In this paper,an Automated Brain Image Analysis(ABIA)system that classifies the Magnetic Resonance Imaging(MRI)of human brain is presented.The classification of MRI images into normal or low grade or high grade plays ...In this paper,an Automated Brain Image Analysis(ABIA)system that classifies the Magnetic Resonance Imaging(MRI)of human brain is presented.The classification of MRI images into normal or low grade or high grade plays a vital role for the early diagnosis.The Non-Subsampled Shearlet Transform(NSST)that captures more visual information than conventional wavelet transforms is employed for feature extraction.As the feature space of NSST is very high,a statistical t-test is applied to select the dominant directional sub-bands at each level of NSST decomposition based on sub-band energies.A combination of features that includes Gray Level Co-occurrence Matrix(GLCM)based features,Histograms of Positive Shearlet Coefficients(HPSC),and Histograms of Negative Shearlet Coefficients(HNSC)are estimated.The combined feature set is utilized in the classification phase where a hybrid approach is designed with three classifiers;k-Nearest Neighbor(kNN),Naive Bayes(NB)and Support Vector Machine(SVM)classifiers.The output of individual trained classifiers for a testing input is hybridized to take a final decision.The quantitative results of ABIA system on Repository of Molecular Brain Neoplasia Data(REMBRANDT)database show the overall improved performance in comparison with a single classifier model with accuracy of 99% for normal/abnormal classification and 98% for low and high risk classification.展开更多
Medical image analysis is an active research topic,with thousands of studies published in the past few years.Transfer learning(TL)including convolutional neural networks(CNNs)focused to enhance efficiency on an innova...Medical image analysis is an active research topic,with thousands of studies published in the past few years.Transfer learning(TL)including convolutional neural networks(CNNs)focused to enhance efficiency on an innovative task using the knowledge of the same tasks learnt in advance.It has played a major role in medical image analysis since it solves the data scarcity issue along with that it saves hardware resources and time.This study develops an EnhancedTunicate SwarmOptimization withTransfer Learning EnabledMedical Image Analysis System(ETSOTL-MIAS).The goal of the ETSOTL-MIAS technique lies in the identification and classification of diseases through medical imaging.The ETSOTL-MIAS technique involves the Chan Vese segmentation technique to identify the affected regions in the medical image.For feature extraction purposes,the ETSOTL-MIAS technique designs a modified DarkNet-53 model.To avoid the manual hyperparameter adjustment process,the ETSOTLMIAS technique exploits the ETSO algorithm,showing the novelty of the work.Finally,the classification of medical images takes place by random forest(RF)classifier.The performance validation of the ETSOTL-MIAS technique is tested on a benchmark medical image database.The extensive experimental analysis showed the promising performance of the ETSOTL-MIAS technique under different measures.展开更多
The continuous growth in the scale of unmanned aerial vehicle (UAV) applications in transmission line inspection has resulted in a corresponding increase in the demand for UAV inspection image processing. Owing to its...The continuous growth in the scale of unmanned aerial vehicle (UAV) applications in transmission line inspection has resulted in a corresponding increase in the demand for UAV inspection image processing. Owing to its excellent performance in computer vision, deep learning has been applied to UAV inspection image processing tasks such as power line identification and insulator defect detection. Despite their excellent performance, electric power UAV inspection image processing models based on deep learning face several problems such as a small application scope, the need for constant retraining and optimization, and high R&D monetary and time costs due to the black-box and scene data-driven characteristics of deep learning. In this study, an automated deep learning system for electric power UAV inspection image analysis and processing is proposed as a solution to the aforementioned problems. This system design is based on the three critical design principles of generalizability, extensibility, and automation. Pre-trained models, fine-tuning (downstream task adaptation), and automated machine learning, which are closely related to these design principles, are reviewed. In addition, an automated deep learning system architecture for electric power UAV inspection image analysis and processing is presented. A prototype system was constructed and experiments were conducted on the two electric power UAV inspection image analysis and processing tasks of insulator self-detonation and bird nest recognition. The models constructed using the prototype system achieved 91.36% and 86.13% mAP for insulator self-detonation and bird nest recognition, respectively. This demonstrates that the system design concept is reasonable and the system architecture feasible .展开更多
Malaria is an important and worldwide fatal disease that has been widely reported by the World Health Organization(WHO),and it has about 219 million cases worldwide,with 435,000 of those mortal.The common malaria diag...Malaria is an important and worldwide fatal disease that has been widely reported by the World Health Organization(WHO),and it has about 219 million cases worldwide,with 435,000 of those mortal.The common malaria diagnosis approach is heavily reliant on highly trained experts,who use a microscope to examine the samples.Therefore,there is a need to create an automated solution for the diagnosis of malaria.One of the main objectives of this work is to create a design tool that could be used to diagnose malaria from the image of a blood sample.In this paper,we firstly developed a graphical user interface that could be used to help segment red blood cells and infected cells and allow the users to analyze the blood samples.Secondly,a Feed-forward Neural Network(FNN)is designed to classify the cells into two classes.The achieved results show that the proposed techniques can be used to detect malaria,as it has achieved 92%accuracy with a database that contains 27,560 benchmark images.展开更多
BACKGROUND Congestive hepatopathy,also known as nutmeg liver,is liver damage secondary to chronic heart failure(HF).Its morphological characteristics in terms of medical imaging are not defined and remain unclear.AIM ...BACKGROUND Congestive hepatopathy,also known as nutmeg liver,is liver damage secondary to chronic heart failure(HF).Its morphological characteristics in terms of medical imaging are not defined and remain unclear.AIM To leverage machine learning to capture imaging features of congestive hepatopathy using incidentally acquired computed tomography(CT)scans.METHODS We retrospectively analyzed 179 chronic HF patients who underwent echocardiography and CT within one year.Right HF severity was classified into three grades.Liver CT images at the paraumbilical vein level were used to develop a ResNet-based machine learning model to predict tricuspid regurgitation(TR)severity.Model accuracy was compared with that of six gastroenterology and four radiology experts.RESULTS In the included patients,120 were male(mean age:73.1±14.4 years).The accuracy of the results predicting TR severity from a single CT image for the machine learning model was significantly higher than the average accuracy of the experts.The model was found to be exceptionally reliable for predicting severe TR.CONCLUSION Deep learning models,particularly those using ResNet architectures,can help identify morphological changes associated with TR severity,aiding in early liver dysfunction detection in patients with HF,thereby improving outcomes.展开更多
Medical image analysis has become a cornerstone of modern healthcare,driven by the exponential growth of data from imaging modalities such as MRI,CT,PET,ultrasound,and X-ray.Traditional machine learning methods have m...Medical image analysis has become a cornerstone of modern healthcare,driven by the exponential growth of data from imaging modalities such as MRI,CT,PET,ultrasound,and X-ray.Traditional machine learning methods have made early contributions;however,recent advancements in deep learning(DL)have revolutionized the field,offering state-of-the-art performance in image classification,segmentation,detection,fusion,registration,and enhancement.This comprehensive review presents an in-depth analysis of deep learning methodologies applied across medical image analysis tasks,highlighting both foundational models and recent innovations.The article begins by introducing conventional techniques and their limitations,setting the stage for DL-based solutions.Core DL architectures,including Convolutional Neural Networks(CNNs),Recurrent Neural Networks(RNNs),Generative Adversarial Networks(GANs),Vision Transformers(ViTs),and hybrid models,are discussed in detail,including their advantages and domain-specific adaptations.Advanced learning paradigms such as semi-supervised learning,selfsupervised learning,and few-shot learning are explored for their potential to mitigate data annotation challenges in clinical datasets.This review further categorizes major tasks in medical image analysis,elaborating on how DL techniques have enabled precise tumor segmentation,lesion detection,modality fusion,super-resolution,and robust classification across diverse clinical settings.Emphasis is placed on applications in oncology,cardiology,neurology,and infectious diseases,including COVID-19.Challenges such as data scarcity,label imbalance,model generalizability,interpretability,and integration into clinical workflows are critically examined.Ethical considerations,explainable AI(XAI),federated learning,and regulatory compliance are discussed as essential components of real-world deployment.Benchmark datasets,evaluation metrics,and comparative performance analyses are presented to support future research.The article concludes with a forward-looking perspective on the role of foundation models,multimodal learning,edge AI,and bio-inspired computing in the future of medical imaging.Overall,this review serves as a valuable resource for researchers,clinicians,and developers aiming to harness deep learning for intelligent,efficient,and clinically viable medical image analysis.展开更多
Optimizing root system architecture(RSA)is essential for plants because of its critical role in acquiring water and nutrients from the soil.However,the subterranean nature of roots complicates the measurement of RSA t...Optimizing root system architecture(RSA)is essential for plants because of its critical role in acquiring water and nutrients from the soil.However,the subterranean nature of roots complicates the measurement of RSA traits.Recently developed rhizobox methods allow for the rapid acquisition of root images.Nevertheless,effective and precise approaches for extracting RSA features from these images remain underdeveloped.Deep learning(DL)technology can enhance image segmentation and facilitate RSA trait extraction.However,comprehensive pipelines that integrate DL technologies into image-based root phenotyping techniques are still scarce,hampering their implementation.To address this challenge,we present a reproducible pipeline(faCRSA)for automated RSA traits analysis,consisting of three modules:(1)the RSA traits extraction module functions to segment soil-root images and calculate RSA traits.A lightweight convolutional neural network(CNN)named RootSeg was proposed for efficient and accurate segmentation;(2)the data storage module,which stores image and text data from other modules;and(3)the web application module,which allows researchers to analyze data online in a user-friendly manner.The correlation coefficients(R^(2))of total root length,root surface area,and root volume calculated from faCRSA and manually measured results were 0.96**,0.97**,and 0.93**,respectively,with root mean square errors(RMSE)of 8.13 cm,1.68 cm^(2),and 0.05 cm^(3),processed at a rate of 9.74 s per image,indicating satisfying accuracy.faCRSA has also demonstrated satisfactory performance in dynamically monitoring root system changes under various stress conditions,such as drought or waterlogging.The detailed code and deployable package of faCRSA are provided for researchers with the potential to replace manual and semi-automated methods.展开更多
Integrating multiple medical imaging techniques,including Magnetic Resonance Imaging(MRI),Computed Tomography,Positron Emission Tomography(PET),and ultrasound,provides a comprehensive view of the patient health status...Integrating multiple medical imaging techniques,including Magnetic Resonance Imaging(MRI),Computed Tomography,Positron Emission Tomography(PET),and ultrasound,provides a comprehensive view of the patient health status.Each of these methods contributes unique diagnostic insights,enhancing the overall assessment of patient condition.Nevertheless,the amalgamation of data from multiple modalities presents difficulties due to disparities in resolution,data collection methods,and noise levels.While traditional models like Convolutional Neural Networks(CNNs)excel in single-modality tasks,they struggle to handle multi-modal complexities,lacking the capacity to model global relationships.This research presents a novel approach for examining multi-modal medical imagery using a transformer-based system.The framework employs self-attention and cross-attention mechanisms to synchronize and integrate features across various modalities.Additionally,it shows resilience to variations in noise and image quality,making it adaptable for real-time clinical use.To address the computational hurdles linked to transformer models,particularly in real-time clinical applications in resource-constrained environments,several optimization techniques have been integrated to boost scalability and efficiency.Initially,a streamlined transformer architecture was adopted to minimize the computational load while maintaining model effectiveness.Methods such as model pruning,quantization,and knowledge distillation have been applied to reduce the parameter count and enhance the inference speed.Furthermore,efficient attention mechanisms such as linear or sparse attention were employed to alleviate the substantial memory and processing requirements of traditional self-attention operations.For further deployment optimization,researchers have implemented hardware-aware acceleration strategies,including the use of TensorRT and ONNX-based model compression,to ensure efficient execution on edge devices.These optimizations allow the approach to function effectively in real-time clinical settings,ensuring viability even in environments with limited resources.Future research directions include integrating non-imaging data to facilitate personalized treatment and enhancing computational efficiency for implementation in resource-limited environments.This study highlights the transformative potential of transformer models in multi-modal medical imaging,offering improvements in diagnostic accuracy and patient care outcomes.展开更多
Deep learning models have become a core technological tool in the field of medical image analysis.However,these models often suffer from a lack of transparency in their decision-making processes,leading to challenges ...Deep learning models have become a core technological tool in the field of medical image analysis.However,these models often suffer from a lack of transparency in their decision-making processes,leading to challenges related to trust and interpret ability in clinical applications.To address this issue,explainable artificial intelligence(XAI)techniques have been applied to medical image analysis.While showing promising potential,XAI also brings significant ethical risks in practice—most notably,the problem of spurious explanations.Such explanations may rise further concerns regarding patient privacy,data security,and the attribution of decisionmaking authority in medical contexts.This paper analyzes the application of XAI methods—particularly saliency aps—in medical image interpretation,identifies the underlying causes of spurious explanations,and proposes possible mitigation strategies.The aim is to contribute to the responsible and sustainable integration of explainable AI into clinical practice.展开更多
Given the importance of sentiment analysis in diverse environments,various methods are used for image sentiment analysis,including contextual sentiment analysis that utilizes character and scene relationships.However,...Given the importance of sentiment analysis in diverse environments,various methods are used for image sentiment analysis,including contextual sentiment analysis that utilizes character and scene relationships.However,most existing works employ character faces in conjunction with context,yet lack the capacity to analyze the emotions of characters in unconstrained environments,such as when their faces are obscured or blurred.Accordingly,this article presents the Adaptive Multi-Channel Sentiment Analysis Network(AMSA),a contextual image sentiment analysis framework,which consists of three channels:body,face,and context.AMSA employs Multi-task Cascaded Convolutional Networks(MTCNN)to detect faces within body frames;if detected,facial features are extracted and fused with body and context information for emotion recognition.If not,the model leverages body and context features alone.Meanwhile,to address class imbalance in the EMOTIC dataset,Focal Loss is introduced to improve classification performance,especially for minority emotion categories.Experimental results have shown that certain sentiment categories with lower representation in the dataset demonstrate leading classification accuracy,the AMSA yields a 2.53%increase compared with state-of-the-art methods.展开更多
The detection and characterization of non-metallic inclusions are essential for clean steel production.Recently,imaging analysis combined with high-dimensional data processing of metallic materials using artificial in...The detection and characterization of non-metallic inclusions are essential for clean steel production.Recently,imaging analysis combined with high-dimensional data processing of metallic materials using artificial intelligence(AI)-based machine learning(ML)has developed rapidly.This technique has achieved impressive results in the field of inclusion classification in process metallurgy.The present study surveys the ML modeling of inclusion prediction in advanced steels,including the detection,classification,and feature prediction of inclusions in different steel grades.Studies on clean steel with different features based on data and image analysis via ML are summarized.Regarding the data analysis,the inclusion prediction methodology based on ML establishes a connection between the experimental parameters and inclusion characteristics and analyzes the importance of the experimental parameters.Regarding the image analysis,the focus is placed on the classification of different types of inclusions via deep learning,in comparison with data analysis.Finally,further development of inclusion analyses using ML-based methods is recommended.This work paves the way for the application of AIbased methodologies for ultraclean-steel studies from a sustainable metallurgy perspective.展开更多
Recent studies indicate that millions of individuals suffer from renal diseases,with renal carcinoma,a type of kidney cancer,emerging as both a chronic illness and a significant cause of mortality.Magnetic Resonance I...Recent studies indicate that millions of individuals suffer from renal diseases,with renal carcinoma,a type of kidney cancer,emerging as both a chronic illness and a significant cause of mortality.Magnetic Resonance Imaging(MRI)and Computed Tomography(CT)have become essential tools for diagnosing and assessing kidney disorders.However,accurate analysis of thesemedical images is critical for detecting and evaluating tumor severity.This study introduces an integrated hybrid framework that combines three complementary deep learning models for kidney tumor segmentation from MRI images.The proposed framework fuses a customized U-Net and Mask R-CNN using a weighted scheme to achieve semantic and instance-level segmentation.The fused outputs are further refined through edge detection using Stochastic FeatureMapping Neural Networks(SFMNN),while volumetric consistency is ensured through Improved Mini-Batch K-Means(IMBKM)clustering integrated with an Encoder-Decoder Convolutional Neural Network(EDCNN).The outputs of these three stages are combined through a weighted fusion mechanism,with optimal weights determined empirically.Experiments on MRI scans from the TCGA-KIRC dataset demonstrate that the proposed hybrid framework significantly outperforms standalone models,achieving a Dice Score of 92.5%,an IoU of 87.8%,a Precision of 93.1%,a Recall of 90.8%,and a Hausdorff Distance of 2.8 mm.These findings validate that the weighted integration of complementary architectures effectively overcomes key limitations in kidney tumor segmentation,leading to improved diagnostic accuracy and robustness in medical image analysis.展开更多
Underground hydrogen storage has gained interest in recent years due to the enormous demand for clean energy.Hydrogen is more diffusive than air,with a smaller density and lower viscosity.These unique properties intro...Underground hydrogen storage has gained interest in recent years due to the enormous demand for clean energy.Hydrogen is more diffusive than air,with a smaller density and lower viscosity.These unique properties introduce distinctive hydrodynamic phenomena in hydrogen storage,one of which is fingering.Fingering could induce the fluid trapped in small clusters of pores,leading to a dramatic decrease in hydrogen saturation and a lower recovery rate.In this study,numerical simulations are performed at the microscopic scale to understand the evolution of hydrogen saturation and the impacts of injection and withdrawal cycles.Two sets of micromodels with different porosity(0.362 and 0.426)and minimum sizes of pore throats(0.362 mm and 0.181 mm)are developed in the numerical model.A parameter analysis is then conducted to understand the influence of injection velocity(in the range of 10^(-2)m/s to 10^(-5)m/s)and porous structure on the fingering pattern,followed by an image analysis to capture the evolution of the fingering pattern.Viscous fingering,capillary fingering,and crossover fingering are observed and identified under different boundary conditions.The fractal dimension,specific area,mean angle,and entropy of fingers are proposed as geometric descriptors to characterize the shape of the fingering pattern.When porosity increases from 0.362 to 0.426,the saturation of hydrogen increases by 26.2%.Narrower pore throats elevate capillary resistance,which hinders fluid invasion.These results underscore the importance of pore structures and the interaction between viscous and capillary forces for hydrogen recovery efficiency.This work illuminates the influence of the pore structures and the fluid properties on the immiscible displacement of hydrogen and can be further extended to optimize the injection strategy of hydrogen in underground hydrogen storage.展开更多
Methods and procedures of three-dimensional （3D） characterization of the pore structure features in the packed ore particle bed are focused. X-ray computed tomography was applied to deriving the cross-sectional imag...Methods and procedures of three-dimensional （3D） characterization of the pore structure features in the packed ore particle bed are focused. X-ray computed tomography was applied to deriving the cross-sectional images of specimens with single particle size of 1-2, 2-3, 3-4, 4-5, 5-6, 6-7, 7-8, 8-9, 9-10 ram. Based on the in-house developed 3D image analysis programs using Matlab, the volume porosity, pore size distribution and degree of connectivity were calculated and analyzed in detail. The results indicate that the volume porosity, the mean diameter of pores and the effective pore size （d50） increase with the increasing of particle size. Lognormal distribution or Gauss distribution is mostly suitable to model the pore size distribution. The degree of connectivity investigated on the basis of cluster-labeling algorithm also increases with increasing the particle size approximately.展开更多
The changes of retinal nuclear DNA content in rats after death was detected and the relationship between degradation of retinal nuclear DNA and postmortem interval （PMI） was analyzed. Ninety healthy adult SD rats, f...The changes of retinal nuclear DNA content in rats after death was detected and the relationship between degradation of retinal nuclear DNA and postmortem interval （PMI） was analyzed. Ninety healthy adult SD rats, female, weighing 250±10 g, were randomly divided into 15 groups. At 20 ℃, the retinal cells were withdrawn every 2 h within 0 to 28 h after death and stained with Feulgen-Vans. Index of density （ID）, integral absorbance （IA） and average absorbance （AA） in retinal nucleus were analyzed by image analysis system. And the obtained data were subjected to linear regression analysis by using SPSS12.0 software. The results showed that in retinal nucleus, AA and IA were gradually declined with the prolongation of PMI, while ID had an increased tendency. Within 28 h after PMI, the regression equations were as follows： YAA=-0.009XAA＋0.590 （R^2=0.949）, YIA=0.097XIA＋18.903 （R^2=0.968）, YID=0.122XID＋2.246 （R^2=0.951）. It was concluded that retinal nuclear DNA after death in rats was degraded gradually and had a good correlation with PMI.展开更多
Mineral dissemination and pore space distribution in ore particles are important features that influence heap leaching performance.To quantify the mineral dissemination and pore space distribution of an ore particle,a...Mineral dissemination and pore space distribution in ore particles are important features that influence heap leaching performance.To quantify the mineral dissemination and pore space distribution of an ore particle,a cylindrical copper oxide ore sample(I center dot 4.6 mm x 5.6 mm)was scanned using high-resolution X-ray computed tomography(HRXCT),a nondestructive imaging technology,at a spatial resolution of 4.85 mu m.Combined with three-dimensional(3D)image analysis techniques,the main mineral phases and pore space were segmented and the volume fraction of each phase was calculated.In addition,the mass fraction of each mineral phase was estimated and the result was validated with that obtained using traditional techniques.Furthermore,the pore phase features,including the pore size distribution,pore surface area,pore fractal dimension,pore centerline,and the pore connectivity,were investigated quantitatively.The pore space analysis results indicate that the pore size distribution closely fits a log-normal distribution and that the pore space morphology is complicated,with a large surface area and low connectivity.This study demonstrates that the combination of HRXCT and 3D image analysis is an effective tool for acquiring 3D mineralogical and pore structural data.展开更多
With the wide use of high-resolution remotely sensed imagery, the object-oriented remotely sensed informa- tion classification pattern has been intensively studied. Starting with the definition of object-oriented remo...With the wide use of high-resolution remotely sensed imagery, the object-oriented remotely sensed informa- tion classification pattern has been intensively studied. Starting with the definition of object-oriented remotely sensed information classification pattern and a literature review of related research progress, this paper sums up 4 developing phases of object-oriented classification pattern during the past 20 years. Then, we discuss the three aspects of method- ology in detail, namely remotely sensed imagery segmentation, feature analysis and feature selection, and classification rule generation, through comparing them with remotely sensed information classification method based on per-pixel. At last, this paper presents several points that need to be paid attention to in the future studies on object-oriented RS in- formation classification pattern: 1) developing robust and highly effective image segmentation algorithm for multi-spectral RS imagery; 2) improving the feature-set including edge, spatial-adjacent and temporal characteristics; 3) discussing the classification rule generation classifier based on the decision tree; 4) presenting evaluation methods for classification result by object-oriented classification pattern.展开更多
Based on the analysis of high-speed video images, the detachment behavior of dust cake from the ceramic candle filter surface during pulse cleaning process is investigated. The influences of the dust cake loading,the ...Based on the analysis of high-speed video images, the detachment behavior of dust cake from the ceramic candle filter surface during pulse cleaning process is investigated. The influences of the dust cake loading,the reservoir pressure, and the filtration velocity on the cleaning effectiveness are analyzed. Experimental results show that there exists an optimum dust cake thickness for pulse-cleaning process. For thin dust cake, the patchy cleaning exists and the cleaning efficiency is low; if the dust cake is too thick, the pressure drop across the dust cake becomes higher and a higher reservoir pressure may be needed. At the same time there also exists an optimum reservoir pressure for a given filtration condition.展开更多
文摘In this paper,an Automated Brain Image Analysis(ABIA)system that classifies the Magnetic Resonance Imaging(MRI)of human brain is presented.The classification of MRI images into normal or low grade or high grade plays a vital role for the early diagnosis.The Non-Subsampled Shearlet Transform(NSST)that captures more visual information than conventional wavelet transforms is employed for feature extraction.As the feature space of NSST is very high,a statistical t-test is applied to select the dominant directional sub-bands at each level of NSST decomposition based on sub-band energies.A combination of features that includes Gray Level Co-occurrence Matrix(GLCM)based features,Histograms of Positive Shearlet Coefficients(HPSC),and Histograms of Negative Shearlet Coefficients(HNSC)are estimated.The combined feature set is utilized in the classification phase where a hybrid approach is designed with three classifiers;k-Nearest Neighbor(kNN),Naive Bayes(NB)and Support Vector Machine(SVM)classifiers.The output of individual trained classifiers for a testing input is hybridized to take a final decision.The quantitative results of ABIA system on Repository of Molecular Brain Neoplasia Data(REMBRANDT)database show the overall improved performance in comparison with a single classifier model with accuracy of 99% for normal/abnormal classification and 98% for low and high risk classification.
基金support for this work from the Deanship of Scientific Research (DSR),University of Tabuk,Tabuk,Saudi Arabia,under grant number S-1440-0262.
文摘Medical image analysis is an active research topic,with thousands of studies published in the past few years.Transfer learning(TL)including convolutional neural networks(CNNs)focused to enhance efficiency on an innovative task using the knowledge of the same tasks learnt in advance.It has played a major role in medical image analysis since it solves the data scarcity issue along with that it saves hardware resources and time.This study develops an EnhancedTunicate SwarmOptimization withTransfer Learning EnabledMedical Image Analysis System(ETSOTL-MIAS).The goal of the ETSOTL-MIAS technique lies in the identification and classification of diseases through medical imaging.The ETSOTL-MIAS technique involves the Chan Vese segmentation technique to identify the affected regions in the medical image.For feature extraction purposes,the ETSOTL-MIAS technique designs a modified DarkNet-53 model.To avoid the manual hyperparameter adjustment process,the ETSOTLMIAS technique exploits the ETSO algorithm,showing the novelty of the work.Finally,the classification of medical images takes place by random forest(RF)classifier.The performance validation of the ETSOTL-MIAS technique is tested on a benchmark medical image database.The extensive experimental analysis showed the promising performance of the ETSOTL-MIAS technique under different measures.
基金This work was supported by Science and Technology Project of State Grid Corporation“Research on Key Technologies of Power Artificial Intelligence Open Platform”(5700-202155260A-0-0-00).
文摘The continuous growth in the scale of unmanned aerial vehicle (UAV) applications in transmission line inspection has resulted in a corresponding increase in the demand for UAV inspection image processing. Owing to its excellent performance in computer vision, deep learning has been applied to UAV inspection image processing tasks such as power line identification and insulator defect detection. Despite their excellent performance, electric power UAV inspection image processing models based on deep learning face several problems such as a small application scope, the need for constant retraining and optimization, and high R&D monetary and time costs due to the black-box and scene data-driven characteristics of deep learning. In this study, an automated deep learning system for electric power UAV inspection image analysis and processing is proposed as a solution to the aforementioned problems. This system design is based on the three critical design principles of generalizability, extensibility, and automation. Pre-trained models, fine-tuning (downstream task adaptation), and automated machine learning, which are closely related to these design principles, are reviewed. In addition, an automated deep learning system architecture for electric power UAV inspection image analysis and processing is presented. A prototype system was constructed and experiments were conducted on the two electric power UAV inspection image analysis and processing tasks of insulator self-detonation and bird nest recognition. The models constructed using the prototype system achieved 91.36% and 86.13% mAP for insulator self-detonation and bird nest recognition, respectively. This demonstrates that the system design concept is reasonable and the system architecture feasible .
基金This work is partly supported by the Fundamental Research Funds for the Central Universities of China under grants GK202003080the Natural Science Foundation of Shaanxi Province under Grants 2021JM-205the UK Engineering and Physical Sciences Research Council through grants EP/V034111/1.
文摘Malaria is an important and worldwide fatal disease that has been widely reported by the World Health Organization(WHO),and it has about 219 million cases worldwide,with 435,000 of those mortal.The common malaria diagnosis approach is heavily reliant on highly trained experts,who use a microscope to examine the samples.Therefore,there is a need to create an automated solution for the diagnosis of malaria.One of the main objectives of this work is to create a design tool that could be used to diagnose malaria from the image of a blood sample.In this paper,we firstly developed a graphical user interface that could be used to help segment red blood cells and infected cells and allow the users to analyze the blood samples.Secondly,a Feed-forward Neural Network(FNN)is designed to classify the cells into two classes.The achieved results show that the proposed techniques can be used to detect malaria,as it has achieved 92%accuracy with a database that contains 27,560 benchmark images.
基金Supported by Grant-in-Aid for Research on Hepatitis from the Japan Agency for Medical Research and Development,No.24fk0210128h0002Grant-in-Aid for Scientific Research,No.KAKENHI-23K07372.
文摘BACKGROUND Congestive hepatopathy,also known as nutmeg liver,is liver damage secondary to chronic heart failure(HF).Its morphological characteristics in terms of medical imaging are not defined and remain unclear.AIM To leverage machine learning to capture imaging features of congestive hepatopathy using incidentally acquired computed tomography(CT)scans.METHODS We retrospectively analyzed 179 chronic HF patients who underwent echocardiography and CT within one year.Right HF severity was classified into three grades.Liver CT images at the paraumbilical vein level were used to develop a ResNet-based machine learning model to predict tricuspid regurgitation(TR)severity.Model accuracy was compared with that of six gastroenterology and four radiology experts.RESULTS In the included patients,120 were male(mean age:73.1±14.4 years).The accuracy of the results predicting TR severity from a single CT image for the machine learning model was significantly higher than the average accuracy of the experts.The model was found to be exceptionally reliable for predicting severe TR.CONCLUSION Deep learning models,particularly those using ResNet architectures,can help identify morphological changes associated with TR severity,aiding in early liver dysfunction detection in patients with HF,thereby improving outcomes.
文摘Medical image analysis has become a cornerstone of modern healthcare,driven by the exponential growth of data from imaging modalities such as MRI,CT,PET,ultrasound,and X-ray.Traditional machine learning methods have made early contributions;however,recent advancements in deep learning(DL)have revolutionized the field,offering state-of-the-art performance in image classification,segmentation,detection,fusion,registration,and enhancement.This comprehensive review presents an in-depth analysis of deep learning methodologies applied across medical image analysis tasks,highlighting both foundational models and recent innovations.The article begins by introducing conventional techniques and their limitations,setting the stage for DL-based solutions.Core DL architectures,including Convolutional Neural Networks(CNNs),Recurrent Neural Networks(RNNs),Generative Adversarial Networks(GANs),Vision Transformers(ViTs),and hybrid models,are discussed in detail,including their advantages and domain-specific adaptations.Advanced learning paradigms such as semi-supervised learning,selfsupervised learning,and few-shot learning are explored for their potential to mitigate data annotation challenges in clinical datasets.This review further categorizes major tasks in medical image analysis,elaborating on how DL techniques have enabled precise tumor segmentation,lesion detection,modality fusion,super-resolution,and robust classification across diverse clinical settings.Emphasis is placed on applications in oncology,cardiology,neurology,and infectious diseases,including COVID-19.Challenges such as data scarcity,label imbalance,model generalizability,interpretability,and integration into clinical workflows are critically examined.Ethical considerations,explainable AI(XAI),federated learning,and regulatory compliance are discussed as essential components of real-world deployment.Benchmark datasets,evaluation metrics,and comparative performance analyses are presented to support future research.The article concludes with a forward-looking perspective on the role of foundation models,multimodal learning,edge AI,and bio-inspired computing in the future of medical imaging.Overall,this review serves as a valuable resource for researchers,clinicians,and developers aiming to harness deep learning for intelligent,efficient,and clinically viable medical image analysis.
基金supported by the projects of the National Key Research and Development Program of China(2024YFD2301305)Jiangsu Innovation Support Program for International Science and Technology Cooperation Project(BZ2023049)+2 种基金the projects of the National Natural Science Foundation of China(32272213)the China Agriculture Research System(CARS-03)Jiangsu Collaborative Innovation Center for Modern Crop Production(JCIC-MCP).
文摘Optimizing root system architecture(RSA)is essential for plants because of its critical role in acquiring water and nutrients from the soil.However,the subterranean nature of roots complicates the measurement of RSA traits.Recently developed rhizobox methods allow for the rapid acquisition of root images.Nevertheless,effective and precise approaches for extracting RSA features from these images remain underdeveloped.Deep learning(DL)technology can enhance image segmentation and facilitate RSA trait extraction.However,comprehensive pipelines that integrate DL technologies into image-based root phenotyping techniques are still scarce,hampering their implementation.To address this challenge,we present a reproducible pipeline(faCRSA)for automated RSA traits analysis,consisting of three modules:(1)the RSA traits extraction module functions to segment soil-root images and calculate RSA traits.A lightweight convolutional neural network(CNN)named RootSeg was proposed for efficient and accurate segmentation;(2)the data storage module,which stores image and text data from other modules;and(3)the web application module,which allows researchers to analyze data online in a user-friendly manner.The correlation coefficients(R^(2))of total root length,root surface area,and root volume calculated from faCRSA and manually measured results were 0.96**,0.97**,and 0.93**,respectively,with root mean square errors(RMSE)of 8.13 cm,1.68 cm^(2),and 0.05 cm^(3),processed at a rate of 9.74 s per image,indicating satisfying accuracy.faCRSA has also demonstrated satisfactory performance in dynamically monitoring root system changes under various stress conditions,such as drought or waterlogging.The detailed code and deployable package of faCRSA are provided for researchers with the potential to replace manual and semi-automated methods.
基金supported by the Deanship of Research and Graduate Studies at King Khalid University under Small Research Project grant number RGP1/139/45.
文摘Integrating multiple medical imaging techniques,including Magnetic Resonance Imaging(MRI),Computed Tomography,Positron Emission Tomography(PET),and ultrasound,provides a comprehensive view of the patient health status.Each of these methods contributes unique diagnostic insights,enhancing the overall assessment of patient condition.Nevertheless,the amalgamation of data from multiple modalities presents difficulties due to disparities in resolution,data collection methods,and noise levels.While traditional models like Convolutional Neural Networks(CNNs)excel in single-modality tasks,they struggle to handle multi-modal complexities,lacking the capacity to model global relationships.This research presents a novel approach for examining multi-modal medical imagery using a transformer-based system.The framework employs self-attention and cross-attention mechanisms to synchronize and integrate features across various modalities.Additionally,it shows resilience to variations in noise and image quality,making it adaptable for real-time clinical use.To address the computational hurdles linked to transformer models,particularly in real-time clinical applications in resource-constrained environments,several optimization techniques have been integrated to boost scalability and efficiency.Initially,a streamlined transformer architecture was adopted to minimize the computational load while maintaining model effectiveness.Methods such as model pruning,quantization,and knowledge distillation have been applied to reduce the parameter count and enhance the inference speed.Furthermore,efficient attention mechanisms such as linear or sparse attention were employed to alleviate the substantial memory and processing requirements of traditional self-attention operations.For further deployment optimization,researchers have implemented hardware-aware acceleration strategies,including the use of TensorRT and ONNX-based model compression,to ensure efficient execution on edge devices.These optimizations allow the approach to function effectively in real-time clinical settings,ensuring viability even in environments with limited resources.Future research directions include integrating non-imaging data to facilitate personalized treatment and enhancing computational efficiency for implementation in resource-limited environments.This study highlights the transformative potential of transformer models in multi-modal medical imaging,offering improvements in diagnostic accuracy and patient care outcomes.
文摘Deep learning models have become a core technological tool in the field of medical image analysis.However,these models often suffer from a lack of transparency in their decision-making processes,leading to challenges related to trust and interpret ability in clinical applications.To address this issue,explainable artificial intelligence(XAI)techniques have been applied to medical image analysis.While showing promising potential,XAI also brings significant ethical risks in practice—most notably,the problem of spurious explanations.Such explanations may rise further concerns regarding patient privacy,data security,and the attribution of decisionmaking authority in medical contexts.This paper analyzes the application of XAI methods—particularly saliency aps—in medical image interpretation,identifies the underlying causes of spurious explanations,and proposes possible mitigation strategies.The aim is to contribute to the responsible and sustainable integration of explainable AI into clinical practice.
文摘Given the importance of sentiment analysis in diverse environments,various methods are used for image sentiment analysis,including contextual sentiment analysis that utilizes character and scene relationships.However,most existing works employ character faces in conjunction with context,yet lack the capacity to analyze the emotions of characters in unconstrained environments,such as when their faces are obscured or blurred.Accordingly,this article presents the Adaptive Multi-Channel Sentiment Analysis Network(AMSA),a contextual image sentiment analysis framework,which consists of three channels:body,face,and context.AMSA employs Multi-task Cascaded Convolutional Networks(MTCNN)to detect faces within body frames;if detected,facial features are extracted and fused with body and context information for emotion recognition.If not,the model leverages body and context features alone.Meanwhile,to address class imbalance in the EMOTIC dataset,Focal Loss is introduced to improve classification performance,especially for minority emotion categories.Experimental results have shown that certain sentiment categories with lower representation in the dataset demonstrate leading classification accuracy,the AMSA yields a 2.53%increase compared with state-of-the-art methods.
基金support from the National Key Research and Development Program of China(No.2024YFB3713705)is acknowledgedWangzhong Mu would like to acknowledge the Strategic Mobility,Sweden(SSF,No.SM22-0039)+1 种基金the Swedish Foundation for International Cooperation in Research and Higher Education(STINT,No.IB2022-9228)the Jernkontoret(Sweden)for supporting this clean steel research.Gonghao Lian would like to acknowledge China Scholarship Council(CSC,No.202306080032).
文摘The detection and characterization of non-metallic inclusions are essential for clean steel production.Recently,imaging analysis combined with high-dimensional data processing of metallic materials using artificial intelligence(AI)-based machine learning(ML)has developed rapidly.This technique has achieved impressive results in the field of inclusion classification in process metallurgy.The present study surveys the ML modeling of inclusion prediction in advanced steels,including the detection,classification,and feature prediction of inclusions in different steel grades.Studies on clean steel with different features based on data and image analysis via ML are summarized.Regarding the data analysis,the inclusion prediction methodology based on ML establishes a connection between the experimental parameters and inclusion characteristics and analyzes the importance of the experimental parameters.Regarding the image analysis,the focus is placed on the classification of different types of inclusions via deep learning,in comparison with data analysis.Finally,further development of inclusion analyses using ML-based methods is recommended.This work paves the way for the application of AIbased methodologies for ultraclean-steel studies from a sustainable metallurgy perspective.
基金funded by the Ongoing Research Funding Program-Research Chairs(ORF-RC-2025-2400),King Saud University,Riyadh,Saudi Arabia。
文摘Recent studies indicate that millions of individuals suffer from renal diseases,with renal carcinoma,a type of kidney cancer,emerging as both a chronic illness and a significant cause of mortality.Magnetic Resonance Imaging(MRI)and Computed Tomography(CT)have become essential tools for diagnosing and assessing kidney disorders.However,accurate analysis of thesemedical images is critical for detecting and evaluating tumor severity.This study introduces an integrated hybrid framework that combines three complementary deep learning models for kidney tumor segmentation from MRI images.The proposed framework fuses a customized U-Net and Mask R-CNN using a weighted scheme to achieve semantic and instance-level segmentation.The fused outputs are further refined through edge detection using Stochastic FeatureMapping Neural Networks(SFMNN),while volumetric consistency is ensured through Improved Mini-Batch K-Means(IMBKM)clustering integrated with an Encoder-Decoder Convolutional Neural Network(EDCNN).The outputs of these three stages are combined through a weighted fusion mechanism,with optimal weights determined empirically.Experiments on MRI scans from the TCGA-KIRC dataset demonstrate that the proposed hybrid framework significantly outperforms standalone models,achieving a Dice Score of 92.5%,an IoU of 87.8%,a Precision of 93.1%,a Recall of 90.8%,and a Hausdorff Distance of 2.8 mm.These findings validate that the weighted integration of complementary architectures effectively overcomes key limitations in kidney tumor segmentation,leading to improved diagnostic accuracy and robustness in medical image analysis.
基金supported by the National Key Research and Development Project(Grant No.2023YFE0110900)the National Natural Science Foundation of China(Grant Nos.42320104003,42477168).
文摘Underground hydrogen storage has gained interest in recent years due to the enormous demand for clean energy.Hydrogen is more diffusive than air,with a smaller density and lower viscosity.These unique properties introduce distinctive hydrodynamic phenomena in hydrogen storage,one of which is fingering.Fingering could induce the fluid trapped in small clusters of pores,leading to a dramatic decrease in hydrogen saturation and a lower recovery rate.In this study,numerical simulations are performed at the microscopic scale to understand the evolution of hydrogen saturation and the impacts of injection and withdrawal cycles.Two sets of micromodels with different porosity(0.362 and 0.426)and minimum sizes of pore throats(0.362 mm and 0.181 mm)are developed in the numerical model.A parameter analysis is then conducted to understand the influence of injection velocity(in the range of 10^(-2)m/s to 10^(-5)m/s)and porous structure on the fingering pattern,followed by an image analysis to capture the evolution of the fingering pattern.Viscous fingering,capillary fingering,and crossover fingering are observed and identified under different boundary conditions.The fractal dimension,specific area,mean angle,and entropy of fingers are proposed as geometric descriptors to characterize the shape of the fingering pattern.When porosity increases from 0.362 to 0.426,the saturation of hydrogen increases by 26.2%.Narrower pore throats elevate capillary resistance,which hinders fluid invasion.These results underscore the importance of pore structures and the interaction between viscous and capillary forces for hydrogen recovery efficiency.This work illuminates the influence of the pore structures and the fluid properties on the immiscible displacement of hydrogen and can be further extended to optimize the injection strategy of hydrogen in underground hydrogen storage.
基金Projects(50934002,51074013,51304076,51104100)supported by the National Natural Science Foundation of ChinaProject(IRT0950)supported by the Program for Changjiang Scholars Innovative Research Team in Universities,ChinaProject(2012M510007)supported by China Postdoctoral Science Foundation
文摘Methods and procedures of three-dimensional （3D） characterization of the pore structure features in the packed ore particle bed are focused. X-ray computed tomography was applied to deriving the cross-sectional images of specimens with single particle size of 1-2, 2-3, 3-4, 4-5, 5-6, 6-7, 7-8, 8-9, 9-10 ram. Based on the in-house developed 3D image analysis programs using Matlab, the volume porosity, pore size distribution and degree of connectivity were calculated and analyzed in detail. The results indicate that the volume porosity, the mean diameter of pores and the effective pore size （d50） increase with the increasing of particle size. Lognormal distribution or Gauss distribution is mostly suitable to model the pore size distribution. The degree of connectivity investigated on the basis of cluster-labeling algorithm also increases with increasing the particle size approximately.
基金This project was supported by a grant from Hubei Provincial Natural Sciences Foundation of China (No. 2004 ABA200).
文摘The changes of retinal nuclear DNA content in rats after death was detected and the relationship between degradation of retinal nuclear DNA and postmortem interval （PMI） was analyzed. Ninety healthy adult SD rats, female, weighing 250±10 g, were randomly divided into 15 groups. At 20 ℃, the retinal cells were withdrawn every 2 h within 0 to 28 h after death and stained with Feulgen-Vans. Index of density （ID）, integral absorbance （IA） and average absorbance （AA） in retinal nucleus were analyzed by image analysis system. And the obtained data were subjected to linear regression analysis by using SPSS12.0 software. The results showed that in retinal nucleus, AA and IA were gradually declined with the prolongation of PMI, while ID had an increased tendency. Within 28 h after PMI, the regression equations were as follows： YAA=-0.009XAA＋0.590 （R^2=0.949）, YIA=0.097XIA＋18.903 （R^2=0.968）, YID=0.122XID＋2.246 （R^2=0.951）. It was concluded that retinal nuclear DNA after death in rats was degraded gradually and had a good correlation with PMI.
基金financially supported by the National Natural Science Foundation of China(No.51304076)the Natural Science Foundation of Hunan Province,China(No.14JJ4064)
文摘Mineral dissemination and pore space distribution in ore particles are important features that influence heap leaching performance.To quantify the mineral dissemination and pore space distribution of an ore particle,a cylindrical copper oxide ore sample(I center dot 4.6 mm x 5.6 mm)was scanned using high-resolution X-ray computed tomography(HRXCT),a nondestructive imaging technology,at a spatial resolution of 4.85 mu m.Combined with three-dimensional(3D)image analysis techniques,the main mineral phases and pore space were segmented and the volume fraction of each phase was calculated.In addition,the mass fraction of each mineral phase was estimated and the result was validated with that obtained using traditional techniques.Furthermore,the pore phase features,including the pore size distribution,pore surface area,pore fractal dimension,pore centerline,and the pore connectivity,were investigated quantitatively.The pore space analysis results indicate that the pore size distribution closely fits a log-normal distribution and that the pore space morphology is complicated,with a large surface area and low connectivity.This study demonstrates that the combination of HRXCT and 3D image analysis is an effective tool for acquiring 3D mineralogical and pore structural data.
基金Under the auspices of the National Natural Science Foundation of China (No. 40301038), Talents Recruitment Foun-dation of Nanjing University
文摘With the wide use of high-resolution remotely sensed imagery, the object-oriented remotely sensed informa- tion classification pattern has been intensively studied. Starting with the definition of object-oriented remotely sensed information classification pattern and a literature review of related research progress, this paper sums up 4 developing phases of object-oriented classification pattern during the past 20 years. Then, we discuss the three aspects of method- ology in detail, namely remotely sensed imagery segmentation, feature analysis and feature selection, and classification rule generation, through comparing them with remotely sensed information classification method based on per-pixel. At last, this paper presents several points that need to be paid attention to in the future studies on object-oriented RS in- formation classification pattern: 1) developing robust and highly effective image segmentation algorithm for multi-spectral RS imagery; 2) improving the feature-set including edge, spatial-adjacent and temporal characteristics; 3) discussing the classification rule generation classifier based on the decision tree; 4) presenting evaluation methods for classification result by object-oriented classification pattern.
基金Supported by the National Natural Science Foundation of China (No. 50376042)Doctoral Program Foundation of Institute of Higher Education of China (20040425007).
文摘Based on the analysis of high-speed video images, the detachment behavior of dust cake from the ceramic candle filter surface during pulse cleaning process is investigated. The influences of the dust cake loading,the reservoir pressure, and the filtration velocity on the cleaning effectiveness are analyzed. Experimental results show that there exists an optimum dust cake thickness for pulse-cleaning process. For thin dust cake, the patchy cleaning exists and the cleaning efficiency is low; if the dust cake is too thick, the pressure drop across the dust cake becomes higher and a higher reservoir pressure may be needed. At the same time there also exists an optimum reservoir pressure for a given filtration condition.
