Feature Learning Research Articles

Integrative analysis of single-cell transcriptomic and multilayer signaling networks in glioma reveal tumor progression stage

IntroductionTumor microenvironments (TMEs) encompass complex ecosystems of cancer cells, infiltrating immune cells, and diverse cell types. Intercellular and intracellular signals within the TME significantly influence cancer progression and therapeutic outcomes. Although computational tools are available to study TME interactions, explicitly modeling tumor progression across different cancer types remains a challenge.MethodsThis study introduces a comprehensive framework utilizing single-cell RNA sequencing (scRNA-seq) data within a multilayer network model, designed to investigate molecular changes across glioma progression stages. The heterogeneous, multilayered network model replicates the hierarchical structure of biological systems, from genetic building blocks to cellular functions and phenotypic manifestations.ResultsApplying this framework to glioma scRNA-seq data allowed complex network analysis of different cancer stages, revealing significant ligand‒receptor interactions and key ligand‒receptor-transcription factor (TF) axes, along with their associated biological pathways. Differential network analysis between grade III and grade IV glioma highlighted the most critical nodes and edges involved in interaction rewiring. Pathway enrichment analysis identified four essential genes—PDGFA (ligand), PDGFRA (receptor), CREB1 (TF), and PLAT (target gene)—involved in the Receptor Tyrosine Kinases (RTK) signaling pathway, which plays a pivotal role in glioma progression from grade III to grade IV.DiscussionThese genes emerged as significant features for machine learning in predicting glioma progression stages, achieving 87% accuracy and 93% AUC in a 3-year survival prediction through Kaplan-Meier analysis. This framework provides deeper insights into the cellular machinery of glioma, revealing key molecular relationships that may inform prognosis and therapeutic strategies.

An historical overview of artificial intelligence for diagnosis of major depressive disorder

The Artificial Intelligence (AI) technology holds immense potential in the realm of automated diagnosis for Major Depressive Disorder (MDD), yet it is not without potential shortcomings. This paper systematically reviews the research progresses of integrating AI technology with depression diagnosis and provides a comprehensive analysis of existing research findings. In this context, we observe that the knowledge-driven first-generation of depression diagnosis methods could only address deterministic issues in structured information, with the selection of depression-related features directly influencing identification outcomes. The data-driven second-generation of depression diagnosis methods achieved automatic learning of features but required substantial high-quality clinical data, and the results were often obtained solely from the black-box models which lack sufficient explainability. In an effort to overcome the limitations of the preceding approaches, the third-generation of depression diagnosis methods combined the strengths of knowledge-driven and data-driven approaches. Through the fusion of information, the diagnostic accuracy is greatly enhanced, but the interpretability remains relatively weak. In order to enhance interpretability and introduce diagnostic criteria, this paper offers a new approach using Large Language Models (LLMs) as AI agents for assisting the depression diagnosis. Finally, we also discuss the potential advantages and challenges associated with this approach. This newly proposed innovative approach has the potential to offer new perspectives and solutions in the diagnosis of depression.

Physics-aware cross-domain fusion aids learning-driven computer-generated holography

The rapid advancement of computer-generated holography has bridged deep learning with traditional optical principles in recent years. However, a critical challenge in this evolution is the efficient and accurate conversion from the amplitude to phase domain for high-quality phase-only hologram (POH) generation. Existing computational models often struggle to address the inherent complexities of optical phenomena, compromising the conversion process. In this study, we present the cross-domain fusion network (CDFN), an architecture designed to tackle the complexities involved in POH generation. The CDFN employs a multi-stage (MS) mechanism to progressively learn the translation from amplitude to phase domain, complemented by the deep supervision (DS) strategy of middle features to enhance task-relevant feature learning from the initial stages. Additionally, we propose an infinite phase mapper (IPM), a phase-mapping function that circumvents the limitations of conventional activation functions and encapsulates the physical essence of holography. Through simulations, our proposed method successfully reconstructs high-quality 2K color images from the DIV2K dataset, achieving an average PSNR of 31.68 dB and SSIM of 0.944. Furthermore, we realize high-quality color image reconstruction in optical experiments. The experimental results highlight the computational intelligence and optical fidelity achieved by our proposed physics-aware cross-domain fusion.

Robust Multi-Class Classification for Real-Time Agricultural Applications Using Efficient and Adaptive Deep Learning

Plant diseases severely affect agricultural productivity, necessitating accurate and rapid detection methods. This research presents a robust, multi-class plant disease classification framework using adaptive deep learning. We utilize pre-trained convolutional neural networks (CNNs), specifically Xception, InceptionResNetV2, InceptionV3, ResNet50, and the proposed EfficientNetB3-based Adaptive Augmented Deep Learning (EfficientNetB3-AADL) model. Our approach leverages transfer learning combined with extensive data augmentation and trimming techniques to enhance model performance and mitigate overfitting. The EfficientNetB3-AADL architecture incorporates convolutional and max pooling layers, regularization strategies, and a dense feature learning layer, optimized to classify 52 disease categories from a publicly available leaf image dataset. The model’s performance is extensively evaluated using metrics such as accuracy, precision, recall, and F1 score. Notably, EfficientNetB3-AADL achieves superior accuracy over 98%, outperforming other CNN models. The proposed methodology highlights the efficacy of compound scaling and adaptive data augmentation in ensuring robust and efficient disease classification, suitable for real-time agricultural applications. This advancement supports sustainable farming by offering a scalable, computationally efficient solution for early and accurate disease detection in diverse crop species.

No-Reference Quality Assessment Based on Dual-Channel Convolutional Neural Network for Underwater Image Enhancement

Underwater images are important for underwater vision tasks, yet their quality often degrades during imaging, promoting the generation of Underwater Image Enhancement (UIE) algorithms. This paper proposes a Dual-Channel Convolutional Neural Network (DC-CNN)-based quality assessment method to evaluate the performance of different UIE algorithms. Specifically, inspired by the intrinsic image decomposition, the enhanced underwater image is decomposed into reflectance with color information and illumination with texture information based on the Retinex theory. Afterward, we design a DC-CNN with two branches to learn color and texture features from reflectance and illumination, respectively, reflecting the distortion characteristics of enhanced underwater images. To integrate the learned features, a feature fusion module and attention mechanism are conducted to align efficiently and reasonably with human visual perception characteristics. Finally, a quality regression module is used to establish the mapping relationship between the extracted features and quality scores. Experimental results on two public enhanced underwater image datasets (i.e., UIQE and SAUD) show that the proposed DC-CNN method outperforms a variety of the existing quality assessment methods.

A 18F-FDG PET/CT-based deep learning-radiomics-clinical model for prediction of cervical lymph node metastasis in esophageal squamous cell carcinoma

Abstract Background To develop an artificial intelligence (AI)-based model using Radiomics, deep learning (DL) features extracted from 18F-fluorodeoxyglucose (18F-FDG) Positron emission tomography/Computed Tomography (PET/CT) images of tumor and cervical lymph node with clinical feature for predicting cervical lymph node metastasis (CLNM) in patients with esophageal squamous cell carcinoma (ESCC). Methods The study included 300 ESCC patients from the First Affiliated Hospital of Zhengzhou University who were divided into a training cohort and an internal testing cohort with an 8:2 ratio. Another 111 patients from Shanghai Chest Hospital were included as the external cohort. For each sample, we extracted 428 PET/CT-based Radiomics features from the gross tumor volume (GTV) and cervical lymph node (CLN) delineated layer by layer and 256 PET/CT-based DL features from the maximum cross-section of GTV and CLN images We input these features into seven different machine learning algorithms and ultimately selected logistic regression (LR) as the model classifier. Subsequently, we evaluated seven models (Clinical, Radiomics, Radiomics-Clinical, DL-Clinical, DL-Radiomics, DL-Radiomics-Clinical) using Radiomics features, DL features and clinical feature. Results The DL-Radiomics-Clinical (DRC) model demonstrated higher AUC of 0.955 and 0.916 compared to the other six models in both internal and external testing cohorts respectively. The DRC model achieved the highest accuracy among the seven models in both the internal and external test sets, with scores of 0.951 and 0.892, respectively. Conclusions Through the combination of Radiomics features and DL features from PET/CT imaging and clinical feature, we developed a predictive model exhibiting exceptional classification capabilities. This model can be considered as a non-invasive method for predication of CLNM in patients with ESCC. It might facilitate decision-making regarding to the extend of lymph node dissection, and to select candidates for postoperative adjuvant therapy.

Robust-DefReg: A Robust Coarse to Fine Non-rigid Point Cloud Registration Method based on Graph Convolutional Neural Networks

Abstract Point cloud registration is a critical process in computer vision and measurement science, aimed at determining transformations between corresponding sets of points for accurate spatial alignment. In particular, non-rigid registration involves estimating flexible transformations that map a source point cloud to a target point cloud, even under conditions of stretching, compression, or other complex deformations. This task becomes especially challenging when addressing measurement-specific issues like varying degrees of deformation, noise, and outliers, all of which can impact measurement accuracy and reliability.
This paper introduces Robust-DefReg, a novel method for non-rigid point cloud registration that applies graph convolutional networks (GCNNs) within a coarse-to-fine registration framework. This end-to-end pipeline harnesses global feature learning to establish robust correspondences and precise transformations, enabling high accuracy across different deformation scales and noise levels. A key contribution of Robust-DefReg is its demonstrated resilience to various challenges, such as substantial deformations, noise, and outliers, factors often underreported in existing registration literature.
In addition, we present SynBench, a comprehensive benchmark dataset specifically designed for evaluating non-rigid point cloud registration in realistic measurement scenarios. Unlike previous datasets, SynBench incorporates a range of challenges, making it a valuable tool for the fair assessment of registration methods in measurement applications. Experimental results on SynBench and additional datasets show that Robust-DefReg consistently outperforms state-of-the-art methods, offering higher registration accuracy and robustness, even with up to 45% outliers.
SynBench and the Robust-DefReg source code are publicly accessible for further research and development at https://doi.org/10.11588/data/R9IKCF and https://github.com/m-kinz/Robust-DefReg, respectively.

HRCUNet: Hierarchical Region Contrastive Learning for Segmentation of Breast Tumors in DCE‐MRI

ABSTRACTSegmenting breast tumors from dynamic contrast‐enhanced magnetic resonance images is a critical step in the early detection and diagnosis of breast cancer. However, this task becomes significantly more challenging due to the diverse shapes and sizes of tumors, which make it difficult to establish a unified perception field for modeling them. Moreover, tumor regions are often subtle or imperceptible during early detection, exacerbating the issue of extreme class imbalance. This imbalance can lead to biased training and challenge accurately segmenting tumor regions from the predominant normal tissues. To address these issues, we propose a hierarchical region contrastive learning approach for breast tumor segmentation. Our approach introduces a novel hierarchical region contrastive learning loss function that addresses the class imbalance problem. This loss function encourages the model to create a clear separation between feature embeddings by maximizing the inter‐class margin and minimizing the intra‐class distance across different levels of the feature space. In addition, we design a novel Attention‐based 3D Multi‐scale Feature Fusion Residual Module to explore more granular multi‐scale representations to improve the feature learning ability of tumors. Extensive experiments on two breast DCE‐MRI datasets demonstrate that the proposed algorithm is more competitive against several state‐of‐the‐art approaches under different segmentation metrics.

An Enhanced Feature Learning CNN Using Auto-encoder for Partially Occluded Target Recognition in SAR Imagery

An Enhanced Feature Learning CNN Using Auto-encoder for Partially Occluded Target Recognition in SAR Imagery

Deep Learning‐Based Anomaly Diagnosis Method for Capacitive Voltage Transformer Measurement Error in PIoT

ABSTRACTIn the era of power Internet of Things (PIoT), the accuracy of capacitive voltage transformers (CVTs) is crucial for maintaining the reliability of voltage measurement and protection systems in smart grids, thereby contributing to overall grid stability and efficiency. Accurate and timely detection of anomalies in CVT measurement errors is essential for preventing equipment failures, reducing maintenance costs, and improving overall system reliability. However, existing anomaly diagnosis methods often rely on statistical analysis and rule‐based approaches, which have limitations in capturing complex patterns and adapting to evolving anomaly types. This paper proposes a novel deep learning‐based anomaly diagnosis method for CVT measurement error in PIoT, called LSTM‐CVT. The proposed method leverages a long short‐term memory (LSTM) neural network architecture with three key strategies: bidirectional temporal dependency capture, hierarchical feature learning, and joint anomaly diagnosis and error estimation. The experimental results demonstrate the superior performance of LSTM‐CVT compared to state‐of‐the‐art baseline methods.

AI-Driven Fraud Detection in Banking: Enhancing Transaction Security

An important step forward in risk management and fraud detection has been achieved with the integration of Artificial Intelligence (AI) in the banking sector. In this paper, we take a look at how AI has revolutionized various fields, shedding light on the benefits and drawbacks of this technology. The effects of AI on risk management are complex. More complex credit risk assessment models are made possible by algorithms that can see patterns in massive datasets that people might miss. When it comes to market and liquidity issues, real-time transaction monitoring is absolutely essential for quick risk mitigation. Automating compliance with regulatory norms is another critical function of AI, which helps to decrease human mistake and assures quick adaptability to changes in regulations. The automation of mundane processes and the reinforcement of cybersecurity measures further reduce operational risks. By examining client behaviour and transaction data, enhanced algorithms may adeptly spot anomalies that could indicate fraud. Artificial intelligence's capacity to foresee future events enables it to foil possible fraud attempts. The systems are designed to respond to changing fraudster strategies with its adaptive learning feature.

SAM-Net: Spatio-Temporal Sequence Typhoon Cloud Image Prediction Net with Self-Attention Memory

Cloud image prediction is a spatio-temporal sequence prediction task, similar to video prediction. Spatio-temporal sequence prediction involves learning from historical data and using the learned features to generate future images. In this process, the changes in time and space are crucial for spatio-temporal sequence prediction models. However, most models now rely on stacking convolutional layers to obtain local spatial features. In response to the complex changes in cloud position and shape in cloud images, the prediction module of the model needs to be able to extract both global and local spatial features from the cloud images. In addition, for irregular cloud motion, more attention should be paid to the spatio-temporal sequence features between input cloud image frames in the temporal sequence prediction module, considering the extraction of temporal features with long temporal dependencies, so that the spatio-temporal sequence prediction network can learn cloud motion trends more accurately. To address these issues, we have introduced an innovative model called SAM-Net. The self-attention module of this model aims to extract an inner image frame’s spatial features of global and local dependencies. In addition, a memory mechanism has been added to the self-attention module to extract interframe features with long temporal and spatial dependencies. Our method shows better performance than the PredRNN-v2 model on publicly available datasets such as MovingMNIST and KTH. We achieved the best performance in both the 4-time-step and 10-time-step typhoon cloud image predictions. On a cloud dataset consisting of 10 time steps, we observed a decrease in MSE of 180.58, a decrease in LPIPS of 0.064, an increase in SSIM of 0.351, and a significant improvement in PSNR of 5.56 compared to PredRNN-v2.

Enhancing multiclass COVID-19 prediction with ESN-MDFS: Extreme smart network using mean dropout feature selection technique

Deep learning and artificial intelligence offer promising tools for improving the accuracy and efficiency of diagnosing various lung conditions using portable chest x-rays (CXRs). This study explores this potential by leveraging a large dataset containing over 6,000 CXR images from publicly available sources. These images encompass COVID-19 cases, normal cases, and patients with viral or bacterial pneumonia. The research proposes a novel approach called "Enhancing COVID Prediction with ESN-MDFS" that utilizes a combination of an Extreme Smart Network (ESN) and a Mean Dropout Feature Selection Technique (MDFS). This study aimed to enhance multi-class lung condition detection in portable chest X-rays by combining static texture features with dynamic deep learning features extracted from a pre-trained VGG-16 model. To optimize performance, preprocessing, data imbalance, and hyperparameter tuning were meticulously addressed. The proposed ESN-MDFS model achieved a peak accuracy of 96.18% with an AUC of 1.00 in a six-fold cross-validation. Our findings demonstrate the model’s superior ability to differentiate between COVID-19, bacterial pneumonia, viral pneumonia, and normal conditions, promising significant advancements in diagnostic accuracy and efficiency.

Augmented Deep Transfer Learning for SRP Condition Monitoring via Physically-Informed WGAN-GP Approach

Abstract Traditional dynamometer card sensors are costly and complex, making them unsuitable for real-time sucker rod pumping (SRP) well diagnostics. Recently, SRP diagnosis models using motor power curves offer an alternative, but irregular power curves and limited labeled data present challenges. To address this, we propose a deep transfer model enhanced by Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP) data augmentation. First, a physics-driven model reconstructs polished rod torque from motor power, emphasizing fault features. Second, WGAN-GP generates faulty SRP torque-displacement samples to expand training data. Finally, a fully parameter-tuned deep transfer SRP diagnosis framework is established, which improves the automatic learning of advanced fault features and enhances diagnostic accuracy using the augmented dataset. Experiments confirm the model's superior performance and generalization.

BIOM-67. DIFFERENTIAL DNA METHYLATION PATTERNS IN THE CSF OF PATIENTS WITH DIFFUSE GLIOMAS

Abstract Studies have shown that different central nervous system (CNS) tumor types exhibit unique DNA methylation profiles. DNA methylation-based machine learning classification of CNS tumor tissue has been demonstrated to be highly-accurate in differentiating between tumor types, including between diffuse glioma subtypes. While similar analyses have been performed using cell-free DNA (cfDNA) from blood, few studies have looked at the potential of DNA methylation as a biomarker for diffuse gliomas using CSF-derived cfDNA. In this study, we analyzed CSF cfDNA methylation patterns in diffuse glioma patients. DNA methylation profiling of CSF of patients with glioblastoma, IDH-wildtype (GBM) (n=37), IDH-mutant diffuse gliomas (n=12), and non-neoplastic conditions (n=5) was performed using cell-free methylated DNA immunoprecipitation and high-throughput sequencing (cfMeDIP-seq). Methylation quantification was estimated for each genomic region. Significance testing was performed between diagnostic groups to discover differentially-methylated regions (DMRs). The 300 most differentially-methylated regions (DMRs) for each comparison (IDH-wildtype vs. IDH-mutant, IDH-mutant vs. control, IDH-wildtype vs. control), for a total of 900 DMRs, were utilized as features in dimensionality reduction analysis and machine learning. Out of ~1×107 defined genomic regions, there were ~7×105 DMRs between GBM and control samples and ~1.5×106 DMRs between GBM and IDH-mutant samples, accounting for ~7% and ~15% of the genome, respectively. Uniform manifold approximation and projection (UMAP) and hierarchical clustering analysis revealed that samples were mostly clustered by diagnosis. Random forest analysis of the 900 DMRs with an 80/20 train/test split resulted in an overall accuracy of 0.815. In conclusion, CSF cfDNA methylation profiles of diffuse glioma patients exhibit differential methylation patterns between IDH-mutant and IDH-wildtype tumors, as well as between tumor and control, and could be informative biomarkers for the diagnosis of patients with diffuse gliomas.

Recent kernel methods for interacting particle systems: first numerical results

Abstract Interacting particle systems (IPSs) are a very important class of dynamical systems, arising in different domains like biology, physics, sociology and engineering. In many applications, these systems can be very large, making their simulation and control, as well as related numerical tasks, very challenging. Kernel methods, a powerful tool in machine learning, offer promising approaches for analyzing and managing IPS. This paper provides a comprehensive study of applying kernel methods to IPS, including the development of numerical schemes and the exploration of mean-field limits. We present novel applications and numerical experiments demonstrating the effectiveness of kernel methods for surrogate modelling and state-dependent feature learning in IPS. Our findings highlight the potential of these methods for advancing the study and control of large-scale IPS.

Shufflenetv2UNet: An improved neural network model for grassland sample coverage extraction

AbstractAccurate extraction of grassland sample coverage is crucial for regional ecological environment monitoring. Due to the strong feature learning capability, high flexibility, and scalability of deep learning methods, they have great potential in grassland sample extraction modelling. However, we still lack a model that can achieve both lightweight structure and effective performance for small object segmentation to considering the small target characteristics of grassland vegetation and the requirements for model deployment in later stages. Here, we combined the UNet model, which performs well in small target segmentation, with the lightweight network Shufflenetv2 model, proposing an improved UNet neural network, Shufflenetv2UNet, for grassland sample coverage extraction. The core of Shufflenetv2UNet is the removal of maximum pooling and double‐layer convolution modules from downsampling in the UNet neural network. In addition, the Inverted Residual Block structure module from Shufflenetv2 was added to achieve a lightweight model and improved extraction accuracy. The Shufflenetv2UNet achieves an accuracy of 98.23%, with a parameter size of 50.74 M, and a model inference speed of 0.004 s. Compared to existing extraction methods, this model has advantages in prediction accuracy, parameter size, and model inference speed. Moreover, Shufflenetv2UNet achieved different types of grassland sample coverage extractions, with good robustness, generalization, and universality, enabling investigators to quickly and accurately obtain grassland sample coverage. This allows more dynamic and accurate ground measurement data for regional grassland environmental monitoring.

A novel benign and malignant classification model for lung nodules based on multi-scale interleaved fusion integrated network

AbstractOne of the precursors of lung cancer is the presence of lung nodules, and accurate identification of their benign or malignant nature is important for the long-term survival of patients. With the development of artificial intelligence, deep learning has become the main method for lung nodule classification. However, successful deep learning models usually require large number of parameters and carefully annotated data. In the field of medical images, the availability of such data is usually limited, which makes deep networks often perform poorly on new test data. In addition, the model based on the linear stacked single branch structure hinders the extraction of multi-scale features and reduces the classification performance. In this paper, to address this problem, we propose a lightweight interleaved fusion integration network with multi-scale feature learning modules, called MIFNet. The MIFNet consists of a series of MIF blocks that efficiently combine multiple convolutional layers containing 1 × 1 and 3 × 3 convolutional kernels with shortcut links to extract multiscale features at different levels and preserving them throughout the block. The model has only 0.7 M parameters and requires low computational cost and memory space compared to many ImageNet pretrained CNN architectures. The proposed MIFNet conducted exhaustive experiments on the reconstructed LUNA16 dataset, achieving impressive results with 94.82% accuracy, 97.34% F1 value, 96.74% precision, 97.10% sensitivity, and 84.75% specificity. The results show that our proposed deep integrated network achieves higher performance than pre-trained deep networks and state-of-the-art methods. This provides an objective and efficient auxiliary method for accurately classifying the type of lung nodule in medical images.

Attention Pyramid Networks for Object Detection With Semantic Information Fusion

Pyramid context information fusion has limitations in distinguishing background and important semantic information. The attention mechanism solves the limitations, but the detection effect for small target detection and occluded object detection is not good. In order to solve the above problems, a target detection method based on attention pyramid information fusion (APIFNet) is proposed. It consists of Dynamic Pyramid Attention Fusion Module (DPAF) and Attention Semantic Contextual Information Module (ACM). First, the DPAF module fuses information at different scales and simultaneously guides the learning of low-level features by high-level features to enhance semantic information and spatial details. Then, in order to detect small targets and occluded objects, the ACM module effectively enhances the detection ability of small targets by emphasizing the importance of foreground contextual semantic information and suppressing unimportant semantic information. Ultimately, APIFNet outperforms other methods in terms of evaluation performance on the COCO and PASCAL VOC data sets.

Empowering Urdu sentiment analysis: an attention-based stacked CNN-Bi-LSTM DNN with multilingual BERT

Sentiment analysis (SA) as a research field has gained popularity among the researcher throughout the globe over the past 10 years. Deep neural networks (DNN) and word vector models are employed nowadays and perform well in sentiment analysis. Among the different deep neural networks utilized for SA globally, Bi-directional long short-term memory (Bi-LSTM), BERT, and CNN models have received much attention. Even though these models can process a wide range of text types, Because DNNs treat different features the same, using these models in the feature learning phase of a DNN model leads to the creation of a feature space with very high dimensionality. We suggest an attention-based, stacked, two-layer CNN-Bi-LSTM DNN to overcome these glitches. After local feature extraction, by applying stacked two-layer Bi-LSTM, our proposed model extracted coming and outgoing sequences by seeing sequential data streams in backward and forward directions. The output of the stacked two-layer Bi-LSTM is supplied to the attention layer to assign various words with varying values. A second Bi-LSTM layer is constructed atop the initial layer in the suggested network to increase performance. Various experiments have been conducted to evaluate the effectiveness of our proposed model on two Urdu sentiment analysis datasets named as UCSA-21 and UCSA, and an accuracies of 83.12% and 78.91% achieved, respectively.