasets and clinical examination cases and achieve remarkable results. This study may pave the way for clinical practices.

MLMSeg: A multi-view learning model for ultrasound thyroid nodule segmentation.

Accurate segmentation of the thyroid gland in ultrasound images is an essential initial step in distinguishing between benign and malignant nodules, thus facilitating early diagnosis. Most existing deep learning-based methods to segment thyroid nodules are learned from only a single view or two views, which limits the performance of segmenting nodules at different scales in complex ultrasound scanning environments. To address this limitation, this study proposes a multi-view learning model, abbreviated as MLMSeg. First, a deep convolutional neural network is introduced to encode the features of the local view. Second, a multi-channel transformer module is designed to capture long-range dependency correlations of global view between different nodules. Third, there are semantic relationships of structural view between features of different layers. For example, low-level features and high-level features are endowed with hidden relationships in the feature space. To this end, a cross-layer graph convolutional module is proposed to adaptively learn the correlations of high-level and low-level features by constructing graphs across different layers. In addition, in the view fusion, a channel-aware graph attention block is devised to fuse the features from the aforementioned views for accurate segmentation of thyroid nodules. To demonstrate the effectiveness of the proposed method, extensive comparative experiments were conducted with 14 baseline methods. MLMSeg achieved higher Dice coefficients (92.10% and 83.84%) and Intersection over Union scores (86.60% and 73.52%) on two different thyroid datasets. The exceptional segmentation capability of MLMSeg for thyroid nodules can greatly assist in localizing thyroid nodules and facilitating more precise measurements of their transverse and longitudinal diameters, which is of significant clinical relevance for the diagnosis of thyroid nodules.

Controllable van der Waals gaps by water adsorption.

Van der Waals (vdW) gaps with ångström-scale heights can confine molecules or ions to an ultimately small scale, providing an alternative way to tune material properties and explore microscopic phenomena. Modulation of the height of vdW gaps between two-dimensional (2D) materials is challenging due to the vdW interaction. Here we report a general approach to control the vdW gap by preadsorption of water molecules on the material surface. By controlling the saturation vapour pressure of water vapour, we can precisely control the adsorption level of water molecules and vary the height of the vdW gaps of MoS2 homojunctions from 5.5 Å to 53.6 Å. This technique can be further applied to other homo- and heterojunctions, constructing controlled vdW gaps in 2D artificial superlattices and in 2D/3D and 3D/3D heterojunctions. Engineering the vdW gap has great practical potential to modulate the device performance, as evidenced by the vdW-gap-dependent diode characteristics of the MoS2/gap/MoS2 junction. Our work introduces a general strategy of molecular preadsorption that can extend to various precursors, creating more tunability and variability in vdW material systems.

A knowledge-interpretable multi-task learning framework for automated thyroid nodule diagnosis in ultrasound videos.

Ultrasound has become the most widely used modality for thyroid nodule diagnosis, due to its portability, real-time feedback, lack of toxicity, and low cost. Recently, the computer-aided diagnosis (CAD) of thyroid nodules has attracted significant attention. However, most existing techniques can only be applied to either static images with prominent features (manually selected from scanning videos) or rely on 'black boxes' that cannot provide interpretable results. In this study, we develop a user-friendly framework for the automated diagnosis of thyroid nodules in ultrasound videos, by simulating the typical diagnostic workflow used by radiologists. This process consists of two orderly part-to-whole tasks. The first interprets the characteristics of each image using prior knowledge, to obtain corresponding frame-wise TI-RADS scores. Associated embedded representations not only provide diagnostic information for radiologists but also reduce computational costs. The second task models temporal contextual information in an embedding vector sequence and selectively enhances important information to distinguish benign and malignant thyroid nodules, thereby improving the efficiency and generalizability of the proposed framework. Experimental results demonstrated this approach outperformed other state-of-the-art video classification methods. In addition to assisting radiologists in understanding model predictions, these CAD results could further ease diagnostic workloads and improve patient care.

TransFSM: Fetal Anatomy Segmentation and Biometric Measurement in Ultrasound Images Using a Hybrid Transformer.

Biometric parameter measurements are powerful tools for evaluating a fetus's gestational age, growth pattern, and abnormalities in a 2D ultrasound. However, it is still challenging to measure fetal biometric parameters automatically due to the indiscriminate confusing factors, limited foreground-background contrast, variety of fetal anatomy shapes at different gestational ages, and blurry anatomical boundaries in ultrasound images. The performance of a standard CNN architecture is limited for these tasks due to the restricted receptive field. We propose a novel hybrid Transformer framework, TransFSM, to address fetal multi-anatomy segmentation and biometric measurement tasks. Unlike the vanilla Transformer based on a single-scale input, TransFSM has a deformable self-attention mechanism so it can effectively process multi-scale information to segment fetal anatomy with irregular shapes and different sizes. We devised a BAD to capture more intrinsic local details using boundary-wise prior knowledge, which compensates for the defects of the Transformer in extracting local features. In addition, a Transformer auxiliary segment head is designed to improve mask prediction by learning the semantic correspondence of the same pixel categories and feature discriminability among different pixel categories. Extensive experiments were conducted on clinical cases and benchmark datasets for anatomy segmentation and biometric measurement tasks. The experiment results indicate that our method achieves state-of-the-art performance in seven evaluation metrics compared with CNN-based, Transformer-based, and hybrid approaches. By Knowledge distillation, the proposed TransFSM can create a more compact and efficient model with high deploying potential in resource-constrained scenarios. Our study serves as a unified framework for biometric estimation across multiple anatomical regions to monitor fetal growth in clinical practice.

Multiagent-System-Based Attention Mechanism for Predicting Product Popularity: Handling Positive-Negative Diffusion Over Social Networks.

This brief is concerned with the prediction problem of product popularity under a social network (SN) with positive-negative diffusion (PND). First, a PND model is proposed to enable the simulation of product diffusion, and three user states are defined. Second, an optimal and precise feature vector of every user is extracted through a multi-agent-system-based attention mechanism (MASAM) that is devised. The weight matrix shared in the mechanism of all agents is learned using a distributed learning algorithm provided in MASAM. Third, an MAS model for product diffusion on SN is established based on the feature representations from MASAM. Rules for agent interaction during PND diffusion are suggested, which accelerate the simulation of information spread in SN. Finally, comprehensive experiments are conducted to verify the effectiveness and efficiency of the proposed models and algorithms in prediction and to compare their performance with baseline methods. Furthermore, a case study is provided to illustrate the applicability and extendibility of the developed algorithm.

A geometry-enhanced graph neural network for learning the smoothness of glassy dynamics from static structure.

Modeling the dynamics of glassy systems has been challenging in physics for several decades. Recent studies have shown the efficacy of Graph Neural Networks (GNNs) in capturing particle dynamics from the graph structure of glassy systems. However, current GNN methods do not take the dynamic patterns established by neighboring particles explicitly into account. In contrast to these approaches, this paper introduces a novel dynamical parameter termed "smoothness" based on the theory of graph signal processing, which explores the dynamic patterns from a graph perspective. Present graph-based approaches encode structural features without considering smoothness constraints, leading to a weakened correlation between structure and dynamics, particularly on short timescales. To address this limitation, we propose a Geometry-enhanced Graph Neural Network (Geo-GNN) to learn the smoothness of dynamics. Results demonstrate that our method outperforms state-of-the-art baselines in predicting glassy dynamics. Ablation studies validate the effectiveness of each proposed component in capturing smoothness within dynamics. These findings contribute to a deeper understanding of the interplay between glassy dynamics and static structure.

Deep endpoints focusing network under geometric constraints for end-to-end biometric measurement in fetal ultrasound images.

Biometric measurements in fetal ultrasound images are one of the most highly demanding medical image analysis tasks that can directly contribute to diagnosing fetal diseases. However, the natural high-speckle noise and shadows in ultrasound data present big challenges for automatic biometric measurement. Almost all the existing dominant automatic methods are two-stage models, where the key anatomical structures are segmented first and then measured, thus bringing segmentation and fitting errors. What is worse, the results of the second-stage fitting are completely dependent on the good performance of first-stage segmentation, i.e., the segmentation error will lead to a larger fitting error. To this end, we propose a novel end-to-end biometric measurement network, abbreviated as E2EBM-Net, that directly fits the measurement parameters. E2EBM-Net includes a cross-level feature fusion module to extract multi-scale texture information, a hard-soft attention module to improve position sensitivity, and center-focused detectors jointly to achieve accurate localizing and regressing of the measurement endpoints, as well as a loss function with geometric cues to enhance the correlations. To our knowledge, this is the first AI-based application to address the biometric measurement of irregular anatomical structures in fetal ultrasound images with an end-to-end approach. Experiment results showed that E2EBM-Net outperformed the existing methods and achieved the state-of-the-art performance.

Dual Attention Adversarial Attacks With Limited Perturbations.

The construction of undetectable adversarial examples with few perturbances remains a difficult problem in adversarial attacks. At present, most solutions use the standard gradient optimization algorithm to build adversarial examples by applying global perturbations to benign samples and then launch attacks on the targets (e.g., face recognition systems). However, when the perturbance size is limited, the performance of these approaches suffers substantially. The content of crucial places in an image, on the other hand, will impact the final prediction; if these areas can be investigated and limited perturbances introduced, an acceptable adversarial example will be constructed. Based on the foregoing research, this article offers a dual attention adversarial network (DAAN) to produce adversarial examples with limited perturbations. DAAN initially searches for effective areas in an input image using the spatial attention network and channel attention network, and then creates space and channel weights. Following that, these weights direct an encoder and a decoder to generate effective perturbation, which is then combined with the input to produce an adversarial example. Finally, the discriminator determines if the created adversarial examples are true or false, and the attacked model is utilized to determine whether the generated samples fit the attack targets. Extensive studies on various datasets show that DAAN not only delivers the best attack performance across all comparison algorithms with few perturbations, but it can also significantly improve the defensiveness of the attacked models.

Insights into electronic properties of strained two-dimensional semiconductors by out-of-plane bending.

Strain engineering is an important strategy to modulate the electronic and optical properties of two-dimensional (2D) semiconductors. In experiments, an effective and feasible method to induce strains on 2D semiconductors is the out-of-plane bending. However, in contrast to the in-plane methods, it will generate a combined strain effect on 2D semiconductors, which deserves further explorations. In this work, we theoretically investigate the carrier transport-related electronic properties of arsenene, antimonene, phosphorene, and MoS2under the out-of-plane bending. The bending effect can be disassembled into the in-plane and out-of-plane rolling strains. We find that the rolling always degrades the transport performance, while the in-plane strain could boost carrier mobilities by restraining the intervalley scattering. In other words, pursuing the maximum in-plane strain at the expense of minimum rolling should be the primary strategy to promote transports in 2D semiconductors through bending. Electrons in 2D semiconductors usually suffer from the serious intervalley scattering caused by optical phonons. The in-plane strain can break the crystal symmetry and separate nonequivalent energy valleys at band edges energetically, confining carrier transports at the Brillouin zone Γ point and eliminating the intervalley scattering. Investigation results show that the arsenene and antimonene are suitable for the bending technology, because of their small layer thicknesses which can relieve the rolling burden. Their electron and hole mobilities can be doubled simultaneously, compared with their unstrained 2D structures. From this study, the rules for the out-of-plane bending technology towards promoting transport abilities in 2D semiconductors are obtained.

Influence Maximization in Multiagent Systems by a Graph Embedding Method: Dealing With Probabilistically Unstable Links.

This article is concerned with the influence maximization (IM) problem under a network with probabilistically unstable links (PULs) via graph embedding for multiagent systems (MASs). First, two diffusion models, the unstable-link independent cascade (UIC) model and the unstable-link linear threshold (ULT) model, are designed for the IM problem under the network with PULs. Second, the MAS model for the IM problem with PULs is established and a series of interaction rules among agents are built for the MAS model. Third, the similarity of the unstable structure of the nodes is defined and a novel graph embedding method, termed the unstable-similarity2vec (US2vec) approach, is proposed to tackle the IM problem under the network with PULs. According to the embedding results of the US2vec approach, the seed set is figured out by the developed algorithm. Finally, extensive experiments are conducted to: 1) verify the validity of the proposed model and the developed algorithms and 2) illustrate the optimal solution for IM under different scenarios with PULs.

Traffic forecasting with graph spatial-temporal position recurrent network.

With the development of social economy and smart technology, the explosive growth of vehicles has caused traffic forecasting to become a daunting challenge, especially for smart cities. Recent methods exploit graph spatial-temporal characteristics, including constructing the shared patterns of traffic data, and modeling the topological space of traffic data. However, existing methods fail to consider the spatial position information and only utilize little spatial neighborhood information. To tackle above limitation, we design a Graph Spatial-Temporal Position Recurrent Network (GSTPRN) architecture for traffic forecasting. We first construct a position graph convolution module based on self-attention and calculate the dependence strengths among the nodes to capture the spatial dependence relationship. Next, we develop approximate personalized propagation that extends the propagation range of spatial dimension information to obtain more spatial neighborhood information. Finally, we systematically integrate the position graph convolution, approximate personalized propagation and adaptive graph learning into a recurrent network (i.e. Gated Recurrent Units). Experimental evaluation on two benchmark traffic datasets demonstrates that GSTPRN is superior to the state-of-art methods.

Fetal Ultrasound Standard Plane Detection With Coarse-to-Fine Multi-Task Learning.

The ultrasound standard plane plays an important role in prenatal fetal growth parameter measurement and disease diagnosis in prenatal screening. However, obtaining standard planes in a fetal ultrasound video is not only laborious and time-consuming but also depends on the clinical experience of sonographers to a certain extent. To improve the acquisition efficiency and accuracy of the ultrasound standard plane, we propose a novel detection framework that utilizes both the coarse-to-fine detection strategy and multi-task learning mechanism for feature-fused images. First, traditional manually-designed features and deep learning-based features are fused to obtain low-level shared features, which can enhance the model's feature expression ability. Inspired by the process of human recognition, ultrasound standard plane detection is divided into a coarse process of plane type classification and a fine process of standard-or-not detection, which is implemented via an end-to-end multi-task learning network. The region-of-interest area is also recognised in our detection framework to suppress the influence of a variable maternal background. Extensive experiments are conducted on three ultrasound planes of the first-class fetal examination, i.e., the femur, thalamus, and abdomen ultrasound images. The experiment results show that our method outperforms competing methods in terms of accuracy, which demonstrates the efficacy of the proposed method and can reduce the workload of sonographers in prenatal screening.

Interlayer Transition Induced Infrared Response in ReS2/2D Perovskite van der Waals Heterostructure Photodetector.

The emerging Ruddlesden-Popper two-dimensional perovskite (2D PVK) has recently joined the family of 2D semiconductors as a potential competitor for building van der Waals (vdW) heterostructures in future optoelectronics. However, to date, most of the reported heterostructures based on 2D PVKs suffer from poor spectral response that is caused by intrinsic wide bandgap of constituting materials. Herein, a direct heterointerface bandgap (∼0.4 eV) between 2D PVK and ReS2 is demonstrated. The strong interlayer coupling reduces the energy interval at the heterojunction region so that the heterostructure shows high sensitivity with the spectral response expanding to 2000 nm. The large type-II band offsets exceeding 1.1 eV ensure fast photogenerated carriers separation at the heterointerface. When this heterostructure is used as a self-driven photodetector, it exhibits a record high detectivity up to 1.8 × 1014 Jones, surpassing any reported 2D self-driven devices, and an impressive external quantum efficiency of 68%.

A YOLOX-based Deep Instance Segmentation Neural Network for Cardiac Anatomical Structures in Fetal Ultrasound Images.

Echocardiography is an essential procedure for the prenatal examination of the fetus for congenital heart disease (CHD). Accurate segmentation of key anatomical structures in a four-chamber view is an essential step in measuring fetal growth parameters and diagnosing CHD. Currently, most obstetricians perform segmentation tasks manually, but the pixel-level operation is labor-intensive and requires extensive anatomical knowledge and clinical experience. As such, efficiently and accurately detecting structures from real-world fetal ultrasound images is a key challenge. In this paper, we propose a YOLOX-based deep instance segmentation neural network (i.e., IS-YOLOX) for cardiac anatomical structure location and segmentation in fetal ultrasound images. Specifically, we reconstruct a new instance segmentation branch based on a multi-task deep learning framework. We then design a new multi-level non-maximum suppression (NMS) mechanism to further improve the segmentation performance that consists of three levels of selection. Moreover, unlike two-stage instance segmentation approaches, our method does not rely on object detection results. To the best of our knowledge, this is the first study regarding instance segmentation on 13 types of anatomical structures in the fetal four-chamber view. Extensive experiments were carried out on clinical datasets, and the experimental results show that our method outperforms nine competitive baselines.