critical for guiding treatment and improving patient outcomes.Traditional molecular subtyping via immuno-histochemistry(IHC)test is invasive,time-consuming,and may not fully represent tumor heterogeneity.This study pr...critical for guiding treatment and improving patient outcomes.Traditional molecular subtyping via immuno-histochemistry(IHC)test is invasive,time-consuming,and may not fully represent tumor heterogeneity.This study proposes a non-invasive approach using digital mammography images and deep learning algorithm for classifying breast cancer molecular subtypes.Four pretrained models,including two Convolutional Neural Networks(MobileNet_V3_Large and VGG-16)and two Vision Transformers(ViT_B_16 and ViT_Base_Patch16_Clip_224)were fine-tuned to classify images into HER2-enriched,Luminal,Normal-like,and Triple Negative subtypes.Hyperparameter tuning,including learning rate adjustment and layer freezing strategies,was applied to optimize performance.Among the evaluated models,ViT_Base_Patch16_Clip_224 achieved the highest test accuracy(94.44%),with equally high precision,recall,and F1-score of 0.94,demonstrating excellent generalization.MobileNet_V3_Large achieved the same accuracy but showed less training stability.In contrast,VGG-16 recorded the lowest performance,indicating a limitation in its generalizability for this classification task.The study also highlighted the superior performance of the Vision Transformer models over CNNs,particularly due to their ability to capture global contextual features and the benefit of CLIP-based pretraining in ViT_Base_Patch16_Clip_224.To enhance clinical applicability,a graphical user interface(GUI)named“BCMS Dx”was developed for streamlined subtype prediction.Deep learning applied to mammography has proven effective for accurate and non-invasive molecular subtyping.The proposed Vision Transformer-based model and supporting GUI offer a promising direction for augmenting diagnostic workflows,minimizing the need for invasive procedures,and advancing personalized breast cancer management.展开更多
The rapid growth of unlabeled time-series data in domains such as wireless communications,radar,biomedical engineering,and the Internet of Things(IoT)has driven advancements in unsupervised learning.This review synthe...The rapid growth of unlabeled time-series data in domains such as wireless communications,radar,biomedical engineering,and the Internet of Things(IoT)has driven advancements in unsupervised learning.This review synthe-sizes recent progress in applying autoencoders and vision transformers for un-supervised signal analysis,focusing on their architectures,applications,and emerging trends.We explore how these models enable feature extraction,anomaly detection,and classification across diverse signal types,including electrocardiograms,radar waveforms,and IoT sensor data.The review high-lights the strengths of hybrid architectures and self-supervised learning,while identifying challenges in interpretability,scalability,and domain generaliza-tion.By bridging methodological innovations and practical applications,this work offers a roadmap for developing robust,adaptive models for signal in-telligence.展开更多
Accurate estimation of diffuse horizontal irradiance(DHI)is critical for optimising photovoltaic system performance and energy forecasting yet remains challenging in regions lacking comprehensive ground-based instrume...Accurate estimation of diffuse horizontal irradiance(DHI)is critical for optimising photovoltaic system performance and energy forecasting yet remains challenging in regions lacking comprehensive ground-based instrumentation.Recent advancements using Vision Transformers(ViTs)trained on extensive sky image datasets have shown promise in replacing costly irradiance measurement equipment,but the scarcity of long-term,high-quality sky imagery significantly restricts practical implementation.Addressing this critical gap,this study proposes a novel dual-framework approach designed for data-scarce scenarios.First,calculated atmospheric parameters,including extraterrestrial irradiance and cyclic time encodings,are integrated to represent sky conditions without utilising any instrumentation.Next,a sequential pipeline initially predicts synthetic global horizontal irradiance(GHI)and uses it as a feature,to refine DHI estimation.Finally,a dual-parallel architecture simultaneously processes raw and overlay-enhanced fisheye sky images.Overlays are generated through unsupervised,physicsinformed cloud segmentation to highlight dynamic sky features.Empirical validation is performed using data from the Chilbolton Observatory,chosen for its temperate climate and frequent cloud variability.To simulate data-scarce conditions,models are trained on a single month(e.g.,January)and evaluated across a temporally disjoint,full-year test set.Under this setup,the sequential and dual-parallel frameworks achieve RMSE values within 2-3 W/m^(2)and 1-6 W/m^(2),respectively,of a state-of-the-art ViT trained on the complete dataset.By combining physics-informed modelling with unsupervised segmentation,the proposed method provides a scalable and cost-effective solution for DHI estimation,advancing solar resource assessment in data-constrained environments.展开更多
Lung cancer remains a major global health challenge,with early diagnosis crucial for improved patient survival.Traditional diagnostic techniques,including manual histopathology and radiological assessments,are prone t...Lung cancer remains a major global health challenge,with early diagnosis crucial for improved patient survival.Traditional diagnostic techniques,including manual histopathology and radiological assessments,are prone to errors and variability.Deep learning methods,particularly Vision Transformers(ViT),have shown promise for improving diagnostic accuracy by effectively extracting global features.However,ViT-based approaches face challenges related to computational complexity and limited generalizability.This research proposes the DualSet ViT-PSO-SVM framework,integrating aViTwith dual attentionmechanisms,Particle Swarm Optimization(PSO),and SupportVector Machines(SVM),aiming for efficient and robust lung cancer classification acrossmultiple medical image datasets.The study utilized three publicly available datasets:LIDC-IDRI,LUNA16,and TCIA,encompassing computed tomography(CT)scans and histopathological images.Data preprocessing included normalization,augmentation,and segmentation.Dual attention mechanisms enhanced ViT’s feature extraction capabilities.PSO optimized feature selection,and SVM performed classification.Model performance was evaluated on individual and combined datasets,benchmarked against CNN-based and standard ViT approaches.The DualSet ViT-PSO-SVM significantly outperformed existing methods,achieving superior accuracy rates of 97.85%(LIDC-IDRI),98.32%(LUNA16),and 96.75%(TCIA).Crossdataset evaluations demonstrated strong generalization capabilities and stability across similar imagingmodalities.The proposed framework effectively bridges advanced deep learning techniques with clinical applicability,offering a robust diagnostic tool for lung cancer detection,reducing complexity,and improving diagnostic reliability and interpretability.展开更多
Foreign body classification on coal conveyor belts is a critical component of intelligent coal mining systems.Previous approaches have primarily utilized convolutional neural networks(CNNs)to effectively integrate spa...Foreign body classification on coal conveyor belts is a critical component of intelligent coal mining systems.Previous approaches have primarily utilized convolutional neural networks(CNNs)to effectively integrate spatial and semantic information.However,the performance of CNN-based methods remains limited in classification accuracy,primarily due to insufficient exploration of local image characteristics.Unlike CNNs,Vision Transformer(ViT)captures discriminative features by modeling relationships between local image patches.However,such methods typically require a large number of training samples to perform effectively.In the context of foreign body classification on coal conveyor belts,the limited availability of training samples hinders the full exploitation of Vision Transformer’s(ViT)capabilities.To address this issue,we propose an efficient approach,termed Key Part-level Attention Vision Transformer(KPA-ViT),which incorporates key local information into the transformer architecture to enrich the training information.It comprises three main components:a key-point detection module,a key local mining module,and an attention module.To extract key local regions,a key-point detection strategy is first employed to identify the positions of key points.Subsequently,the key local mining module extracts the relevant local features based on these detected points.Finally,an attention module composed of self-attention and cross-attention blocks is introduced to integrate global and key part-level information,thereby enhancing the model’s ability to learn discriminative features.Compared to recent transformer-based frameworks—such as ViT,Swin-Transformer,and EfficientViT—the proposed KPA-ViT achieves performance improvements of 9.3%,6.6%,and 2.8%,respectively,on the CUMT-BelT dataset,demonstrating its effectiveness.展开更多
Accurate plant species classification is essential for many applications,such as biodiversity conservation,ecological research,and sustainable agricultural practices.Traditional morphological classification methods ar...Accurate plant species classification is essential for many applications,such as biodiversity conservation,ecological research,and sustainable agricultural practices.Traditional morphological classification methods are inherently slow,labour-intensive,and prone to inaccuracies,especiallywhen distinguishing between species exhibiting visual similarities or high intra-species variability.To address these limitations and to overcome the constraints of imageonly approaches,we introduce a novel Artificial Intelligence-driven framework.This approach integrates robust Vision Transformer(ViT)models for advanced visual analysis with a multi-modal data fusion strategy,incorporating contextual metadata such as precise environmental conditions,geographic location,and phenological traits.This combination of visual and ecological cues significantly enhances classification accuracy and robustness,proving especially vital in complex,heterogeneous real-world environments.The proposedmodel achieves an impressive 97.27%of test accuracy,andMean Reciprocal Rank(MRR)of 0.9842 that demonstrates strong generalization capabilities.Furthermore,efficient utilization of high-performance GPU resources(RTX 3090,18 GB memory)ensures scalable processing of highdimensional data.Comparative analysis consistently confirms that ourmetadata fusion approach substantially improves classification performance,particularly formorphologically similar species,and through principled self-supervised and transfer learning from ImageNet,the model adapts efficiently to new species,ensuring enhanced generalization.This comprehensive approach holds profound practical implications for precise conservation initiatives,rigorous ecological monitoring,and advanced agricultural management.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
This paper tackles the high computational/space complexity associated with multi-head self-attention(MHSA)in vanilla vision transformers.To this end,we propose hierarchical MHSA(H-MHSA),a novel approach that computes ...This paper tackles the high computational/space complexity associated with multi-head self-attention(MHSA)in vanilla vision transformers.To this end,we propose hierarchical MHSA(H-MHSA),a novel approach that computes self-attention in a hierarchical fashion.Specifically,we first divide the input image into patches as commonly done,and each patch is viewed as a token.Then,the proposed H-MHSA learns token relationships within local patches,serving as local relationship modeling.Then,the small patches are merged into larger ones,and H-MHSA models the global dependencies for the small number of the merged tokens.At last,the local and global attentive features are aggregated to obtain features with powerful representation capacity.Since we only calculate attention for a limited number of tokens at each step,the computational load is reduced dramatically.Hence,H-MHSA can efficiently model global relationships among tokens without sacrificing fine-grained information.With the H-MHSA module incorporated,we build a family of hierarchical-attention-based transformer networks,namely HAT-Net.To demonstrate the superiority of HAT-Net in scene understanding,we conduct extensive experiments on fundamental vision tasks,including image classification,semantic segmentation,object detection and instance segmentation.Therefore,HAT-Net provides a new perspective for vision transformers.Code and pretrained models are available at https://github.com/yun-liu/HAT-Net.展开更多
Most polyp segmentation methods use convolutional neural networks(CNNs)as their backbone,leading to two key issues when exchanging information between the encoder and decoder:(1)taking into account the differences in ...Most polyp segmentation methods use convolutional neural networks(CNNs)as their backbone,leading to two key issues when exchanging information between the encoder and decoder:(1)taking into account the differences in contribution between different-level features,and(2)designing an effective mechanism for fusing these features.Unlike existing CNN-based methods,we adopt a transformer encoder,which learns more powerful and robust representations.In addition,considering the image acquisition influence and elusive properties of polyps,we introduce three standard modules,including a cascaded fusion module(CFM),a camouflage identification module(CIM),and a similarity aggregation module(SAM).Among these,the CFM is used to collect the semantic and location information of polyps from high-level features;the CIM is applied to capture polyp information disguised in low-level features,and the SAM extends the pixel features of the polyp area with high-level semantic position information to the entire polyp area,thereby effectively fusing cross-level features.The proposed model,named Polyp-PVT,effectively suppresses noises in the features and significantly improves their expressive capabilities.Extensive experiments on five widely adopted datasets show that the proposed model is more robust to various challenging situations(e.g.,appearance changes,small objects,and rotation)than existing representative methods.The proposed model is available at https://github.com/DengPingFan/Polyp-PVT.展开更多
Detecting pavement cracks is critical for road safety and infrastructure management.Traditional methods,relying on manual inspection and basic image processing,are time-consuming and prone to errors.Recent deep-learni...Detecting pavement cracks is critical for road safety and infrastructure management.Traditional methods,relying on manual inspection and basic image processing,are time-consuming and prone to errors.Recent deep-learning(DL)methods automate crack detection,but many still struggle with variable crack patterns and environmental conditions.This study aims to address these limitations by introducing the Masker Transformer,a novel hybrid deep learning model that integrates the precise localization capabilities of Mask Region-based Convolutional Neural Network(Mask R-CNN)with the global contextual awareness of Vision Transformer(ViT).The research focuses on leveraging the strengths of both architectures to enhance segmentation accuracy and adaptability across different pavement conditions.We evaluated the performance of theMaskerTransformer against other state-of-theartmodels such asU-Net,TransformerU-Net(TransUNet),U-NetTransformer(UNETr),SwinU-NetTransformer(Swin-UNETr),You Only Look Once version 8(YoloV8),and Mask R-CNN using two benchmark datasets:Crack500 and DeepCrack.The findings reveal that the MaskerTransformer significantly outperforms the existing models,achieving the highest Dice SimilarityCoefficient(DSC),precision,recall,and F1-Score across both datasets.Specifically,the model attained a DSC of 80.04%on Crack500 and 91.37%on DeepCrack,demonstrating superior segmentation accuracy and reliability.The high precision and recall rates further substantiate its effectiveness in real-world applications,suggesting that the Masker Transformer can serve as a robust tool for automated pavement crack detection,potentially replacing more traditional methods.展开更多
基金funded by the Ministry of Higher Education(MoHE)Malaysia through the Fundamental Research Grant Scheme—Early Career Researcher(FRGS-EC),grant number FRGSEC/1/2024/ICT02/UNIMAP/02/8.
文摘critical for guiding treatment and improving patient outcomes.Traditional molecular subtyping via immuno-histochemistry(IHC)test is invasive,time-consuming,and may not fully represent tumor heterogeneity.This study proposes a non-invasive approach using digital mammography images and deep learning algorithm for classifying breast cancer molecular subtypes.Four pretrained models,including two Convolutional Neural Networks(MobileNet_V3_Large and VGG-16)and two Vision Transformers(ViT_B_16 and ViT_Base_Patch16_Clip_224)were fine-tuned to classify images into HER2-enriched,Luminal,Normal-like,and Triple Negative subtypes.Hyperparameter tuning,including learning rate adjustment and layer freezing strategies,was applied to optimize performance.Among the evaluated models,ViT_Base_Patch16_Clip_224 achieved the highest test accuracy(94.44%),with equally high precision,recall,and F1-score of 0.94,demonstrating excellent generalization.MobileNet_V3_Large achieved the same accuracy but showed less training stability.In contrast,VGG-16 recorded the lowest performance,indicating a limitation in its generalizability for this classification task.The study also highlighted the superior performance of the Vision Transformer models over CNNs,particularly due to their ability to capture global contextual features and the benefit of CLIP-based pretraining in ViT_Base_Patch16_Clip_224.To enhance clinical applicability,a graphical user interface(GUI)named“BCMS Dx”was developed for streamlined subtype prediction.Deep learning applied to mammography has proven effective for accurate and non-invasive molecular subtyping.The proposed Vision Transformer-based model and supporting GUI offer a promising direction for augmenting diagnostic workflows,minimizing the need for invasive procedures,and advancing personalized breast cancer management.
文摘The rapid growth of unlabeled time-series data in domains such as wireless communications,radar,biomedical engineering,and the Internet of Things(IoT)has driven advancements in unsupervised learning.This review synthe-sizes recent progress in applying autoencoders and vision transformers for un-supervised signal analysis,focusing on their architectures,applications,and emerging trends.We explore how these models enable feature extraction,anomaly detection,and classification across diverse signal types,including electrocardiograms,radar waveforms,and IoT sensor data.The review high-lights the strengths of hybrid architectures and self-supervised learning,while identifying challenges in interpretability,scalability,and domain generaliza-tion.By bridging methodological innovations and practical applications,this work offers a roadmap for developing robust,adaptive models for signal in-telligence.
基金support of the Engineering and Phys-ical Sciences Research Council(EPSRC)Doctoral Training Partnership(DTP)Fundingsupport from EPSRC grant EP/X033333/1.
文摘Accurate estimation of diffuse horizontal irradiance(DHI)is critical for optimising photovoltaic system performance and energy forecasting yet remains challenging in regions lacking comprehensive ground-based instrumentation.Recent advancements using Vision Transformers(ViTs)trained on extensive sky image datasets have shown promise in replacing costly irradiance measurement equipment,but the scarcity of long-term,high-quality sky imagery significantly restricts practical implementation.Addressing this critical gap,this study proposes a novel dual-framework approach designed for data-scarce scenarios.First,calculated atmospheric parameters,including extraterrestrial irradiance and cyclic time encodings,are integrated to represent sky conditions without utilising any instrumentation.Next,a sequential pipeline initially predicts synthetic global horizontal irradiance(GHI)and uses it as a feature,to refine DHI estimation.Finally,a dual-parallel architecture simultaneously processes raw and overlay-enhanced fisheye sky images.Overlays are generated through unsupervised,physicsinformed cloud segmentation to highlight dynamic sky features.Empirical validation is performed using data from the Chilbolton Observatory,chosen for its temperate climate and frequent cloud variability.To simulate data-scarce conditions,models are trained on a single month(e.g.,January)and evaluated across a temporally disjoint,full-year test set.Under this setup,the sequential and dual-parallel frameworks achieve RMSE values within 2-3 W/m^(2)and 1-6 W/m^(2),respectively,of a state-of-the-art ViT trained on the complete dataset.By combining physics-informed modelling with unsupervised segmentation,the proposed method provides a scalable and cost-effective solution for DHI estimation,advancing solar resource assessment in data-constrained environments.
文摘Lung cancer remains a major global health challenge,with early diagnosis crucial for improved patient survival.Traditional diagnostic techniques,including manual histopathology and radiological assessments,are prone to errors and variability.Deep learning methods,particularly Vision Transformers(ViT),have shown promise for improving diagnostic accuracy by effectively extracting global features.However,ViT-based approaches face challenges related to computational complexity and limited generalizability.This research proposes the DualSet ViT-PSO-SVM framework,integrating aViTwith dual attentionmechanisms,Particle Swarm Optimization(PSO),and SupportVector Machines(SVM),aiming for efficient and robust lung cancer classification acrossmultiple medical image datasets.The study utilized three publicly available datasets:LIDC-IDRI,LUNA16,and TCIA,encompassing computed tomography(CT)scans and histopathological images.Data preprocessing included normalization,augmentation,and segmentation.Dual attention mechanisms enhanced ViT’s feature extraction capabilities.PSO optimized feature selection,and SVM performed classification.Model performance was evaluated on individual and combined datasets,benchmarked against CNN-based and standard ViT approaches.The DualSet ViT-PSO-SVM significantly outperformed existing methods,achieving superior accuracy rates of 97.85%(LIDC-IDRI),98.32%(LUNA16),and 96.75%(TCIA).Crossdataset evaluations demonstrated strong generalization capabilities and stability across similar imagingmodalities.The proposed framework effectively bridges advanced deep learning techniques with clinical applicability,offering a robust diagnostic tool for lung cancer detection,reducing complexity,and improving diagnostic reliability and interpretability.
基金funded by the National Key Research and Development Program of China(grant number 2023YFC2907600)the National Natural Science Foundation of China(grant number 52504132)Tiandi Science and Technology Co.,Ltd.Science and Technology Innovation Venture Capital Special Project(grant number 2023-TD-ZD011-004).
文摘Foreign body classification on coal conveyor belts is a critical component of intelligent coal mining systems.Previous approaches have primarily utilized convolutional neural networks(CNNs)to effectively integrate spatial and semantic information.However,the performance of CNN-based methods remains limited in classification accuracy,primarily due to insufficient exploration of local image characteristics.Unlike CNNs,Vision Transformer(ViT)captures discriminative features by modeling relationships between local image patches.However,such methods typically require a large number of training samples to perform effectively.In the context of foreign body classification on coal conveyor belts,the limited availability of training samples hinders the full exploitation of Vision Transformer’s(ViT)capabilities.To address this issue,we propose an efficient approach,termed Key Part-level Attention Vision Transformer(KPA-ViT),which incorporates key local information into the transformer architecture to enrich the training information.It comprises three main components:a key-point detection module,a key local mining module,and an attention module.To extract key local regions,a key-point detection strategy is first employed to identify the positions of key points.Subsequently,the key local mining module extracts the relevant local features based on these detected points.Finally,an attention module composed of self-attention and cross-attention blocks is introduced to integrate global and key part-level information,thereby enhancing the model’s ability to learn discriminative features.Compared to recent transformer-based frameworks—such as ViT,Swin-Transformer,and EfficientViT—the proposed KPA-ViT achieves performance improvements of 9.3%,6.6%,and 2.8%,respectively,on the CUMT-BelT dataset,demonstrating its effectiveness.
文摘Accurate plant species classification is essential for many applications,such as biodiversity conservation,ecological research,and sustainable agricultural practices.Traditional morphological classification methods are inherently slow,labour-intensive,and prone to inaccuracies,especiallywhen distinguishing between species exhibiting visual similarities or high intra-species variability.To address these limitations and to overcome the constraints of imageonly approaches,we introduce a novel Artificial Intelligence-driven framework.This approach integrates robust Vision Transformer(ViT)models for advanced visual analysis with a multi-modal data fusion strategy,incorporating contextual metadata such as precise environmental conditions,geographic location,and phenological traits.This combination of visual and ecological cues significantly enhances classification accuracy and robustness,proving especially vital in complex,heterogeneous real-world environments.The proposedmodel achieves an impressive 97.27%of test accuracy,andMean Reciprocal Rank(MRR)of 0.9842 that demonstrates strong generalization capabilities.Furthermore,efficient utilization of high-performance GPU resources(RTX 3090,18 GB memory)ensures scalable processing of highdimensional data.Comparative analysis consistently confirms that ourmetadata fusion approach substantially improves classification performance,particularly formorphologically similar species,and through principled self-supervised and transfer learning from ImageNet,the model adapts efficiently to new species,ensuring enhanced generalization.This comprehensive approach holds profound practical implications for precise conservation initiatives,rigorous ecological monitoring,and advanced agricultural management.
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.
基金supported by A*STAR Career Development Fund,Singapore(No.C233312006)。
文摘This paper tackles the high computational/space complexity associated with multi-head self-attention(MHSA)in vanilla vision transformers.To this end,we propose hierarchical MHSA(H-MHSA),a novel approach that computes self-attention in a hierarchical fashion.Specifically,we first divide the input image into patches as commonly done,and each patch is viewed as a token.Then,the proposed H-MHSA learns token relationships within local patches,serving as local relationship modeling.Then,the small patches are merged into larger ones,and H-MHSA models the global dependencies for the small number of the merged tokens.At last,the local and global attentive features are aggregated to obtain features with powerful representation capacity.Since we only calculate attention for a limited number of tokens at each step,the computational load is reduced dramatically.Hence,H-MHSA can efficiently model global relationships among tokens without sacrificing fine-grained information.With the H-MHSA module incorporated,we build a family of hierarchical-attention-based transformer networks,namely HAT-Net.To demonstrate the superiority of HAT-Net in scene understanding,we conduct extensive experiments on fundamental vision tasks,including image classification,semantic segmentation,object detection and instance segmentation.Therefore,HAT-Net provides a new perspective for vision transformers.Code and pretrained models are available at https://github.com/yun-liu/HAT-Net.
文摘Most polyp segmentation methods use convolutional neural networks(CNNs)as their backbone,leading to two key issues when exchanging information between the encoder and decoder:(1)taking into account the differences in contribution between different-level features,and(2)designing an effective mechanism for fusing these features.Unlike existing CNN-based methods,we adopt a transformer encoder,which learns more powerful and robust representations.In addition,considering the image acquisition influence and elusive properties of polyps,we introduce three standard modules,including a cascaded fusion module(CFM),a camouflage identification module(CIM),and a similarity aggregation module(SAM).Among these,the CFM is used to collect the semantic and location information of polyps from high-level features;the CIM is applied to capture polyp information disguised in low-level features,and the SAM extends the pixel features of the polyp area with high-level semantic position information to the entire polyp area,thereby effectively fusing cross-level features.The proposed model,named Polyp-PVT,effectively suppresses noises in the features and significantly improves their expressive capabilities.Extensive experiments on five widely adopted datasets show that the proposed model is more robust to various challenging situations(e.g.,appearance changes,small objects,and rotation)than existing representative methods.The proposed model is available at https://github.com/DengPingFan/Polyp-PVT.
文摘Detecting pavement cracks is critical for road safety and infrastructure management.Traditional methods,relying on manual inspection and basic image processing,are time-consuming and prone to errors.Recent deep-learning(DL)methods automate crack detection,but many still struggle with variable crack patterns and environmental conditions.This study aims to address these limitations by introducing the Masker Transformer,a novel hybrid deep learning model that integrates the precise localization capabilities of Mask Region-based Convolutional Neural Network(Mask R-CNN)with the global contextual awareness of Vision Transformer(ViT).The research focuses on leveraging the strengths of both architectures to enhance segmentation accuracy and adaptability across different pavement conditions.We evaluated the performance of theMaskerTransformer against other state-of-theartmodels such asU-Net,TransformerU-Net(TransUNet),U-NetTransformer(UNETr),SwinU-NetTransformer(Swin-UNETr),You Only Look Once version 8(YoloV8),and Mask R-CNN using two benchmark datasets:Crack500 and DeepCrack.The findings reveal that the MaskerTransformer significantly outperforms the existing models,achieving the highest Dice SimilarityCoefficient(DSC),precision,recall,and F1-Score across both datasets.Specifically,the model attained a DSC of 80.04%on Crack500 and 91.37%on DeepCrack,demonstrating superior segmentation accuracy and reliability.The high precision and recall rates further substantiate its effectiveness in real-world applications,suggesting that the Masker Transformer can serve as a robust tool for automated pavement crack detection,potentially replacing more traditional methods.
