Efficiently capturing multi-scale local information and building long-range dependencies among pixels are essential for medical image segmentation because of the various sizes and shapes of the lesion regions or organ...Efficiently capturing multi-scale local information and building long-range dependencies among pixels are essential for medical image segmentation because of the various sizes and shapes of the lesion regions or organs.In this paper,we propose the multi-scale cross-axis attention(MCA)mechanism to address these challenges through enhanced axial attention.To address the issues of insufficient learning of positional bias and limited long-distance interaction in axial attention caused by the small dataset,we propose using a dual cross-attention mechanism instead of axial attention to enhance global information capture.Meanwhile,to compensate for the lack of explicit attention to local information in axial attention,we use multiple convolutions of strip-shaped kernels with different kernel sizes in each axial attention path,which improves the efficiency of MCA in local information encoding.By integrating MCA into the multi-scale cross-axis attention network(MSCAN)backbone,we develop our network architecture,termed MCANet.With merely 4 M+parameters,MCANet outperforms previous heavyweight approaches(e.g.,swin transformer-based methods)across four challenging tasks:skin lesion segmentation,nuclei segmentation,abdominal multi-organ segmentation,and polyp segmentation.The code is available at https://github.com/haoshao-nku/medical_seg.展开更多
With the springing up of the MEMS industry,research on accelerometers is focused on miniaturization, integration,high reliability,and high resolution,and shares extensive application prospects in military and civil fi...With the springing up of the MEMS industry,research on accelerometers is focused on miniaturization, integration,high reliability,and high resolution,and shares extensive application prospects in military and civil fields.Comparing with the traditional single cantilever beam structure or ＂cantilever-mass＂ structure,the proposed structure of ＂8-beams/mass＂ with its varistor completely symmetric distribution in micro-silicon piezoresistive triaxial accelerometer in this paper has a higher axial sensitivity and smaller cross-axis sensitivity.Adopting ANSYS, the process of structural analysis and the manufacturing flow of sensing unit are showed.In dynamic testing conditions, it can be concluded that the axial sensitivity of x,y,and z are Sx-48μV/g,Sy = 54μV/g and Sz = 217μV/g respectively,and the nonlinearities are 0.4%,0.6%and 0.4%.展开更多
基金supported by the Science and Technology Support Program of Tianjin,China(No.23JCZDJC01050).
文摘Efficiently capturing multi-scale local information and building long-range dependencies among pixels are essential for medical image segmentation because of the various sizes and shapes of the lesion regions or organs.In this paper,we propose the multi-scale cross-axis attention(MCA)mechanism to address these challenges through enhanced axial attention.To address the issues of insufficient learning of positional bias and limited long-distance interaction in axial attention caused by the small dataset,we propose using a dual cross-attention mechanism instead of axial attention to enhance global information capture.Meanwhile,to compensate for the lack of explicit attention to local information in axial attention,we use multiple convolutions of strip-shaped kernels with different kernel sizes in each axial attention path,which improves the efficiency of MCA in local information encoding.By integrating MCA into the multi-scale cross-axis attention network(MSCAN)backbone,we develop our network architecture,termed MCANet.With merely 4 M+parameters,MCANet outperforms previous heavyweight approaches(e.g.,swin transformer-based methods)across four challenging tasks:skin lesion segmentation,nuclei segmentation,abdominal multi-organ segmentation,and polyp segmentation.The code is available at https://github.com/haoshao-nku/medical_seg.
基金supported by the International Science & Technology Cooperation Program of China(No.61011140351)the National High Technology Research and Development Program of China(No.2011AA040404)the National Natural Science Foundation of China(No. 51075375)
文摘With the springing up of the MEMS industry,research on accelerometers is focused on miniaturization, integration,high reliability,and high resolution,and shares extensive application prospects in military and civil fields.Comparing with the traditional single cantilever beam structure or ＂cantilever-mass＂ structure,the proposed structure of ＂8-beams/mass＂ with its varistor completely symmetric distribution in micro-silicon piezoresistive triaxial accelerometer in this paper has a higher axial sensitivity and smaller cross-axis sensitivity.Adopting ANSYS, the process of structural analysis and the manufacturing flow of sensing unit are showed.In dynamic testing conditions, it can be concluded that the axial sensitivity of x,y,and z are Sx-48μV/g,Sy = 54μV/g and Sz = 217μV/g respectively,and the nonlinearities are 0.4%,0.6%and 0.4%.
