Information Extraction Research Articles

Revolutionizing Cytology and Cytopathology with Natural Language Processing and Chatbot Technologies: A Narrative Review on Current Trends and Future Directions.

The application of chatbots and Natural Language Processing (NLP) in cytology and cytopathology is an emerging field, which is currently characterized by a limited but growing body of research. Here, a narrative review has been proposed utilizing a standardized checklist and quality control procedure for including scientific papers. This narrative review explores the early developments and potential future impact of these technologies in medical diagnostics. The current literature, comprising 11 studies (after excluding comments, letters, and editorials) suggests that chatbots and NLP offer significant opportunities to enhance diagnostic accuracy, streamline clinical workflows, and improve patient engagement. By automating the extraction and classification of medical information, these technologies can reduce human error and increase precision. They also promise to make patient information more accessible and facilitate complex decision-making processes, thereby fostering greater patient involvement in healthcare. Despite these promising prospects, several challenges need to be addressed for the full potential of these technologies to be realized. These include the need for data standardization, mitigation of biases in Artificial Intelligence (AI) systems, and comprehensive clinical validation. Furthermore, ethical, privacy, and legal considerations must be navigated carefully to ensure responsible AI deployment. Compared to the more established fields of histology, histopathology, and especially radiology, the integration of digital tools in cytology and cytopathology is still in its infancy. Building on the advancements in related fields, especially radiology's experience with digital integration, where these technologies already offer promising solutions in mentoring, second opinions, and education, we can leverage this knowledge to further develop chatbots and natural language processing in cytology and cytopathology. Overall, this review underscores the transformative potential of these technologies while outlining the critical areas for future research and development.

WGAN-based multi-structure segmentation of vertebral cross-section MRI using ResU-Net and clustered transformer

The vertebral foramen, lamina, and vertebral body are three critical components of the spine structure, essential for maintaining spinal connectivity and stability. Accurately segmenting lumbar structures such as the vertebral body, vertebral foramen, and lamina in MRI cross-sections helps doctors better understand and diagnose the pathological causes of spine-related diseases. This study presents a multi-structure semantic segmentation method for vertebral transverse section MRI slices using WGAN with a residual U-Net and clustered Transformer. The generator network was replaced with a combination of a residual U-Net and a clustered Transformer-based segmentation network. The enhanced U-Net encoder, utilizing dilated convolutions and residual structures, improved multi-scale feature extraction capabilities. Meanwhile, the clustered Transformer structure, with reduced progressive linear complexity, ensured the extraction of global positional information. The results of multiple experiments show that the Dice coefficient for vertebral body segmentation increased by 3.1%, the Hausdorff distance decreased by 0.6 mm, mIOU improved by 4.1–96.2%, and PPV increased by 2.0–98.8% compared to mainstream segmentation models. These improvements are statistically significant (p < 0.05).Ablation experiments further validated the effectiveness of the proposed enhanced modules in improving segmentation accuracy for the three target structures.

Art Pattern Design Based on Visual Semantic Segmentation and Evolutionary Algorithm Optimization

This paper aims to address the issue of traditional art patterns not fully leveraging contextual information in image semantic segmentation. It proposes a multi-level feature aggregation method for image semantic segmentation based on visual Transformer and coordinate attention adjustment (CA Adjustment). Initially, the input image is segmented into slices for linear projection, incorporating learnable positional embeddings to derive an encoded input sequence. Subsequently, employing a visual Transformer-based encoder, the image is transformed into patches to capture global context across the network. CA Adjustment is then introduced to adaptively merge fusion features and visual backbone features using weighted adjustments. Lastly, a multi-objective grasshopper optimization algorithm is designed based on a clustering evolution mechanism to further enhance algorithm performance. Extensive experimental results demonstrate that our proposed algorithm effectively models global contextual information for image feature extraction. Compared to existing advanced algorithms, our method achieves high segmentation accuracy in semantic segmentation tasks, consistently exceeding 95%. Ablation experiments further validate the efficacy of our approach, underscoring its broad applicability in high-precision image semantic segmentation.

Exploration of an intrinsically explainable self-attention based model for prototype generation on single-channel EEG sleep stage classification

Prototype-based methods in deep learning offer interpretable explanations for decisions by comparing inputs to typical representatives in the data. This study explores the adaptation of SESM, a self-attention-based prototype method successful in electrocardiogram (ECG) tasks, for electroencephalogram (EEG) signals. The architecture is evaluated on sleep stage classification, exploring its efficacy in predicting stages with single-channel EEG. The model achieves comparable test accuracy compared to EEGNet, a state-of-the-art black-box architecture for EEG classification. The generated prototypical components are exaimed qualitatively and using the area over the perterbation curve (AOPC) indicate some alignment with expected bio-markers for different sleep stages such as alpha spindles and slow waves in non-REM sleep, but the results are severely limited by the model’s ability to only extract and present information in the time-domain. Ablation studies are used to explore the impact of kernel size, number of heads, and diversity threshold on model performance and explainability. This study represents the first application of a self-attention based prototype method to EEG data and provides a step forward in explainable AI for EEG data analysis.

Vision-guided crack identification and size quantification framework for dam underwater concrete structures

Remotely operated vehicles with cameras provide a non-contact inspection solution for dam underwater structural defect detection. However, manual information extraction methods suffer from problems like labor costs and high misjudgment. This study proposes an integrated dam underwater crack identification and size quantification framework using machine vision and deep learning. First, a real-time automatic detection framework for dam underwater cracks is built via You Only Look Once v5 (YOLOv5), and four different backbone detectors are introduced to balance detection accuracy and speed. Then, the Swin-Transformer module is inserted into the YOLOv5 model to enhance its feature extraction capability and small object detection capability. Next, a method for measuring the true size of cracks was constructed based on deep learning and infrared laser rangefinders. In this study, physical model experiments and actual engineering projects are combined to validate the generalization capability of the proposed crack identification and size quantification method. Experimental results show that the Swin-Transformer-based YOLOv5 model effectively balances detection accuracy and speed with a precision of 0.986, a recall of 0.979, a mean average precision of 0.985, and a frame rate of 68 frames per second in detecting underwater crack images with 768 × 576 pixels. In addition, the proposed method can accurately identify and detect cracks in complex underwater scenes, including obstacle interference, tilt shooting angle, uneven illumination, and turbid water scenarios. Moreover, the method proposed in this paper can quantify the overall size and geometric parameters of underwater cracks with relatively small errors.

Network Embedding Technology Based on Breadth Learning for Information Extraction and Review in Social Media

Network Embedding Technology Based on Breadth Learning for Information Extraction and Review in Social Media

SentinelGuard Pro: Deploying Cutting‐Edge FusionNet for Unerring Detection and Enforcement of Wrong Parking Incidents

ABSTRACTWrong parking incidents pose a pervasive challenge in urban environments, disrupting the smooth flow of traffic, compromising safety and contributing to various logistical issues. Unauthorized parking occurs when vehicles are parked in locations not designated for such purposes, leading to a myriad of problems for both authorities and the general public. This research introduces a pioneering approach to confront the persistent challenge of unauthorized parking incidents in urban environments. The study focuses on harnessing the advanced capabilities of the FusionNet model to enhance the accuracy of license plate detection. This paper introduces the YOLO v8 Model, a deep learning architecture designed to enhance urban parking management by accurately detecting vehicles parked in unauthorized slots. The objective is to enhance parking management efficiency by accurately detecting vehicles and their occupancy status in designated parking areas. The methodology begins with data collection and preprocessing of images of parking spaces, followed by the training of YOLO v8 to identify vehicles and parking spaces in real time. Leveraging a diverse dataset encompassing various parking scenarios, including instances of unauthorized parking, the model achieves an accuracy of 98.50% in identifying vehicles outside designated areas. This model segments characters from detected license plates, enabling the accurate extraction of alphanumeric information associated with each vehicle. The integrated system provides timely identification of parking violations and facilitates effective enforcement actions through captured license plate data. Results demonstrate the model's effectiveness in real‐world scenarios, showcasing its potential for improving urban safety and efficiency. The implementation of FusionNet in the Python programming language, the proposed solution aims to streamline parking management, improve compliance with parking regulations and enhance overall urban mobility., with robust precision 96.17%, specificity 97.42% and sensitivity 96.19%, surpassing other MobileNet, CNN, ANN, DNN and EfficientNet models.

DocNER: Document Chat Assistance with NER

The increasing amounts of high volume unstructured medical data, such as research and clinical articles, impede efficient information extraction. Since the medical texts have a complex structure, many times these approaches do not work, leading to significant losses. The current approaches are not robust to a wide spectrum of medical content as they are too specific to the given domain.This study proposes a recent innovative approach that involves the combination of Named Entity Recognition (NER) and Retrieval-Augmented Generation (RAG) to enhance entity extraction and give useful bibliographic information from medical data respectively. The objectives of our method are to improve accuracy, accommodate different kinds of documents, and assist in clinical research and patient care.

Chemometrics: A Vital Implement for Understanding the Water Structures by Near‐Infrared Spectroscopy

ABSTRACTWater structures take an important role in chemical and biological systems, because the structure and function of a molecule may depend on the structure of water with which the molecule interacts. Near‐infrared (NIR) spectroscopy has been proven to be powerful in analyzing the structure of water due to its sensitive response to OH. However, chemometrics is vitally important in the analysis of NIR spectrum of water due to the low resolution of the spectrum and the complexity of the water structures. In this review, chemometric methods for structural analysis of water in aqueous systems, particularly in chemical and biological processes, by NIR spectroscopy were summarized, from the improvement of spectral resolution to the effective extraction of the spectral information of different water structures. Through the changes of the spectral features of the water structures, the structural transformation of proteins, thermo‐responsive polymers, antifreeze agents, as well as the structural variation of water in the transformation were elucidated. Water was proved to be a good probe for analyzing the structure and interactions in aqueous solutions and chemical/biological processes by NIR spectroscopy.

Global, regional, and national burden of ischemic stroke attributable to active smoking, 1990-2021.

Ischemic stroke is a major global health issue, with active smoking identified as a key modifiable risk factor. This study examines the burden of ischemic stroke due to active smoking from 1990 to 2021, across different sociodemographic contexts. Data from the Global Burden of Disease (GBD) 2021 database were used to extract information on mortality and disability-adjusted life years (DALYs) attributable to active smoking-related ischemic stroke. Countries and regions were categorized by the sociodemographic index (SDI) into five levels. Statistical analyses were conducted using R Studio, including the calculation of estimated annual percentage change (EAPC) and joinpoint regression models. In 2021, there were 342674 deaths globally due to ischemic stroke caused by active smoking, with an age-standardized mortality rate (ASMR) of 4.06 and a population-attributable fraction (PAF) of 9.54%. The number of deaths increased by 35.59% from 1990 to 2021, with males aged ≥70 years experiencing the largest increase. The global age-standardized DALY rate in 2021 was 98.29, with an overall increase in DALYs by 33.55% from 1990. Regional analysis revealed significant disparities, with the middle SDI region reporting the highest number of deaths and DALYs, while the high SDI region reported the lowest. Geographically, East Asia had the highest burden in 2021. Nationally, China had the highest number of deaths and DALYs due to smoking-related ischemic stroke. This study highlights the significant global burden of ischemic stroke attributable to active smoking and the critical need for targeted smoking cessation programs and stroke prevention strategies. Despite overall declines in ASMR and age-standardized DALY rates, the burden varies significantly across different regions and sociodemographic groups. Effective public health interventions, particularly in low- to middle-SDI regions, are essential to mitigate the impact of smoking-related ischemic stroke and improve global health outcomes.

Comprehensive Characterization of Oxidative Stress-Modulating Chemicals Using GPT-Based Text Mining.

The screening of hazardous environmental pollutants is hindered by the limited availability of toxicological databases. Large language model (LLM)-based text mining holds the potential to automatically extract complex toxicological information from the literature. Due to its relevance to diseases and the challenge of comprehensive characterization, oxidative stress serves as a suitable case for research by texting mining. In this study, a robust workflow utilizing a LLM (i.e., GPT-4) was developed to extract information on oxidative stress tests, including data collection, text preprocessing, prompt engineering, and performance evaluation procedures. A total of 17,780 relevant records were extracted from 7166 articles, covering 2558 unique compounds. A rising interest in oxidative stress was observed over the past two decades. A list of known prooxidants (n = 1416) and antioxidants (n = 1102) was established, with the leading chemical categories being pharmaceuticals, pesticides, and metals for prooxidants and pharmaceuticals and flavonoids for antioxidants. Structural alert analysis identified potential prooxidant (e.g., chlorobenzene, nitrobenzene, and tertiary amines) and antioxidant (e.g., flavonoid and thiol) substructures. These findings illustrate the feasibility of building toxicological databases through LLM-based text mining in a cost-efficient manner, and the information obtained from the technique holds significant promise for future applications in environmental and health research.

Causal representation learning through higher-level information extraction

The large gap between the generalization level of state-of-the-art machine learning and human learning systems calls for the development of artificial intelligence (AI) models that are truly inspired by human cognition. In tasks related to image analysis, searching for pixel-level regularities has reached a power of information extraction still far from what humans capture with image-based observations. This leads to poor generalization when even small shifts occur at the level of the observations. We explore a perspective on this problem that is directed to learning the generative process with causality-related foundations, using models capable of combining symbolic manipulation, probabilistic reasoning, and pattern recognition abilities. We briefly review and explore connections of research from machine learning, cognitive science, and related fields of human behavior to support our perspective for the direction to more robust and human-like artificial learning systems.

Knowledge-embedded multi-layer collaborative adaptive fusion network: Addressing challenges in foggy conditions and complex imaging

Infrared and visible image fusion aims at generating high-quality images that serve both human and machine visual perception under extreme imaging conditions. However, current fusion methods primarily rely on datasets comprising infrared and visible images captured under clear weather conditions. When applied to real-world scenarios, image fusion tasks inevitably encounter challenges posed by adverse weather conditions such as heavy fog, resulting in difficulties in obtaining effective information and inferior visual perception. To address these challenges, this paper proposes a Mean Teacher-based Self-supervised Image Restoration and multimodal Image Fusion joint learning network (SIRIFN), which enhances the robustness of the fusion network in adverse weather conditions by employing deep supervision from a guiding network to the learning network. Furthermore, to enhance the network’s information extraction and integration capabilities, our Multi-level Feature Collaborative adaptive Reconstruction Network (MFCRNet) is introduced, which adopts a multi-branch, multi-scale design, with differentiated processing strategies for different features. This approach preserves rich texture information while maintaining semantic consistency from the source images. Extensive experiments demonstrate that SIRIFN outperforms current state-of-the-art algorithms in both visual quality and quantitative evaluation. Specifically, the joint implementation of image restoration and multimodal fusion provides more effective information for visual tasks under extreme weather conditions, thereby facilitating downstream visual tasks.

Practical hardware demonstration of a multi-sensor goal-oriented semantic signal processing and communications network

Recent advancements in machine learning, particularly real-time extraction of rich semantic information, reshape signal processing techniques and related hardware architectures. To address the highly challenging requirements of next-generation signal processing applications in networked platforms, we investigate low-power hardware implementation alternatives for a multi-sensor, goal-oriented semantic communications network. Specifically, we focus on cost-effective Raspberry Pis in a multi-sensor semantic video communication application, showcasing adaptability from traditional CPU/GPU configurations. Additionally, we provide a preliminary investigation on implementing semantic extraction tasks through in-memory computation using memristor arrays to further emphasize the potential future of low-power low-cost semantic signal processing. Hardware demonstrations using Raspberry Pi 4Bs and simulations with in-memory computation architectures offer promising hardware architectures with cost-effective and low-power sensor alternatives to the next-generation semantic signal processing applications and semantic communication systems.

Optimizing Healthcare Data Extraction with AI: A Path to Improved Patient Care

This study analyses the role of AI in enhancing the healthcare system by improving data extraction techniques to address various challenges in the healthcare sector. The roles of artificial intelligence (AI), like natural language processing (NLP) and machine learning (ML) in healthcare, allow the automation of the extraction of unstructured clinical information from records. This results in quicker diagnosis, enhanced treatment outcomes, and reduced paperwork. It has allowed for reduced burnout among healthcare professionals who would otherwise extract this data manually, allowing them to focus on providing better quality care to patients. While AI has been implemented in this sector, much must be done to achieve its full potential.

Towards Integrating Federated Learning with Split Learning via Spatio-temporal Graph Framework for Brain Disease Prediction.

Functional Magnetic Resonance Imaging (fMRI) is used for extracting blood oxygen signals from brain regions to map brain functional connectivity for brain disease prediction. Despite its effectiveness, fMRI has not been widely used: on the one hand, collecting and labeling the data is time-consuming and costly, which limits the amount of valid data collected at a single healthcare site; on the other hand, integrating data from multiple sites is challenging due to data privacy restrictions. To address these issues, we propose a novel, integrated Federated learning and Split learning Spatio-temporal Graph framework (FS2G). Specifically, we introduce federated learning and split learning techniques to split a spatio-temporal model into a client temporal model and a server spatial model. In the client temporal model, we propose a time-aware mechanism to focus on changes in brain functional states and use an InceptionTime model to extract information about changes in the brain states of each subject. In the server spatial model, we propose a united graph convolutional network to integrate multiple graph convolutional networks. Integrating federated learning and split learning, FS2G can utilize multi-site fMRI data without violating data privacy protection and reduce the risk of overfitting as it is capable of learning from limited training data sets. Moreover, it boosts the extraction of spatio-temporal features of fMRI using spatio-temporal graph networks. Experiments on ABIDE and ADHD200 datasets demonstrate that our proposed method outperforms state-of-the-art methods. In addition, we explore biomarkers associated with brain disease prediction using community discovery algorithms using intermediate results of FS2G. The source code is available at https://github.com/yutian0315/FS2G.

An unpaired SAR-to-optical image translation method based on Schrödinger bridge network and multi-scale feature fusion

SAR-to-optical (S2O) translation is able to covert SAR into optical images, which help the interpreter to extract information efficiently. In the absence of strictly matched datasets, it is difficult for existing methods to complete training on unpaired data with a minimum amount of data. By employing the recent Schrödinger bridge-based transformation framework, a multiscale axial residual module (MARM) based on the concept of multi-scale feature fusion has been proposed in this paper. To enable efficient translation of SAR to optical images, the generator and discriminator of the model have been designed. Extensive experiments on the SEN1-2 dataset conducted, and the results show the superiority of the proposed method in terms of the generation quality. Compared with the classical CycleGAN, the proposed method can improve the FID metrics by 42.05%.

Fusing semantic and syntactic information for aspect sentiment triplet extraction

Aspect Sentiment Triplet Extraction (ASTE) aims to extract aspect terms, sentiment polarity and opinion terms explaining the reason for the sentiment from a sentence in the form of triplets. Many existing studies model the context by graph neural networks to learn relevant information from the generated graphs. However, some sentences may have syntactic errors or lack significant grammar, which may lead to poor results on the dataset of the model. In this paper, we propose the Fusing Semantic and Syntactic Information for Aspect Sentiment Triplet Extraction (FSSI) model, which incorporates both syntactic structure and semantic relevance in the context. Specifically, we construct a syntactic graph convolutional network to obtain comprehensive syntactic structure information and a semantic graph convolutional network to obtain global semantic relevance of sentences using the self-attention mechanism. Furthermore, we concatenate the graph representations generated by the two graph convolution networks to obtain the final enhanced representation. Finally, we apply an effective inference strategy to extract triplets. Extensive experimental results on benchmark datasets show that our model outperforms state-of-the-art approaches.

BreakNet: discontinuity-resilient multi-scale transformer segmentation of retinal layers

Visible light optical coherence tomography (vis-OCT) is gaining traction for retinal imaging due to its high resolution and functional capabilities. However, the significant absorption of hemoglobin in the visible light range leads to pronounced shadow artifacts from retinal blood vessels, posing challenges for accurate layer segmentation. In this study, we present BreakNet, a multi-scale Transformer-based segmentation model designed to address boundary discontinuities caused by these shadow artifacts. BreakNet utilizes hierarchical Transformer and convolutional blocks to extract multi-scale global and local feature maps, capturing essential contextual, textural, and edge characteristics. The model incorporates decoder blocks that expand pathways to enhance the extraction of fine details and semantic information, ensuring precise segmentation. Evaluated on rodent retinal images acquired with prototype vis-OCT, BreakNet demonstrated superior performance over state-of-the-art segmentation models, such as TCCT-BP and U-Net, even when faced with limited-quality ground truth data. Our findings indicate that BreakNet has the potential to significantly improve retinal quantification and analysis.

Gearbox Fault Diagnosis Based on Adaptive Variational Mode Decomposition-Stationary Wavelet Transform and Ensemble Refined Composite Multiscale Fluctuation Dispersion Entropy.

Existing gearbox fault diagnosis methods are prone to noise interference and cannot extract comprehensive fault signals, leading to misdiagnosis or missed diagnosis. This paper proposes a method for gearbox fault diagnosis based on adaptive variational mode decomposition-stationary wavelet transform (AVMD-SWT) and ensemble refined composite multiscale fluctuation dispersion entropy (ERCMFDE). Initially, the kurtosis coefficient and autocorrelation coefficient are presented, and the Intrinsic Mode Functions are denoised through the application of AVMD-SWT. Secondly, the coarse-grained processing method of composite multiscale fluctuation dispersion entropy is extended to encompass three additional approaches: first-order central moment, second-order central moment, and third-order central moment. This enables the comprehensive extraction of feature information from the time series, thereby facilitating the formation of an initial hybrid feature set. Subsequently, recursive feature elimination (RFE) is employed for feature selection. Ultimately, the outcomes of the faults diagnoses are derived through the utilization of a Support Vector Machine with a Sparrow Search Algorithm (SSA-SVM), with the actual faults data collection and analysis conducted on an experimental platform for gearbox fault diagnosis. The experiments demonstrate that the method can accurately identify gearbox faults and achieve a high diagnostic accuracy of 98.78%.