Dual Modulation Research Articles

DAFE-MSGAT: Dual-Attention Feature Extraction and Multi-Scale Graph Attention Network for Polyphonic Piano Transcription

Automatic music transcription (AMT) aims to convert raw audio signals into symbolic music. This is a highly challenging task in the fields of signal processing and artificial intelligence, and it holds significant application value in music information retrieval (MIR). Existing methods based on convolutional neural networks (CNNs) often fall short in capturing the time-frequency characteristics of audio signals and tend to overlook the interdependencies between notes when processing polyphonic piano with multiple simultaneous notes. To address these issues, we propose a dual attention feature extraction and multi-scale graph attention network (DAFE-MSGAT). Specifically, we design a dual attention feature extraction module (DAFE) to enhance the frequency and time-domain features of the audio signal, and we utilize a long short-term memory network (LSTM) to capture the temporal features within the audio signal. We introduce a multi-scale graph attention network (MSGAT), which leverages the various implicit relationships between notes to enhance the interaction between different notes. Experimental results demonstrate that our model achieves high accuracy in detecting the onset and offset of notes on public datasets. In both frame-level and note-level metrics, DAFE-MSGAT achieves performance comparable to the state-of-the-art methods, showcasing exceptional transcription capabilities.

Multi-attention associate prediction network for visual tracking

Classification-regression prediction networks have realized impressive success in several modern deep trackers. However, there is an inherent difference between classification and regression tasks, so they have diverse even opposite demands for feature matching. Existed models always ignore the key issue and only employ a unified matching block in two task branches, decaying the decision quality. Besides, these models also struggle with decision misalignment situation. In this paper, we propose a multi-attention associate prediction network (MAPNet) to tackle the above problems. Concretely, two novel matchers, i.e., category-aware matcher and spatial-aware matcher, are first designed for feature comparison by integrating self, cross, channel or spatial attentions organically. They are capable of fully capturing the category-related semantics for classification and the local spatial contexts for regression, respectively. Then, we present a dual alignment module to enhance the correspondences between two branches, which is useful to find the optimal tracking solution. Finally, we describe a Siamese tracker built upon the proposed prediction network, which achieves the leading performance on five tracking benchmarks, consisting of LaSOT, TrackingNet, GOT-10k, TNL2k and UAV123, and surpasses other state-of-the-art approaches.

DACG: Dual Attention and Context Guidance model for radiology report generation

Medical images are an essential basis for radiologists to write radiology reports and greatly help subsequent clinical treatment. The task of generating automatic radiology reports aims to alleviate the burden of clinical doctors writing reports and has received increasing attention this year, becoming an important research hotspot. However, there are severe issues of visual and textual data bias and long text generation in the medical field. Firstly, Abnormal areas in radiological images only account for a small portion, and most radiological reports only involve descriptions of normal findings. Secondly, there are still significant challenges in generating longer and more accurate descriptive texts for radiology report generation tasks. In this paper, we propose a new Dual Attention and Context Guidance (DACG) model to alleviate visual and textual data bias and promote the generation of long texts. We use a Dual Attention Module, including a Position Attention Block and a Channel Attention Block, to extract finer position and channel features from medical images, enhancing the image feature extraction ability of the encoder. We use the Context Guidance Module to integrate contextual information into the decoder and supervise the generation of long texts. The experimental results show that our proposed model achieves state-of-the-art performance on the most commonly used IU X-ray and MIMIC-CXR datasets. Further analysis also proves that our model can improve reporting through more accurate anomaly detection and more detailed descriptions. The source code is available at https://github.com/LangWY/DACG.

OTFS With Dual Frequency Index Modulation

OTFS With Dual Frequency Index Modulation

Characteristic modules and Gorenstein (co-)homological dimension of groups

In this paper, we examine the Gorenstein dimension of modules over the group algebra kG of a group G with coefficients in a commutative ring k. As a Gorenstein analogue of the classical case, we bound this dimension in terms of the Gorenstein dimension of the underlying k-module and the Gorenstein dimension of G over k. Our method is based on the notion of a characteristic module for G, introduced by the second author, and uses the stability properties of the Gorenstein categories. We also examine the class of hierarchically decomposable groups defined by Kropholler and use the module of bounded Z-valued functions on such a group G to characterize the Gorenstein flat ZG-modules, in terms of flat modules, and the Gorenstein injective ZG-modules, in terms of injective modules (by complete analogy with the characterization of Gorenstein projective ZG-modules, in terms of projective modules, obtained by Dembegioti and the second author). It follows that, for a group G in Kropholler's class, (a) any Gorenstein projective ZG-module is Gorenstein flat and (b) a ZG-module is Gorenstein flat if its Pontryagin dual module is Gorenstein injective.

Spatio-temporal Dual Graph Neural Networks for Travel Time Estimation

Travel time estimation is one of the core tasks for the development of intelligent transportation systems. Most previous works model the road segments or intersections separately by learning their spatio-temporal characteristics to estimate travel time. However, due to the continuous alternations of the road segments and intersections in a path, the dynamic features are supposed to be coupled and interactive. Therefore, modeling one of them limits further improvement in accuracy of estimating travel time. To address the above problems, a novel graph-based deep learning framework for travel time estimation is proposed in this article, namely, Spatio-temporal Dual Graph Neural Networks (STDGNN). Specifically, we first establish the node-wise and edge-wise graphs to, respectively, characterize the adjacency relations of intersections and that of road segments. To extract the joint spatio-temporal correlations of the intersections and road segments, we adopt the spatio-temporal dual graph learning approach that incorporates multiple spatial-temporal dual graph learning modules with multi-scale network architectures for capturing multi-level spatial-temporal information from the dual graph. Finally, we employ the multi-task learning approach to estimate the travel time of a given whole route, each road segment and intersection simultaneously. We conduct extensive experiments to evaluate our proposed model on three real-world trajectory datasets, and the experimental results show that STDGNN significantly outperforms several state-of-art baselines.

Dual phase hierarchical brain tumour detection and segmentation using unet based skip guidance residual convolutional transformer

Brain tumour segmentation and classification is an important approach to understanding tumour characteristics effectively. Accurate lesion segmentation requires many image modalities with varied contrasts. As a result, manual segmentation may not be suitable for large-scale research. The research has shown interest in developing various learning algorithms to detect and segment brain tumour (BT) images accurately based on the set of experimental images. Some of the issues noticed in the existing models, such as high computational complexity, high processing time, and less accuracy, were identified. In order to overcome these existing issues, a novel Dual-phase channel attentional Residual convolutional transformer Unet model (DCARCTUnet) is introduced for segmenting brain tumours. Initially, the input images are passed into the pre-processing stage to remove noise and blur. The Efficient Net model identified the pre-processed image as either normal or tumour. The proposed segmentation model can be integrated into three stages, namely dual phase channel attention module (DCAM), residual convolutional transformer Unet model and Pyramid skip guidance module (PSM). The proposed method uses the BRATS 2020–2021 dataset to analyze the results accurately. The results showed that the proposed model has a dice coefficient value of 0.99 and a Jaccard coefficient value of 0.98. Furthermore, performance analysis is carried out to demonstrate the effectiveness of the proposed model.

Dual modulation of homogeneous nanomaterialization and electrochemical activation enhancing zinc ion storage

Dual modulation of homogeneous nanomaterialization and electrochemical activation enhancing zinc ion storage

Enhanced Recognition Memory through Dual Modulation of Brain Carbonic Anhydrases and Cholinesterases.

This study introduces a novel multitargeting strategy that combines carbonic anhydrase (CA) activators and cholinesterase (ChE) inhibitors to enhance cognitive functions. A series of tacrine-based derivatives with amine/amino acid moieties were synthesized and evaluated for their dual activity on brain CA isoforms and ChEs (AChE and BChE). Several derivatives, notably compounds 26, 30, 34, and 40, demonstrated potent CA activation, particularly of hCA II and VII, and strong ChE inhibition with subnanomolar to low nanomolar IC50 values. In vivo studies using a mouse model of social recognition memory showed that these derivatives significantly improved memory consolidation at doses 10-100 times lower than the reference compounds (either alone or in combination). Molecular modeling and ADMET predictions elucidated the compound binding modes and confirmed favorable pharmacokinetic and safety profiles. The findings suggest that dual modulation of CA and ChE activities is a promising strategy for treating cognitive deficits associated with neurodegenerative and psychiatric disorders.

Attention redirection transformer with semantic oriented learning for unbiased scene graph generation

Scene Graph Generation (SGG) plays an important role in scene understanding because all of the objects and relations in an image can be abstracted into a concise topological graph. Due to the complexity of visual scenes, including mutual occlusion between objects and semantic ambiguity, SGG is still a challenging task. Most of the existing models only focused on the context of a single object while contexts provided by paired objects are ignored. In this paper, we propose an Attention Redirection Transformer (ART) to extract pair-level contexts specifically, which is divided into an attention distraction stage and an attention integration stage. In this way, the attention of the model is forced to be redirected, which explores the implicit information in the background. In addition, to incorporate the semantic information of predicates, a Semantic Oriented Learning Module (SOL) is designed, which may assist in getting better textual semantics and also prompts cross-modal information fusion. At last, a self-diversity driven Dual Translation Embedding Module (DTM) is designed, which refines representations of subject and object and makes them distinct. Experimental results on the Visual Genome dataset demonstrate the effectiveness of our proposed method. Moreover, our method outperforms state-of-the-art methods on the mR@K metric. The source codes are released on Github: https://github.com/Nora-Zhang98/ART-SOL.

A Novel Feedforward Scheme for Enhancing Dynamic Performance of Vector-Controlled Dual Active Bridge Converter with Dual Phase Shift Modulation for Fast Battery Charging Systems

This paper proposes a novel feedforward control scheme to achieve a very smooth transition from Constant Current (CC) to Constant Voltage (CV) charging modes, the commonly used method for electric vehicle charging applications. Furthermore, a three-loop model-independent Linear Active Disturbance Rejection Control (LADRC)-based system is proposed, replacing the traditional two-loop Proportional-Integral (PI) control system. The extra loop performs a decoupled dq vector control of the inductor current, which is typically not used in single-phase Dual Active Bridge (DAB) systems. This additional loop not only facilitates the optimal determination of both internal and external phase shift angles of a Dual-Phase Shift (DPS) modulator but also lowers the peak input current of the converter, allowing for lower-rated switches. Numerical simulations using MATLAB/Simulink demonstrate the robustness of the proposed control strategy against both input voltage disturbances and load disturbances during the transition from CC to CV charging modes. Hence, the dynamic performance of the charging system is significantly improved with minimal controller effort.

Dual Kv7.2/3-TRPV1 modulators inhibit nociceptor hyperexcitability and alleviate pain without target-related side effects.

Persistent or chronic pain is the primary reason people seek medical care, yet current therapies are either limited in efficacy or cause intolerable side effects. Diverse mechanisms contribute to the basic phenomena of nociceptor hyperexcitability that initiates and maintains pain. Two prominent players in the modulation of nociceptor hyperexcitability are the transient receptor potential vanilloid type 1 (TRPV1) ligand-gated ion channel and the voltage-gated potassium channel, Kv7.2/3, that reciprocally regulate neuronal excitability. Across many drug development programs targeting either TRPV1 or Kv7.2/3, significant evidence has been accumulated to support these as highly relevant targets; however, side effects that are poorly separated from efficacy have limited the successful clinical translation of numerous Kv7.2/3 and TRPV1 drug development programs. We report here the pharmacological profile of 3 structurally related small molecule analogues that demonstrate a novel mechanism of action (MOA) of dual modulation of Kv7.2/3 and TRPV1. Specifically, these compounds simultaneously activate Kv7.2/3 and enable unexpected specific and potent inhibition of TRPV1. This in vitro potency translated to significant analgesia in vivo in several animal models of acute and chronic pain. Importantly, this specific MOA is not associated with any previously described Kv7.2/3 or TRPV1 class-specific side effects. We suggest that the therapeutic potential of this MOA is derived from the selective and specific targeting of a subpopulation of nociceptors found in rodents and humans. This efficacy and safety profile supports the advancement of dual TRPV1-Kv7.2/3 modulating compounds into preclinical and clinical development for the treatment of chronic pain.

Characterizing third-order intermodulation distortion of photodetectors based on de-coupling and de-embedding modulation distortion of modulators

A novel, to the best of our knowledge, electro-optical modulation method is proposed for measuring third-order intermodulation distortion of photodetectors (PDs) based on de-coupling and de-embedding modulation distortion of modulators. The method utilizes dual parallel intensity modulation to generate electro-optical stimulus signals with fast and fine sweeping capability, and it eliminates the nonlinear impact of modulators by using low-frequency bias swing, allowing a direct extraction of the third-order output intercept point (OIP3) of PD from the combined nonlinear response contributed by both the modulators and the PD. The OIP3 of PD is frequency-swept measured with our method and compared to those with the conventional method to check for consistency. The proposed method enables a modulator-distortion-free, fast, and fine sweeping measurement of PDs using a simple system.

Dual-Electrically Configurable MoTe2/In2S3 Phototransistor toward Multifunctional Applications.

Photodetectors, essential for a wide range of optoelectronic applications in both military and civilian sectors, face challenges in balancing responsivity, detectivity, and response time due to their inherent unidirectional carrier transport mechanism. Multifunctional photodetectors that address these trade-offs are highly sought after for their potential to reduce costs, simplify system design, and surpass Moore's Law limitations. Herein, we present a multimodal phototransistor based on a 2D MoTe2/In2S3 heterostructure. Through dual electrical modulation employing bias voltage and gate voltage, we engineer the energy band to achieve switchable photoresponse mechanisms between photoconductive and photovoltaic modes. In photoconductive mode, the device exhibits a responsivity of 320 A/W and a specific detectivity of 1.2 × 1013 Jones. Meanwhile, in photovoltaic mode, it exhibits a light on/off ratio of 2 × 105 and response speed of 0.68/0.60 ms. These capabilities enable multifunctional applications such as high-resolution imaging across various wavelengths, a conceptual optoelectronic logic gate, and dual-channel optical communication. This work makes an advancement in the development of future multifunctional optoelectronic devices.

MEDANet: More Efficient Dual Attention Network for Scene Segmentation

The dual attention module is a potent semantic segmentation technique renowned for its capabilities, yet it often faces significant computational demands and GPU memory usage. To tackle these challenges, we introduce an advanced dual perception network comprising two modules: A streamlined Multi-scale Efficient Position Attention Module (MEPAM) and an optimized Efficient Channel Attention Module (MECAM). MEPAM incorporates multi-scale global average pooling into the Position Attention Module (PAM), substantially cutting computational overhead and memory consumption without compromising performance. Meanwhile, MECAM integrates compressed convolutions into the Channel Attention Module (CAM), improving segmentation accuracy and inference speed compared to conventional methods like DANet. Our approach underwent comprehensive evaluation on a semantic segmentation benchmark dataset, showcasing superior performance. For instance, on the Cityscapes dataset, our method achieves an IoU of 82.2%. In terms of efficiency gains, MEPAM operates nearly 1.97 times faster than the standard PAM module on GPU, while requiring 7.55 times less memory with a [Formula: see text] input. Similarly, MECAM achieves approximately 2.2 times faster processing than CAM, while cutting GPU memory usage by 7.53 times. This innovative dual perception network not only enhances segmentation accuracy and speed but also addresses the computational challenges associated with traditional dual attention modules.

Hierarchical feature aggregation with mixed attention mechanism for single-cell RNA-seq analysis

Single-cell RNA sequencing (scRNA-seq) analysis is capable of elucidating cell heterogeneity and diversity, making cell-level biological research possible. Cell type clusterization is one of the main goals of scRNA-seq analysis. However, existing methods lack a flexible aggregation mechanism to fuse node attribute features and graph structural features adaptively. They also ignore multi-scale information embedded in different layers of networks. Here, we propose AGAC (Attention-driven Graph Attentional Clustering), a unified computational framework that applies hierarchical feature aggregation with mixed attentional mechanisms to scRNA-seq data clustering. Firstly, AGAC learns the attribute features of cells through graph attention autoencoder and the graph structural features among cells through graph neural network (GNN) respectively. Secondly, AGAC utilizes the heterogeneous intelligent fusion module to fuse these two kinds of features dynamically, as well as the multi-scale intelligent fusion module to aggregate the multi-scale information embedded in different layers of the GNN adaptively. In addition, AGAC adopts a dual self-supervision module for end-to-end training and synchronous optimization. Experimental results on 13 real scRNA-seq datasets demonstrate that AGAC is more effective than existing baseline methods because it takes into account the discriminative information in the network comprehensively and generates clustering results directly. The source code of AGAC can be downloaded at https://github.com/zzyqh/AGAC.

First-in-Class Mitogen-Activated Protein Kinase p38α: MAPK-Activated Protein Kinase-2 (MK2) Dual Signal Modulator with Anti-inflammatory and Endothelial-stabilizing Properties.

We previously identified a small molecule, UM101, predicted to bind to the substrate-binding groove of p38aMitogen-activated Protein Kinase (MAPK) near the binding site of its proinflammatory substrate, MAPK-activated protein kinase (MK2). UM101 exhibited anti-inflammatory, endothelial-stabilizing, and lung-protective effects. To overcome its limited aqueous solubility and p38a binding affinity, we designed an analog of UM101, GEn-1124, with improved aqueous solubility, stability, and p38a binding affinity. Compared with UM101, GEn-1124 has 18-fold greater p38a-binding affinity as measured by Surface Plasmon Resonance (SPR), 11-fold greater aqueous solubility, enhanced barrier-stabilizing activity in thrombin-stimulated human pulmonary artery endothelial cells (hPAEC) in vitro, and greater lung protection in vivo GEn-1124 improved survival from 10% to 40% in murine acute lung injury (ALI) induced by combined exposure to intratracheal bacterial endotoxin lipopolysaccharide (LPS) instillation and febrile-range hyperthermia (FRH) and from 0% to 50% in a mouse influenza pneumonia model. Gene expression analysis by RNASeq in TNFa-treated hPAEC showed that the gene-modifying effects of GEn-1124 were much more restricted to TNFa-inducible genes than the catalytic site p38 inhibitor, SB203580. Gene expression pathway analysis, confocal immunofluorescence analysis of p38aand MK2 subcellular trafficking, and SPR analysis of phosphorylated p38a:MK2 binding affinity supports a novel mechanism of action. GEn-1124 destabilizes the activated p38a:MK2 complex, dissociates nuclear export of MK2 and p38a, thereby promoting intranuclear retention and enhanced intranuclear signaling by phosphorylated p38a retention, and accelerated inactivation of p38-free cytosolic MK2 by unopposed phosphatases. Significance Statement We describe an analog of our first-in-class small molecule modulator of p38a/MK2 signaling targeted to a pocket near the ED substrate binding domain of p38a, which destabilizes the p38a:MK2 complex without blocking p38 catalytic activity or ablating downstream signaling. The result is a rebalancing of downstream pro- and anti-inflammatory signaling, yielding anti-inflammatory, endothelial-stabilizing, and lung-protective effects with therapeutic potential in ARDS.

CFTR modulators response of S737F and T465N CFTR variants on patient-derived rectal organoids

BackgroundPredictions based on patient-derived materials of CFTR modulators efficacy have been performed lately in patient-derived cells, extending FDA-approved drugs for CF patients harboring rare variants. Here we developed intestinal organoids from subjects carrying S737F- and T465N-CFTR in trans with null alleles to evaluate their functional impact on CFTR protein function and their restoration upon CFTR modulator treatment. The characterization of S737F-CFTR was performed in two subjects recently assessed in nasal epithelial cells but not in colonoids.ResultsOur functional analysis (Ussing chamber) confirmed that S737F-CFTR is a mild variant with residual function as investigated in colonoids of patients with S737F/Dele22-24 and S737F/W1282X genotypes. An increase of current upon Elexacaftor/Tezacaftor/Ivacaftor (ETI) treatment was recorded for the former genotype. T465N is a poorly characterized missense variant that strongly impacts CFTR function, as almost no CFTR-mediated anion secretion was registered for T465N/Q39X colonoids. ETI treatment substantially improved CFTR-mediated anion secretion and increased the rescue of mature CFTR expression compared to either untreated colonoids or to dual CFTR modulator therapies.ConclusionsOur study confirms the presence of a residual function of the S737F variant and its limited response to CFTR modulators while predicting for the first time the potential clinical benefit of Trikafta® for patients carrying the rare T465N variant.

Chemical reaction enhanced graph learning for molecule representation.

Molecular representation learning (MRL) models molecules with low-dimensional vectors to support biological and chemical applications. Current methods primarily rely on intrinsic molecular information to learn molecular representations, but they often overlook effectively integrating domain knowledge into MRL. In this article, we develop a reaction-enhanced graph learning (RXGL) framework for MRL, utilizing chemical reactions as domain knowledge. RXGL introduces dual graph learning modules to model molecule representation. One module employs graph convolutions on molecular graphs to capture molecule structures. The other module constructs a reaction-aware graph from chemical reactions and designs a novel graph attention network on this graph to integrate reaction-level relations into molecular modeling. To refine molecule representations, we design a reaction-based relation learning task, which considers the relations between the reactant and product sides in reactions. In addition, we introduce a cross-view contrastive task to strengthen the cooperative associations between molecular and reaction-aware graph learning. Experiment results show that our RXGL achieves strong performance in various downstream tasks, including product prediction, reaction classification, and molecular property prediction. The code is publicly available at https://github.com/coder-ACAC/RLM.

A lightweight convolutional neural network-based feature extractor for visible images

Feature extraction networks (FENs), as the first stage in many computer vision tasks, play critical roles. Previous studies regarding FENs employed deeper and wider networks to attain higher accuracy, but their approaches were memory-inefficient and computationally intensive. Here, we present an accurate and lightweight feature extractor (RoShuNet) for visible images based on ShuffleNetV2. The provided improvements are threefold. To make ShuffleNetV2 compact without degrading its feature extraction ability, we propose an aggregated dual group convolutional module; to better aid the channel interflow process, we propose a γ-weighted shuffling module; to further reduce the complexity and size of the model, we introduce slimming strategies. Classification experiments demonstrate the state-of-the-art (SOTA) performance of RoShuNet, which yields an increase in accuracy and reduces the complexity and size of the model compared to those of ShuffleNetV2. Generalization experiments verify that the proposed method is also applicable to feature extraction tasks in semantic segmentation and multiple-object tracking scenarios, achieving comparable accuracy to that of other approaches with more memory and greater computational efficiency. Our method provides a novel perspective for designing lightweight models.