Segmenting a breast ultrasound image is still challenging due to the presence of speckle noise,dependency on the operator,and the variation of image quality.This paper presents the UltraSegNet architecture that addres...Segmenting a breast ultrasound image is still challenging due to the presence of speckle noise,dependency on the operator,and the variation of image quality.This paper presents the UltraSegNet architecture that addresses these challenges through three key technical innovations:This work adds three things:(1)a changed ResNet-50 backbone with sequential 3×3 convolutions to keep fine anatomical details that are needed for finding lesion boundaries;(2)a computationally efficient regional attention mechanism that works on high-resolution features without using a transformer’s extra memory;and(3)an adaptive feature fusion strategy that changes local and global featuresbasedonhowthe image isbeing used.Extensive evaluation on two distinct datasets demonstrates UltraSegNet’s superior performance:On the BUSI dataset,it obtains a precision of 0.915,a recall of 0.908,and an F1 score of 0.911.In the UDAIT dataset,it achieves robust performance across the board,with a precision of 0.901 and recall of 0.894.Importantly,these improvements are achieved at clinically feasible computation times,taking 235 ms per image on standard GPU hardware.Notably,UltraSegNet does amazingly well on difficult small lesions(less than 10 mm),achieving a detection accuracy of 0.891.This is a huge improvement over traditional methods that have a hard time with small-scale features,as standard models can only achieve 0.63–0.71 accuracy.This improvement in small lesion detection is particularly crucial for early-stage breast cancer identification.Results from this work demonstrate that UltraSegNet can be practically deployable in clinical workflows to improve breast cancer screening accuracy.展开更多
Heart disease remains a leading cause of morbidity and mortality worldwide,highlighting the need for improved diagnostic methods.Traditional diagnostics face limitations such as reliance on single-modality data and vu...Heart disease remains a leading cause of morbidity and mortality worldwide,highlighting the need for improved diagnostic methods.Traditional diagnostics face limitations such as reliance on single-modality data and vulnerability to apparatus faults,which can reduce accuracy,especially with poor-quality images.Additionally,these methods often require significant time and expertise,making them less accessible in resource-limited settings.Emerging technologies like artificial intelligence and machine learning offer promising solutions by integrating multi-modality data and enhancing diagnostic precision,ultimately improving patient outcomes and reducing healthcare costs.This study introduces Heart-Net,a multi-modal deep learning framework designed to enhance heart disease diagnosis by integrating data from Cardiac Magnetic Resonance Imaging(MRI)and Electrocardiogram(ECG).Heart-Net uses a 3D U-Net for MRI analysis and a Temporal Convolutional Graph Neural Network(TCGN)for ECG feature extraction,combining these through an attention mechanism to emphasize relevant features.Classification is performed using Optimized TCGN.This approach improves early detection,reduces diagnostic errors,and supports personalized risk assessments and continuous health monitoring.The proposed approach results show that Heart-Net significantly outperforms traditional single-modality models,achieving accuracies of 92.56%forHeartnetDataset Ⅰ(HNET-DSⅠ),93.45%forHeartnetDataset Ⅱ(HNET-DSⅡ),and 91.89%for Heartnet Dataset Ⅲ(HNET-DSⅢ),mitigating the impact of apparatus faults and image quality issues.These findings underscore the potential of Heart-Net to revolutionize heart disease diagnostics and improve clinical outcomes.展开更多
基金funded by Princess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2025R435),Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.
文摘Segmenting a breast ultrasound image is still challenging due to the presence of speckle noise,dependency on the operator,and the variation of image quality.This paper presents the UltraSegNet architecture that addresses these challenges through three key technical innovations:This work adds three things:(1)a changed ResNet-50 backbone with sequential 3×3 convolutions to keep fine anatomical details that are needed for finding lesion boundaries;(2)a computationally efficient regional attention mechanism that works on high-resolution features without using a transformer’s extra memory;and(3)an adaptive feature fusion strategy that changes local and global featuresbasedonhowthe image isbeing used.Extensive evaluation on two distinct datasets demonstrates UltraSegNet’s superior performance:On the BUSI dataset,it obtains a precision of 0.915,a recall of 0.908,and an F1 score of 0.911.In the UDAIT dataset,it achieves robust performance across the board,with a precision of 0.901 and recall of 0.894.Importantly,these improvements are achieved at clinically feasible computation times,taking 235 ms per image on standard GPU hardware.Notably,UltraSegNet does amazingly well on difficult small lesions(less than 10 mm),achieving a detection accuracy of 0.891.This is a huge improvement over traditional methods that have a hard time with small-scale features,as standard models can only achieve 0.63–0.71 accuracy.This improvement in small lesion detection is particularly crucial for early-stage breast cancer identification.Results from this work demonstrate that UltraSegNet can be practically deployable in clinical workflows to improve breast cancer screening accuracy.
基金funded by Princess Nourah bint Abdulrahman University Researchers Supporting Project Number(PNURSP2024R435),Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.
文摘Heart disease remains a leading cause of morbidity and mortality worldwide,highlighting the need for improved diagnostic methods.Traditional diagnostics face limitations such as reliance on single-modality data and vulnerability to apparatus faults,which can reduce accuracy,especially with poor-quality images.Additionally,these methods often require significant time and expertise,making them less accessible in resource-limited settings.Emerging technologies like artificial intelligence and machine learning offer promising solutions by integrating multi-modality data and enhancing diagnostic precision,ultimately improving patient outcomes and reducing healthcare costs.This study introduces Heart-Net,a multi-modal deep learning framework designed to enhance heart disease diagnosis by integrating data from Cardiac Magnetic Resonance Imaging(MRI)and Electrocardiogram(ECG).Heart-Net uses a 3D U-Net for MRI analysis and a Temporal Convolutional Graph Neural Network(TCGN)for ECG feature extraction,combining these through an attention mechanism to emphasize relevant features.Classification is performed using Optimized TCGN.This approach improves early detection,reduces diagnostic errors,and supports personalized risk assessments and continuous health monitoring.The proposed approach results show that Heart-Net significantly outperforms traditional single-modality models,achieving accuracies of 92.56%forHeartnetDataset Ⅰ(HNET-DSⅠ),93.45%forHeartnetDataset Ⅱ(HNET-DSⅡ),and 91.89%for Heartnet Dataset Ⅲ(HNET-DSⅢ),mitigating the impact of apparatus faults and image quality issues.These findings underscore the potential of Heart-Net to revolutionize heart disease diagnostics and improve clinical outcomes.
