Myocardial perfusion imaging(MPI),which uses single-photon emission computed tomography(SPECT),is a well-known estimating tool for medical diagnosis,employing the classification of images to show situations in coronar...Myocardial perfusion imaging(MPI),which uses single-photon emission computed tomography(SPECT),is a well-known estimating tool for medical diagnosis,employing the classification of images to show situations in coronary artery disease(CAD).The automatic classification of SPECT images for different techniques has achieved near-optimal accuracy when using convolutional neural networks(CNNs).This paper uses a SPECT classification framework with three steps:1)Image denoising,2)Attenuation correction,and 3)Image classification.Image denoising is done by a U-Net architecture that ensures effective image denoising.Attenuation correction is implemented by a convolution neural network model that can remove the attenuation that affects the feature extraction process of classification.Finally,a novel multi-scale diluted convolution(MSDC)network is proposed.It merges the features extracted in different scales and makes the model learn the features more efficiently.Three scales of filters with size 3×3 are used to extract features.All three steps are compared with state-of-the-art methods.The proposed denoising architecture ensures a high-quality image with the highest peak signal-to-noise ratio(PSNR)value of 39.7.The proposed classification method is compared with the five different CNN models,and the proposed method ensures better classification with an accuracy of 96%,precision of 87%,sensitivity of 87%,specificity of 89%,and F1-score of 87%.To demonstrate the importance of preprocessing,the classification model was analyzed without denoising and attenuation correction.展开更多
In radiology,magnetic resonance imaging(MRI)is an essential diagnostic tool that provides detailed images of a patient’s anatomical and physiological structures.MRI is particularly effective for detecting soft tissue...In radiology,magnetic resonance imaging(MRI)is an essential diagnostic tool that provides detailed images of a patient’s anatomical and physiological structures.MRI is particularly effective for detecting soft tissue anomalies.Traditionally,radiologists manually interpret these images,which can be labor-intensive and time-consuming due to the vast amount of data.To address this challenge,machine learning,and deep learning approaches can be utilized to improve the accuracy and efficiency of anomaly detection in MRI scans.This manuscript presents the use of the Deep AlexNet50 model for MRI classification with discriminative learning methods.There are three stages for learning;in the first stage,the whole dataset is used to learn the features.In the second stage,some layers of AlexNet50 are frozen with an augmented dataset,and in the third stage,AlexNet50 with an augmented dataset with the augmented dataset.This method used three publicly available MRI classification datasets:Harvard whole brain atlas(HWBA-dataset),the School of Biomedical Engineering of Southern Medical University(SMU-dataset),and The National Institute of Neuroscience and Hospitals brain MRI dataset(NINS-dataset)for analysis.Various hyperparameter optimizers like Adam,stochastic gradient descent(SGD),Root mean square propagation(RMS prop),Adamax,and AdamW have been used to compare the performance of the learning process.HWBA-dataset registers maximum classification performance.We evaluated the performance of the proposed classification model using several quantitative metrics,achieving an average accuracy of 98%.展开更多
Many fields,such as neuroscience,are experiencing the vast prolife ration of cellular data,underscoring the need fo r organizing and interpreting large datasets.A popular approach partitions data into manageable subse...Many fields,such as neuroscience,are experiencing the vast prolife ration of cellular data,underscoring the need fo r organizing and interpreting large datasets.A popular approach partitions data into manageable subsets via hierarchical clustering,but objective methods to determine the appropriate classification granularity are missing.We recently introduced a technique to systematically identify when to stop subdividing clusters based on the fundamental principle that cells must differ more between than within clusters.Here we present the corresponding protocol to classify cellular datasets by combining datadriven unsupervised hierarchical clustering with statistical testing.These general-purpose functions are applicable to any cellular dataset that can be organized as two-dimensional matrices of numerical values,including molecula r,physiological,and anatomical datasets.We demonstrate the protocol using cellular data from the Janelia MouseLight project to chara cterize morphological aspects of neurons.展开更多
Machine learning(ML)is increasingly applied for medical image processing with appropriate learning paradigms.These applications include analyzing images of various organs,such as the brain,lung,eye,etc.,to identify sp...Machine learning(ML)is increasingly applied for medical image processing with appropriate learning paradigms.These applications include analyzing images of various organs,such as the brain,lung,eye,etc.,to identify specific flaws/diseases for diagnosis.The primary concern of ML applications is the precise selection of flexible image features for pattern detection and region classification.Most of the extracted image features are irrelevant and lead to an increase in computation time.Therefore,this article uses an analytical learning paradigm to design a Congruent Feature Selection Method to select the most relevant image features.This process trains the learning paradigm using similarity and correlation-based features over different textural intensities and pixel distributions.The similarity between the pixels over the various distribution patterns with high indexes is recommended for disease diagnosis.Later,the correlation based on intensity and distribution is analyzed to improve the feature selection congruency.Therefore,the more congruent pixels are sorted in the descending order of the selection,which identifies better regions than the distribution.Now,the learning paradigm is trained using intensity and region-based similarity to maximize the chances of selection.Therefore,the probability of feature selection,regardless of the textures and medical image patterns,is improved.This process enhances the performance of ML applications for different medical image processing.The proposed method improves the accuracy,precision,and training rate by 13.19%,10.69%,and 11.06%,respectively,compared to other models for the selected dataset.The mean error and selection time is also reduced by 12.56%and 13.56%,respectively,compared to the same models and dataset.展开更多
In response to the scarcity of infrared aircraft samples and the tendency of traditional deep learning to overfit,a few-shot infrared aircraft classification method based on cross-correlation networks is proposed.This...In response to the scarcity of infrared aircraft samples and the tendency of traditional deep learning to overfit,a few-shot infrared aircraft classification method based on cross-correlation networks is proposed.This method combines two core modules:a simple parameter-free self-attention and cross-attention.By analyzing the self-correlation and cross-correlation between support images and query images,it achieves effective classification of infrared aircraft under few-shot conditions.The proposed cross-correlation network integrates these two modules and is trained in an end-to-end manner.The simple parameter-free self-attention is responsible for extracting the internal structure of the image while the cross-attention can calculate the cross-correlation between images further extracting and fusing the features between images.Compared with existing few-shot infrared target classification models,this model focuses on the geometric structure and thermal texture information of infrared images by modeling the semantic relevance between the features of the support set and query set,thus better attending to the target objects.Experimental results show that this method outperforms existing infrared aircraft classification methods in various classification tasks,with the highest classification accuracy improvement exceeding 3%.In addition,ablation experiments and comparative experiments also prove the effectiveness of the method.展开更多
The World Wide Web provides a wealth of information about everything, including contemporary audio and visual art events, which are discussed on media outlets, blogs, and specialized websites alike. This information m...The World Wide Web provides a wealth of information about everything, including contemporary audio and visual art events, which are discussed on media outlets, blogs, and specialized websites alike. This information may become a robust source of real-world data, which may form the basis of an objective data-driven analysis. In this study, a methodology for collecting information about audio and visual art events in an automated manner from a large array of websites is presented in detail. This process uses cutting edge Semantic Web, Web Search and Generative AI technologies to convert website documents into a collection of structured data. The value of the methodology is demonstrated by creating a large dataset concerning audiovisual events in Greece. The collected information includes event characteristics, estimated metrics based on their text descriptions, outreach metrics based on the media that reported them, and a multi-layered classification of these events based on their type, subjects and methods used. This dataset is openly provided to the general and academic public through a Web application. Moreover, each event’s outreach is evaluated using these quantitative metrics, the results are analyzed with an emphasis on classification popularity and useful conclusions are drawn concerning the importance of artistic subjects, methods, and media.展开更多
The rapid growth of machine learning(ML)across fields has intensified the challenge of selecting the right algorithm for specific tasks,known as the Algorithm Selection Problem(ASP).Traditional trial-and-error methods...The rapid growth of machine learning(ML)across fields has intensified the challenge of selecting the right algorithm for specific tasks,known as the Algorithm Selection Problem(ASP).Traditional trial-and-error methods have become impractical due to their resource demands.Automated Machine Learning(AutoML)systems automate this process,but often neglect the group structures and sparsity in meta-features,leading to inefficiencies in algorithm recommendations for classification tasks.This paper proposes a meta-learning approach using Multivariate Sparse Group Lasso(MSGL)to address these limitations.Our method models both within-group and across-group sparsity among meta-features to manage high-dimensional data and reduce multicollinearity across eight meta-feature groups.The Fast Iterative Shrinkage-Thresholding Algorithm(FISTA)with adaptive restart efficiently solves the non-smooth optimization problem.Empirical validation on 145 classification datasets with 17 classification algorithms shows that our meta-learning method outperforms four state-of-the-art approaches,achieving 77.18%classification accuracy,86.07%recommendation accuracy and 88.83%normalized discounted cumulative gain.展开更多
Segmenting a breast ultrasound image is still challenging due to the presence of speckle noise,dependency on the operator,and the variation of image quality.This paper presents the UltraSegNet architecture that addres...Segmenting a breast ultrasound image is still challenging due to the presence of speckle noise,dependency on the operator,and the variation of image quality.This paper presents the UltraSegNet architecture that addresses these challenges through three key technical innovations:This work adds three things:(1)a changed ResNet-50 backbone with sequential 3×3 convolutions to keep fine anatomical details that are needed for finding lesion boundaries;(2)a computationally efficient regional attention mechanism that works on high-resolution features without using a transformer’s extra memory;and(3)an adaptive feature fusion strategy that changes local and global featuresbasedonhowthe image isbeing used.Extensive evaluation on two distinct datasets demonstrates UltraSegNet’s superior performance:On the BUSI dataset,it obtains a precision of 0.915,a recall of 0.908,and an F1 score of 0.911.In the UDAIT dataset,it achieves robust performance across the board,with a precision of 0.901 and recall of 0.894.Importantly,these improvements are achieved at clinically feasible computation times,taking 235 ms per image on standard GPU hardware.Notably,UltraSegNet does amazingly well on difficult small lesions(less than 10 mm),achieving a detection accuracy of 0.891.This is a huge improvement over traditional methods that have a hard time with small-scale features,as standard models can only achieve 0.63–0.71 accuracy.This improvement in small lesion detection is particularly crucial for early-stage breast cancer identification.Results from this work demonstrate that UltraSegNet can be practically deployable in clinical workflows to improve breast cancer screening accuracy.展开更多
As the trend to use the latestmachine learning models to automate requirements engineering processes continues,security requirements classification is tuning into the most researched field in the software engineering ...As the trend to use the latestmachine learning models to automate requirements engineering processes continues,security requirements classification is tuning into the most researched field in the software engineering community.Previous literature studies have proposed numerousmodels for the classification of security requirements.However,adopting those models is constrained due to the lack of essential datasets permitting the repetition and generalization of studies employing more advanced machine learning algorithms.Moreover,most of the researchers focus only on the classification of requirements with security keywords.They did not consider other nonfunctional requirements(NFR)directly or indirectly related to security.This has been identified as a significant research gap in security requirements engineering.The major objective of this study is to propose a security requirements classification model that categorizes security and other relevant security requirements.We use PROMISE_exp and DOSSPRE,the two most commonly used datasets in the software engineering community.The proposed methodology consists of two steps.In the first step,we analyze all the nonfunctional requirements and their relation with security requirements.We found 10 NFRs that have a strong relationship with security requirements.In the second step,we categorize those NFRs in the security requirements category.Our proposedmethodology is a hybridmodel based on the ConvolutionalNeural Network(CNN)and Extreme Gradient Boosting(XGBoost)models.Moreover,we evaluate the model by updating the requirement type column with a binary classification column in the dataset to classify the requirements into security and non-security categories.The performance is evaluated using four metrics:recall,precision,accuracy,and F1 Score with 20 and 28 epochs number and batch size of 32 for PROMISE_exp and DOSSPRE datasets and achieved 87.3%and 85.3%accuracy,respectively.The proposed study shows an enhancement in metrics values compared to the previous literature studies.This is a proof of concept for systematizing the evaluation of security recognition in software systems during the early phases of software development.展开更多
Preservation of the crops depends on early and accurate detection of pests on crops as they cause several diseases decreasing crop production and quality. Several deep-learning techniques have been applied to overcome...Preservation of the crops depends on early and accurate detection of pests on crops as they cause several diseases decreasing crop production and quality. Several deep-learning techniques have been applied to overcome the issue of pest detection on crops. We have developed the YOLOCSP-PEST model for Pest localization and classification. With the Cross Stage Partial Network (CSPNET) backbone, the proposed model is a modified version of You Only Look Once Version 7 (YOLOv7) that is intended primarily for pest localization and classification. Our proposed model gives exceptionally good results under conditions that are very challenging for any other comparable models especially conditions where we have issues with the luminance and the orientation of the images. It helps farmers working out on their crops in distant areas to determine any infestation quickly and accurately on their crops which helps in the quality and quantity of the production yield. The model has been trained and tested on 2 datasets namely the IP102 data set and a local crop data set on both of which it has shown exceptional results. It gave us a mean average precision (mAP) of 88.40% along with a precision of 85.55% and a recall of 84.25% on the IP102 dataset meanwhile giving a mAP of 97.18% on the local data set along with a recall of 94.88% and a precision of 97.50%. These findings demonstrate that the proposed model is very effective in detecting real-life scenarios and can help in the production of crops improving the yield quality and quantity at the same time.展开更多
Flood disasters can have a serious impact on people's production and lives, and can cause hugelosses in lives and property security. Based on multi-source remote sensing data, this study establisheddecision tree c...Flood disasters can have a serious impact on people's production and lives, and can cause hugelosses in lives and property security. Based on multi-source remote sensing data, this study establisheddecision tree classification rules through multi-source and multi-temporal feature fusion, classified groundobjects before the disaster and extracted flood information in the disaster area based on optical imagesduring the disaster, so as to achieve rapid acquisition of the disaster situation of each disaster bearing object.In the case of Qianliang Lake, which suffered from flooding in 2020, the results show that decision treeclassification algorithms based on multi-temporal features can effectively integrate multi-temporal and multispectralinformation to overcome the shortcomings of single-temporal image classification and achieveground-truth object classification.展开更多
Multi-label image classification is a challenging task due to the diverse sizes and complex backgrounds of objects in images.Obtaining class-specific precise representations at different scales is a key aspect of feat...Multi-label image classification is a challenging task due to the diverse sizes and complex backgrounds of objects in images.Obtaining class-specific precise representations at different scales is a key aspect of feature representation.However,existing methods often rely on the single-scale deep feature,neglecting shallow and deeper layer features,which poses challenges when predicting objects of varying scales within the same image.Although some studies have explored multi-scale features,they rarely address the flow of information between scales or efficiently obtain class-specific precise representations for features at different scales.To address these issues,we propose a two-stage,three-branch Transformer-based framework.The first stage incorporates multi-scale image feature extraction and hierarchical scale attention.This design enables the model to consider objects at various scales while enhancing the flow of information across different feature scales,improving the model’s generalization to diverse object scales.The second stage includes a global feature enhancement module and a region selection module.The global feature enhancement module strengthens interconnections between different image regions,mitigating the issue of incomplete represen-tations,while the region selection module models the cross-modal relationships between image features and labels.Together,these components enable the efficient acquisition of class-specific precise feature representations.Extensive experiments on public datasets,including COCO2014,VOC2007,and VOC2012,demonstrate the effectiveness of our proposed method.Our approach achieves consistent performance gains of 0.3%,0.4%,and 0.2%over state-of-the-art methods on the three datasets,respectively.These results validate the reliability and superiority of our approach for multi-label image classification.展开更多
With the rise of encrypted traffic,traditional network analysis methods have become less effective,leading to a shift towards deep learning-based approaches.Among these,multimodal learning-based classification methods...With the rise of encrypted traffic,traditional network analysis methods have become less effective,leading to a shift towards deep learning-based approaches.Among these,multimodal learning-based classification methods have gained attention due to their ability to leverage diverse feature sets from encrypted traffic,improving classification accuracy.However,existing research predominantly relies on late fusion techniques,which hinder the full utilization of deep features within the data.To address this limitation,we propose a novel multimodal encrypted traffic classification model that synchronizes modality fusion with multiscale feature extraction.Specifically,our approach performs real-time fusion of modalities at each stage of feature extraction,enhancing feature representation at each level and preserving inter-level correlations for more effective learning.This continuous fusion strategy improves the model’s ability to detect subtle variations in encrypted traffic,while boosting its robustness and adaptability to evolving network conditions.Experimental results on two real-world encrypted traffic datasets demonstrate that our method achieves a classification accuracy of 98.23% and 97.63%,outperforming existing multimodal learning-based methods.展开更多
The proliferation of Internet of Things(IoT)devices has established edge computing as a critical paradigm for real-time data analysis and low-latency processing.Nevertheless,the distributed nature of edge computing pr...The proliferation of Internet of Things(IoT)devices has established edge computing as a critical paradigm for real-time data analysis and low-latency processing.Nevertheless,the distributed nature of edge computing presents substantial security challenges,rendering it a prominent target for sophisticated malware attacks.Existing signature-based and behavior-based detection methods are ineffective against the swiftly evolving nature of malware threats and are constrained by the availability of resources.This paper suggests the Genetic Encoding for Novel Optimization of Malware Evaluation(GENOME)framework,a novel solution that is intended to improve the performance of malware detection and classification in peripheral computing environments.GENOME optimizes data storage and computa-tional efficiency by converting malware artifacts into compact,structured sequences through a Deoxyribonucleic Acid(DNA)encoding mechanism.The framework employs two DNA encoding algorithms,standard and compressed,which substantially reduce data size while preserving high detection accuracy.The Edge-IIoTset dataset was used to conduct experiments that showed that GENOME was able to achieve high classification performance using models such as Random Forest and Logistic Regression,resulting in a reduction of data size by up to 42%.Further evaluations with the CIC-IoT-23 dataset and Deep Learning models confirmed GENOME’s scalability and adaptability across diverse datasets and algorithms.The potential of GENOME to address critical challenges,such as the rapid mutation of malware,real-time processing demands,and resource limitations,is emphasized in this study.GENOME offers comprehensive protection for peripheral computing environments by offering a security solution that is both efficient and scalable.展开更多
Large amounts of labeled data are usually needed for training deep neural networks in medical image studies,particularly in medical image classification.However,in the field of semi-supervised medical image analysis,l...Large amounts of labeled data are usually needed for training deep neural networks in medical image studies,particularly in medical image classification.However,in the field of semi-supervised medical image analysis,labeled data is very scarce due to patient privacy concerns.For researchers,obtaining high-quality labeled images is exceedingly challenging because it involves manual annotation and clinical understanding.In addition,skin datasets are highly suitable for medical image classification studies due to the inter-class relationships and the inter-class similarities of skin lesions.In this paper,we propose a model called Coalition Sample Relation Consistency(CSRC),a consistency-based method that leverages Canonical Correlation Analysis(CCA)to capture the intrinsic relationships between samples.Considering that traditional consistency-based models only focus on the consistency of prediction,we additionally explore the similarity between features by using CCA.We enforce feature relation consistency based on traditional models,encouraging the model to learn more meaningful information from unlabeled data.Finally,considering that cross-entropy loss is not as suitable as the supervised loss when studying with imbalanced datasets(i.e.,ISIC 2017 and ISIC 2018),we improve the supervised loss to achieve better classification accuracy.Our study shows that this model performs better than many semi-supervised methods.展开更多
Intelligent vehicle applications provide convenience but raise privacy and security concerns.Misuse of sensitive data,including vehicle location,and facial recognition information,poses a threat to user privacy.Hence,...Intelligent vehicle applications provide convenience but raise privacy and security concerns.Misuse of sensitive data,including vehicle location,and facial recognition information,poses a threat to user privacy.Hence,traffic classification is vital for promptly overseeing and controlling applications with sensitive information.In this paper,we propose ETNet,a framework that combines multiple features and leverages self-attention mechanisms to learn deep relationships between packets.ET-Net employs a multisimilarity triplet network to extract features from raw bytes,and exploits self-attention to capture long-range dependencies within packets in a session and contextual information features.Additionally,we utilizing the loss function to more effectively integrate information acquired from both byte sequences and their corresponding lengths.Through simulated evaluations on datasets with similar attributes,ET-Net demonstrates the ability to finely distinguish between nine categories of applications,achieving superior results compared to existing methods.展开更多
BACKGROUND Adenocarcinoma of the esophagogastric junction(AEG)has distinct malignant features compared with other esophageal and gastric cancers.Its management is controversial and largely influenced by tumor location...BACKGROUND Adenocarcinoma of the esophagogastric junction(AEG)has distinct malignant features compared with other esophageal and gastric cancers.Its management is controversial and largely influenced by tumor location and esophageal involve-ment.Hence,understanding the clinicopathological characteristics and prognosis of AEG is essential for optimizing treatment strategies.AIM To evaluate the prognosis and clinicopathological features of patients with AEG,providing insights for management strategies.METHODS This retrospective study analyzed cases with AEG admitted between January 2016 and December 2017.Patients meeting the inclusion criteria were categorized into three groups:Type E[tumors whose center was located within 5 cm above the esophagogastric junction(EGJ)];Type Eg(tumors whose center was situated within 2 cm below the EGJ),with a 2-cm esophageal invasion;Type Ge(tumors whose center was situated within 2 cm below the EGJ),with an esophageal in-vasion of<2 cm,based on tumor location and esophageal involvement.Then,clinicopathological characteristics and survival outcomes of the groups were compared to evaluate the predictive value of the American Joint Committee on Cancer/International Alliance against Cancer 8th edition gastric cancer and eso-phageal adenocarcinoma staging systems.Statistical analysis included survival analysis and Cox regression to assess prognostic factors.RESULTS Totally,153 patients with AEG were included(median follow up:41.1 months;22,31,and 100 patients from type E,Eg,and Ge,respectively),with significant differences in maximum tumor length,esophageal involvement length,tumor type,pathology,differentiation,depth of invasion,and lymph node metastasis between the groups(P<0.05).Lymph node metastasis rates at stations 1,2,3,and 7 were lower in type E than in Eg and Ge(P<0.05).Survival rates for type E(45.5%)were significantly lower than those for Eg(48.4%)and Ge(73.0%)(P=0.001).Type E tumors,vascular infiltration,T3-T4 invasion depth,and lymph node metastasis,were identified as independent prognostic factors(P<0.05).The gastric cancer staging system outperformed the esophageal adenocarcinoma system for type Ge tumors.CONCLUSION Clinicopathological characteristics and prognoses varied between the AEG groups,with type E demonstrating distinct features.The gastric cancer staging system more accurately predicted type Ge AEG prognosis,guiding clinical decision-making.展开更多
In this paper,the sentiment classification method of multimodal adversarial autoencoder is studied.This paper includes the introduction of the multimodal adversarial autoencoder emotion classification method and the e...In this paper,the sentiment classification method of multimodal adversarial autoencoder is studied.This paper includes the introduction of the multimodal adversarial autoencoder emotion classification method and the experiment of the emotion classification method based on the encoder.The experimental analysis shows that the encoder has higher precision than other encoders in emotion classification.It is hoped that this analysis can provide some reference for the emotion classification under the current intelligent algorithm mode.展开更多
This study first analyzes four distinct forms of Fujian folk dance,highlighting the notable differences in their cultural characteristics and dance qualities.It then categorizes these dance forms to align with textboo...This study first analyzes four distinct forms of Fujian folk dance,highlighting the notable differences in their cultural characteristics and dance qualities.It then categorizes these dance forms to align with textbook construction,discussing in depth the principles guiding the development of textbooks that correspond to these classifications.展开更多
The rapid growth of digital data necessitates advanced natural language processing(NLP)models like BERT(Bidi-rectional Encoder Representations from Transformers),known for its superior performance in text classificati...The rapid growth of digital data necessitates advanced natural language processing(NLP)models like BERT(Bidi-rectional Encoder Representations from Transformers),known for its superior performance in text classification.However,BERT’s size and computational demands limit its practicality,especially in resource-constrained settings.This research compresses the BERT base model for Bengali emotion classification through knowledge distillation(KD),pruning,and quantization techniques.Despite Bengali being the sixth most spoken language globally,NLP research in this area is limited.Our approach addresses this gap by creating an efficient BERT-based model for Bengali text.We have explored 20 combinations for KD,quantization,and pruning,resulting in improved speedup,fewer parameters,and reduced memory size.Our best results demonstrate significant improvements in both speed and efficiency.For instance,in the case of mBERT,we achieved a 3.87×speedup and 4×compression ratio with a combination of Distil+Prune+Quant that reduced parameters from 178 to 46 M,while the memory size decreased from 711 to 178 MB.These results offer scalable solutions for NLP tasks in various languages and advance the field of model compression,making these models suitable for real-world applications in resource-limited environments.展开更多
基金the Research Grant of Kwangwoon University in 2024.
文摘Myocardial perfusion imaging(MPI),which uses single-photon emission computed tomography(SPECT),is a well-known estimating tool for medical diagnosis,employing the classification of images to show situations in coronary artery disease(CAD).The automatic classification of SPECT images for different techniques has achieved near-optimal accuracy when using convolutional neural networks(CNNs).This paper uses a SPECT classification framework with three steps:1)Image denoising,2)Attenuation correction,and 3)Image classification.Image denoising is done by a U-Net architecture that ensures effective image denoising.Attenuation correction is implemented by a convolution neural network model that can remove the attenuation that affects the feature extraction process of classification.Finally,a novel multi-scale diluted convolution(MSDC)network is proposed.It merges the features extracted in different scales and makes the model learn the features more efficiently.Three scales of filters with size 3×3 are used to extract features.All three steps are compared with state-of-the-art methods.The proposed denoising architecture ensures a high-quality image with the highest peak signal-to-noise ratio(PSNR)value of 39.7.The proposed classification method is compared with the five different CNN models,and the proposed method ensures better classification with an accuracy of 96%,precision of 87%,sensitivity of 87%,specificity of 89%,and F1-score of 87%.To demonstrate the importance of preprocessing,the classification model was analyzed without denoising and attenuation correction.
文摘In radiology,magnetic resonance imaging(MRI)is an essential diagnostic tool that provides detailed images of a patient’s anatomical and physiological structures.MRI is particularly effective for detecting soft tissue anomalies.Traditionally,radiologists manually interpret these images,which can be labor-intensive and time-consuming due to the vast amount of data.To address this challenge,machine learning,and deep learning approaches can be utilized to improve the accuracy and efficiency of anomaly detection in MRI scans.This manuscript presents the use of the Deep AlexNet50 model for MRI classification with discriminative learning methods.There are three stages for learning;in the first stage,the whole dataset is used to learn the features.In the second stage,some layers of AlexNet50 are frozen with an augmented dataset,and in the third stage,AlexNet50 with an augmented dataset with the augmented dataset.This method used three publicly available MRI classification datasets:Harvard whole brain atlas(HWBA-dataset),the School of Biomedical Engineering of Southern Medical University(SMU-dataset),and The National Institute of Neuroscience and Hospitals brain MRI dataset(NINS-dataset)for analysis.Various hyperparameter optimizers like Adam,stochastic gradient descent(SGD),Root mean square propagation(RMS prop),Adamax,and AdamW have been used to compare the performance of the learning process.HWBA-dataset registers maximum classification performance.We evaluated the performance of the proposed classification model using several quantitative metrics,achieving an average accuracy of 98%.
基金supported in part by NIH grants R01NS39600,U01MH114829RF1MH128693(to GAA)。
文摘Many fields,such as neuroscience,are experiencing the vast prolife ration of cellular data,underscoring the need fo r organizing and interpreting large datasets.A popular approach partitions data into manageable subsets via hierarchical clustering,but objective methods to determine the appropriate classification granularity are missing.We recently introduced a technique to systematically identify when to stop subdividing clusters based on the fundamental principle that cells must differ more between than within clusters.Here we present the corresponding protocol to classify cellular datasets by combining datadriven unsupervised hierarchical clustering with statistical testing.These general-purpose functions are applicable to any cellular dataset that can be organized as two-dimensional matrices of numerical values,including molecula r,physiological,and anatomical datasets.We demonstrate the protocol using cellular data from the Janelia MouseLight project to chara cterize morphological aspects of neurons.
基金the Deanship of Scientifc Research at King Khalid University for funding this work through large group Research Project under grant number RGP2/421/45supported via funding from Prince Sattam bin Abdulaziz University project number(PSAU/2024/R/1446)+1 种基金supported by theResearchers Supporting Project Number(UM-DSR-IG-2023-07)Almaarefa University,Riyadh,Saudi Arabia.supported by the Basic Science Research Program through the National Research Foundation of Korea(NRF)funded by the Ministry of Education(No.2021R1F1A1055408).
文摘Machine learning(ML)is increasingly applied for medical image processing with appropriate learning paradigms.These applications include analyzing images of various organs,such as the brain,lung,eye,etc.,to identify specific flaws/diseases for diagnosis.The primary concern of ML applications is the precise selection of flexible image features for pattern detection and region classification.Most of the extracted image features are irrelevant and lead to an increase in computation time.Therefore,this article uses an analytical learning paradigm to design a Congruent Feature Selection Method to select the most relevant image features.This process trains the learning paradigm using similarity and correlation-based features over different textural intensities and pixel distributions.The similarity between the pixels over the various distribution patterns with high indexes is recommended for disease diagnosis.Later,the correlation based on intensity and distribution is analyzed to improve the feature selection congruency.Therefore,the more congruent pixels are sorted in the descending order of the selection,which identifies better regions than the distribution.Now,the learning paradigm is trained using intensity and region-based similarity to maximize the chances of selection.Therefore,the probability of feature selection,regardless of the textures and medical image patterns,is improved.This process enhances the performance of ML applications for different medical image processing.The proposed method improves the accuracy,precision,and training rate by 13.19%,10.69%,and 11.06%,respectively,compared to other models for the selected dataset.The mean error and selection time is also reduced by 12.56%and 13.56%,respectively,compared to the same models and dataset.
基金Supported by the National Pre-research Program during the 14th Five-Year Plan(514010405)。
文摘In response to the scarcity of infrared aircraft samples and the tendency of traditional deep learning to overfit,a few-shot infrared aircraft classification method based on cross-correlation networks is proposed.This method combines two core modules:a simple parameter-free self-attention and cross-attention.By analyzing the self-correlation and cross-correlation between support images and query images,it achieves effective classification of infrared aircraft under few-shot conditions.The proposed cross-correlation network integrates these two modules and is trained in an end-to-end manner.The simple parameter-free self-attention is responsible for extracting the internal structure of the image while the cross-attention can calculate the cross-correlation between images further extracting and fusing the features between images.Compared with existing few-shot infrared target classification models,this model focuses on the geometric structure and thermal texture information of infrared images by modeling the semantic relevance between the features of the support set and query set,thus better attending to the target objects.Experimental results show that this method outperforms existing infrared aircraft classification methods in various classification tasks,with the highest classification accuracy improvement exceeding 3%.In addition,ablation experiments and comparative experiments also prove the effectiveness of the method.
文摘The World Wide Web provides a wealth of information about everything, including contemporary audio and visual art events, which are discussed on media outlets, blogs, and specialized websites alike. This information may become a robust source of real-world data, which may form the basis of an objective data-driven analysis. In this study, a methodology for collecting information about audio and visual art events in an automated manner from a large array of websites is presented in detail. This process uses cutting edge Semantic Web, Web Search and Generative AI technologies to convert website documents into a collection of structured data. The value of the methodology is demonstrated by creating a large dataset concerning audiovisual events in Greece. The collected information includes event characteristics, estimated metrics based on their text descriptions, outreach metrics based on the media that reported them, and a multi-layered classification of these events based on their type, subjects and methods used. This dataset is openly provided to the general and academic public through a Web application. Moreover, each event’s outreach is evaluated using these quantitative metrics, the results are analyzed with an emphasis on classification popularity and useful conclusions are drawn concerning the importance of artistic subjects, methods, and media.
文摘The rapid growth of machine learning(ML)across fields has intensified the challenge of selecting the right algorithm for specific tasks,known as the Algorithm Selection Problem(ASP).Traditional trial-and-error methods have become impractical due to their resource demands.Automated Machine Learning(AutoML)systems automate this process,but often neglect the group structures and sparsity in meta-features,leading to inefficiencies in algorithm recommendations for classification tasks.This paper proposes a meta-learning approach using Multivariate Sparse Group Lasso(MSGL)to address these limitations.Our method models both within-group and across-group sparsity among meta-features to manage high-dimensional data and reduce multicollinearity across eight meta-feature groups.The Fast Iterative Shrinkage-Thresholding Algorithm(FISTA)with adaptive restart efficiently solves the non-smooth optimization problem.Empirical validation on 145 classification datasets with 17 classification algorithms shows that our meta-learning method outperforms four state-of-the-art approaches,achieving 77.18%classification accuracy,86.07%recommendation accuracy and 88.83%normalized discounted cumulative gain.
基金funded by Princess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2025R435),Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.
文摘Segmenting a breast ultrasound image is still challenging due to the presence of speckle noise,dependency on the operator,and the variation of image quality.This paper presents the UltraSegNet architecture that addresses these challenges through three key technical innovations:This work adds three things:(1)a changed ResNet-50 backbone with sequential 3×3 convolutions to keep fine anatomical details that are needed for finding lesion boundaries;(2)a computationally efficient regional attention mechanism that works on high-resolution features without using a transformer’s extra memory;and(3)an adaptive feature fusion strategy that changes local and global featuresbasedonhowthe image isbeing used.Extensive evaluation on two distinct datasets demonstrates UltraSegNet’s superior performance:On the BUSI dataset,it obtains a precision of 0.915,a recall of 0.908,and an F1 score of 0.911.In the UDAIT dataset,it achieves robust performance across the board,with a precision of 0.901 and recall of 0.894.Importantly,these improvements are achieved at clinically feasible computation times,taking 235 ms per image on standard GPU hardware.Notably,UltraSegNet does amazingly well on difficult small lesions(less than 10 mm),achieving a detection accuracy of 0.891.This is a huge improvement over traditional methods that have a hard time with small-scale features,as standard models can only achieve 0.63–0.71 accuracy.This improvement in small lesion detection is particularly crucial for early-stage breast cancer identification.Results from this work demonstrate that UltraSegNet can be practically deployable in clinical workflows to improve breast cancer screening accuracy.
基金The authors of this study extend their appreciation to the Researchers Supporting Project number(RSPD2025R544),King Saud University,Riyadh,Saudia Arabia.
文摘As the trend to use the latestmachine learning models to automate requirements engineering processes continues,security requirements classification is tuning into the most researched field in the software engineering community.Previous literature studies have proposed numerousmodels for the classification of security requirements.However,adopting those models is constrained due to the lack of essential datasets permitting the repetition and generalization of studies employing more advanced machine learning algorithms.Moreover,most of the researchers focus only on the classification of requirements with security keywords.They did not consider other nonfunctional requirements(NFR)directly or indirectly related to security.This has been identified as a significant research gap in security requirements engineering.The major objective of this study is to propose a security requirements classification model that categorizes security and other relevant security requirements.We use PROMISE_exp and DOSSPRE,the two most commonly used datasets in the software engineering community.The proposed methodology consists of two steps.In the first step,we analyze all the nonfunctional requirements and their relation with security requirements.We found 10 NFRs that have a strong relationship with security requirements.In the second step,we categorize those NFRs in the security requirements category.Our proposedmethodology is a hybridmodel based on the ConvolutionalNeural Network(CNN)and Extreme Gradient Boosting(XGBoost)models.Moreover,we evaluate the model by updating the requirement type column with a binary classification column in the dataset to classify the requirements into security and non-security categories.The performance is evaluated using four metrics:recall,precision,accuracy,and F1 Score with 20 and 28 epochs number and batch size of 32 for PROMISE_exp and DOSSPRE datasets and achieved 87.3%and 85.3%accuracy,respectively.The proposed study shows an enhancement in metrics values compared to the previous literature studies.This is a proof of concept for systematizing the evaluation of security recognition in software systems during the early phases of software development.
基金supported by King Saud University,Riyadh,Saudi Arabia,through the Researchers Supporting Project under Grant RSPD2025R697.
文摘Preservation of the crops depends on early and accurate detection of pests on crops as they cause several diseases decreasing crop production and quality. Several deep-learning techniques have been applied to overcome the issue of pest detection on crops. We have developed the YOLOCSP-PEST model for Pest localization and classification. With the Cross Stage Partial Network (CSPNET) backbone, the proposed model is a modified version of You Only Look Once Version 7 (YOLOv7) that is intended primarily for pest localization and classification. Our proposed model gives exceptionally good results under conditions that are very challenging for any other comparable models especially conditions where we have issues with the luminance and the orientation of the images. It helps farmers working out on their crops in distant areas to determine any infestation quickly and accurately on their crops which helps in the quality and quantity of the production yield. The model has been trained and tested on 2 datasets namely the IP102 data set and a local crop data set on both of which it has shown exceptional results. It gave us a mean average precision (mAP) of 88.40% along with a precision of 85.55% and a recall of 84.25% on the IP102 dataset meanwhile giving a mAP of 97.18% on the local data set along with a recall of 94.88% and a precision of 97.50%. These findings demonstrate that the proposed model is very effective in detecting real-life scenarios and can help in the production of crops improving the yield quality and quantity at the same time.
文摘Flood disasters can have a serious impact on people's production and lives, and can cause hugelosses in lives and property security. Based on multi-source remote sensing data, this study establisheddecision tree classification rules through multi-source and multi-temporal feature fusion, classified groundobjects before the disaster and extracted flood information in the disaster area based on optical imagesduring the disaster, so as to achieve rapid acquisition of the disaster situation of each disaster bearing object.In the case of Qianliang Lake, which suffered from flooding in 2020, the results show that decision treeclassification algorithms based on multi-temporal features can effectively integrate multi-temporal and multispectralinformation to overcome the shortcomings of single-temporal image classification and achieveground-truth object classification.
基金supported by the National Natural Science Foundation of China(62302167,62477013)Natural Science Foundation of Shanghai(No.24ZR1456100)+1 种基金Science and Technology Commission of Shanghai Municipality(No.24DZ2305900)the Shanghai Municipal Special Fund for Promoting High-Quality Development of Industries(2211106).
文摘Multi-label image classification is a challenging task due to the diverse sizes and complex backgrounds of objects in images.Obtaining class-specific precise representations at different scales is a key aspect of feature representation.However,existing methods often rely on the single-scale deep feature,neglecting shallow and deeper layer features,which poses challenges when predicting objects of varying scales within the same image.Although some studies have explored multi-scale features,they rarely address the flow of information between scales or efficiently obtain class-specific precise representations for features at different scales.To address these issues,we propose a two-stage,three-branch Transformer-based framework.The first stage incorporates multi-scale image feature extraction and hierarchical scale attention.This design enables the model to consider objects at various scales while enhancing the flow of information across different feature scales,improving the model’s generalization to diverse object scales.The second stage includes a global feature enhancement module and a region selection module.The global feature enhancement module strengthens interconnections between different image regions,mitigating the issue of incomplete represen-tations,while the region selection module models the cross-modal relationships between image features and labels.Together,these components enable the efficient acquisition of class-specific precise feature representations.Extensive experiments on public datasets,including COCO2014,VOC2007,and VOC2012,demonstrate the effectiveness of our proposed method.Our approach achieves consistent performance gains of 0.3%,0.4%,and 0.2%over state-of-the-art methods on the three datasets,respectively.These results validate the reliability and superiority of our approach for multi-label image classification.
基金supported by the National Key Research and Development Program of China No.2023YFB2705000.
文摘With the rise of encrypted traffic,traditional network analysis methods have become less effective,leading to a shift towards deep learning-based approaches.Among these,multimodal learning-based classification methods have gained attention due to their ability to leverage diverse feature sets from encrypted traffic,improving classification accuracy.However,existing research predominantly relies on late fusion techniques,which hinder the full utilization of deep features within the data.To address this limitation,we propose a novel multimodal encrypted traffic classification model that synchronizes modality fusion with multiscale feature extraction.Specifically,our approach performs real-time fusion of modalities at each stage of feature extraction,enhancing feature representation at each level and preserving inter-level correlations for more effective learning.This continuous fusion strategy improves the model’s ability to detect subtle variations in encrypted traffic,while boosting its robustness and adaptability to evolving network conditions.Experimental results on two real-world encrypted traffic datasets demonstrate that our method achieves a classification accuracy of 98.23% and 97.63%,outperforming existing multimodal learning-based methods.
基金supported by the Institute of Information&Communications Technology Planning&Evaluation(IITP)(Project Nos.RS-2024-00438551,30%,2022-11220701,30%,2021-0-01816,30%)the National Research Foundation of Korea(NRF)grant funded by the Korean Government(Project No.RS2023-00208460,10%).
文摘The proliferation of Internet of Things(IoT)devices has established edge computing as a critical paradigm for real-time data analysis and low-latency processing.Nevertheless,the distributed nature of edge computing presents substantial security challenges,rendering it a prominent target for sophisticated malware attacks.Existing signature-based and behavior-based detection methods are ineffective against the swiftly evolving nature of malware threats and are constrained by the availability of resources.This paper suggests the Genetic Encoding for Novel Optimization of Malware Evaluation(GENOME)framework,a novel solution that is intended to improve the performance of malware detection and classification in peripheral computing environments.GENOME optimizes data storage and computa-tional efficiency by converting malware artifacts into compact,structured sequences through a Deoxyribonucleic Acid(DNA)encoding mechanism.The framework employs two DNA encoding algorithms,standard and compressed,which substantially reduce data size while preserving high detection accuracy.The Edge-IIoTset dataset was used to conduct experiments that showed that GENOME was able to achieve high classification performance using models such as Random Forest and Logistic Regression,resulting in a reduction of data size by up to 42%.Further evaluations with the CIC-IoT-23 dataset and Deep Learning models confirmed GENOME’s scalability and adaptability across diverse datasets and algorithms.The potential of GENOME to address critical challenges,such as the rapid mutation of malware,real-time processing demands,and resource limitations,is emphasized in this study.GENOME offers comprehensive protection for peripheral computing environments by offering a security solution that is both efficient and scalable.
基金sponsored by the National Natural Science Foundation of China Grant No.62271302the Shanghai Municipal Natural Science Foundation Grant 20ZR1423500.
文摘Large amounts of labeled data are usually needed for training deep neural networks in medical image studies,particularly in medical image classification.However,in the field of semi-supervised medical image analysis,labeled data is very scarce due to patient privacy concerns.For researchers,obtaining high-quality labeled images is exceedingly challenging because it involves manual annotation and clinical understanding.In addition,skin datasets are highly suitable for medical image classification studies due to the inter-class relationships and the inter-class similarities of skin lesions.In this paper,we propose a model called Coalition Sample Relation Consistency(CSRC),a consistency-based method that leverages Canonical Correlation Analysis(CCA)to capture the intrinsic relationships between samples.Considering that traditional consistency-based models only focus on the consistency of prediction,we additionally explore the similarity between features by using CCA.We enforce feature relation consistency based on traditional models,encouraging the model to learn more meaningful information from unlabeled data.Finally,considering that cross-entropy loss is not as suitable as the supervised loss when studying with imbalanced datasets(i.e.,ISIC 2017 and ISIC 2018),we improve the supervised loss to achieve better classification accuracy.Our study shows that this model performs better than many semi-supervised methods.
基金supported by National Key Research and Development Program of China(2022YFB3104903)S&T Program of Hebei(No.SZX2020034).
文摘Intelligent vehicle applications provide convenience but raise privacy and security concerns.Misuse of sensitive data,including vehicle location,and facial recognition information,poses a threat to user privacy.Hence,traffic classification is vital for promptly overseeing and controlling applications with sensitive information.In this paper,we propose ETNet,a framework that combines multiple features and leverages self-attention mechanisms to learn deep relationships between packets.ET-Net employs a multisimilarity triplet network to extract features from raw bytes,and exploits self-attention to capture long-range dependencies within packets in a session and contextual information features.Additionally,we utilizing the loss function to more effectively integrate information acquired from both byte sequences and their corresponding lengths.Through simulated evaluations on datasets with similar attributes,ET-Net demonstrates the ability to finely distinguish between nine categories of applications,achieving superior results compared to existing methods.
基金Supported by the Medical Science Research Project of Hebei,No.20211323.
文摘BACKGROUND Adenocarcinoma of the esophagogastric junction(AEG)has distinct malignant features compared with other esophageal and gastric cancers.Its management is controversial and largely influenced by tumor location and esophageal involve-ment.Hence,understanding the clinicopathological characteristics and prognosis of AEG is essential for optimizing treatment strategies.AIM To evaluate the prognosis and clinicopathological features of patients with AEG,providing insights for management strategies.METHODS This retrospective study analyzed cases with AEG admitted between January 2016 and December 2017.Patients meeting the inclusion criteria were categorized into three groups:Type E[tumors whose center was located within 5 cm above the esophagogastric junction(EGJ)];Type Eg(tumors whose center was situated within 2 cm below the EGJ),with a 2-cm esophageal invasion;Type Ge(tumors whose center was situated within 2 cm below the EGJ),with an esophageal in-vasion of<2 cm,based on tumor location and esophageal involvement.Then,clinicopathological characteristics and survival outcomes of the groups were compared to evaluate the predictive value of the American Joint Committee on Cancer/International Alliance against Cancer 8th edition gastric cancer and eso-phageal adenocarcinoma staging systems.Statistical analysis included survival analysis and Cox regression to assess prognostic factors.RESULTS Totally,153 patients with AEG were included(median follow up:41.1 months;22,31,and 100 patients from type E,Eg,and Ge,respectively),with significant differences in maximum tumor length,esophageal involvement length,tumor type,pathology,differentiation,depth of invasion,and lymph node metastasis between the groups(P<0.05).Lymph node metastasis rates at stations 1,2,3,and 7 were lower in type E than in Eg and Ge(P<0.05).Survival rates for type E(45.5%)were significantly lower than those for Eg(48.4%)and Ge(73.0%)(P=0.001).Type E tumors,vascular infiltration,T3-T4 invasion depth,and lymph node metastasis,were identified as independent prognostic factors(P<0.05).The gastric cancer staging system outperformed the esophageal adenocarcinoma system for type Ge tumors.CONCLUSION Clinicopathological characteristics and prognoses varied between the AEG groups,with type E demonstrating distinct features.The gastric cancer staging system more accurately predicted type Ge AEG prognosis,guiding clinical decision-making.
文摘In this paper,the sentiment classification method of multimodal adversarial autoencoder is studied.This paper includes the introduction of the multimodal adversarial autoencoder emotion classification method and the experiment of the emotion classification method based on the encoder.The experimental analysis shows that the encoder has higher precision than other encoders in emotion classification.It is hoped that this analysis can provide some reference for the emotion classification under the current intelligent algorithm mode.
文摘This study first analyzes four distinct forms of Fujian folk dance,highlighting the notable differences in their cultural characteristics and dance qualities.It then categorizes these dance forms to align with textbook construction,discussing in depth the principles guiding the development of textbooks that correspond to these classifications.
文摘The rapid growth of digital data necessitates advanced natural language processing(NLP)models like BERT(Bidi-rectional Encoder Representations from Transformers),known for its superior performance in text classification.However,BERT’s size and computational demands limit its practicality,especially in resource-constrained settings.This research compresses the BERT base model for Bengali emotion classification through knowledge distillation(KD),pruning,and quantization techniques.Despite Bengali being the sixth most spoken language globally,NLP research in this area is limited.Our approach addresses this gap by creating an efficient BERT-based model for Bengali text.We have explored 20 combinations for KD,quantization,and pruning,resulting in improved speedup,fewer parameters,and reduced memory size.Our best results demonstrate significant improvements in both speed and efficiency.For instance,in the case of mBERT,we achieved a 3.87×speedup and 4×compression ratio with a combination of Distil+Prune+Quant that reduced parameters from 178 to 46 M,while the memory size decreased from 711 to 178 MB.These results offer scalable solutions for NLP tasks in various languages and advance the field of model compression,making these models suitable for real-world applications in resource-limited environments.
