<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[Stark_Z的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/Stark_Z</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; Stark_Z]]></copyright><item><title><![CDATA[论文笔记CMU-NET: A STRONG CONVMIXER-BASED MEDICAL ULTRASOUND IMAGE SEGMENTATION NETWORK]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/133162807</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/133162807</guid><author>Stark_Z</author><pubDate>Fri, 22 Sep 2023 11:33:51 +0800</pubDate><description><![CDATA[在这项工作中，提出了CMU-Net，这是一种用于医学超声分割的完全卷积网络。将ConvMixer块引入U形网络架构中，以构建用于获取全局上下文信息的强大编码器，并提出多尺度注意门模块，用于强调有价值的特征以实现高效的跳过连接。]]></description><category></category></item><item><title><![CDATA[论文笔记MALUNet: A Multi-Attention and Light-weight UNet for Skin Lesion Segmentation]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/133161272</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/133161272</guid><author>Stark_Z</author><pubDate>Fri, 22 Sep 2023 11:11:52 +0800</pubDate><description><![CDATA[本文提出了四个模块：（1）扩展门控注意块（DGA），它通过获得全局和局部信息来帮助我们的模型更加关注目标区域。同时，深度可分离卷积进一步减少了参数的数量和计算复杂度。(2)针对轻量级医学分割模型，提出了一种有效的外部注意机制，称为反向外部注意块（IEA），该机制可以增强样本之间的信息关联，获得整体数据集的特征。(3)针对多阶段多尺度信息的获取，提出了针对通道和空间轴的注意桥模块，称为通道注意桥模块（CAB）和空间注意桥模块（SAB），它们可以生成相应的注意图。]]></description><category></category></item><item><title><![CDATA[论文笔记CMUNeXt: An Efficient Medical Image Segmentation Network based on Large Kernel and Skip Fusion]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/132966130</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/132966130</guid><author>Stark_Z</author><pubDate>Mon, 18 Sep 2023 10:37:18 +0800</pubDate><description><![CDATA[CMUNeXt：一种基于大核和跳跃融合的高效医学图像分割网络，一个完全卷积的轻量级医学图像分割网络，遵循U形架构设计，由具有跳过连接的五级编码器-解码器结构组成。]]></description><category></category></item><item><title><![CDATA[论文笔记EGE-UNet: an Efficient Group Enhanced UNet for skin lesion segmentation]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/132556398</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/132556398</guid><author>Stark_Z</author><pubDate>Tue, 29 Aug 2023 11:15:53 +0800</pubDate><description><![CDATA[EGE-UNet：一种高效的群算法用于皮肤病变分割的增强型UNet算法。两个关键模块：组多轴Hadamard产品注意模块（GHPA）和组聚合桥模块（GAB）]]></description><category></category></item><item><title><![CDATA[论文笔记MultiTrans: Multi-scale feature fusion transformer with transfer learning strategy]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/132014954</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/132014954</guid><author>Stark_Z</author><pubDate>Mon, 31 Jul 2023 09:22:28 +0800</pubDate><description><![CDATA[为了辅助医生进行精确分割头颈部危及器官（OARs）分割，提出了一种具有多尺度特征融合模块的MultiTrans框架。此外，我们还提出了一种迁移学习策略，通过将数据量大的器官的权值参数作为数据量小的器官的预训练权值参数，有效地提高了模型对小器官的分割性能。]]></description><category></category></item><item><title><![CDATA[论文笔记TC-Net: A joint learning framework based on CNN and vision transformer for multi-lesion medical]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/131419511</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/131419511</guid><author>Stark_Z</author><pubDate>Tue, 27 Jun 2023 17:13:10 +0800</pubDate><description><![CDATA[提出了一个名为TC-Net的医学图像分割网络；设计了一个局部感知和长距离依赖连接策略（LLCS）聚合从两个子分支获得的特征图；同时提出了一个动态的循环局灶性损失（DCFL），以解决类不平衡的问题。]]></description><category></category></item><item><title><![CDATA[论文笔记A dual attention-guided 3D convolution network for automatic segmentation of prostate and tumor]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/131028662</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/131028662</guid><author>Stark_Z</author><pubDate>Sun, 04 Jun 2023 09:16:15 +0800</pubDate><description><![CDATA[提供了一个独特的双注意力机制3D分割网络架构，用于评估DWI上的前列腺和肿瘤；同时设计了视觉注意力以及多尺度注意力来更好的提取有效]]></description><category></category></item><item><title><![CDATA[论文笔记MRLA-Net: A tumor segmentation network embedded with a multiple receptive-field lesion attention]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/130840736</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/130840736</guid><author>Stark_Z</author><pubDate>Wed, 24 May 2023 10:31:10 +0800</pubDate><description><![CDATA[本文提出了一种将PET-CT诊断视野与病灶形态特征相结合的病灶注意力模块]]></description><category></category></item><item><title><![CDATA[论文笔记MSAA-Net: a multi-scale attention-aware U-Net is used to segment the liver]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/130371808</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/130371808</guid><author>Stark_Z</author><pubDate>Tue, 25 Apr 2023 20:07:28 +0800</pubDate><description><![CDATA[提出了一个使用多尺度注意力感知的U-Net肝脏分割网络，并设计了一个多尺度注意力门结构（MAG）以替代跳跃连接以此优化合并不同大小的对象。]]></description><category></category></item><item><title><![CDATA[论文笔记Dual Cross-Attention for Medical Image Segmentation]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/130222635</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/130222635</guid><author>Stark_Z</author><pubDate>Tue, 18 Apr 2023 15:54:23 +0800</pubDate><description><![CDATA[用于医学图像分割的双交叉注意力]]></description><category></category></item><item><title><![CDATA[论文笔记Enhancing Medical Image Segmentation with TransCeption: A Multi-Scale Feature Fusion Approach]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/130084528</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/130084528</guid><author>Stark_Z</author><pubDate>Tue, 11 Apr 2023 16:02:21 +0800</pubDate><description><![CDATA[使用TransCeption增强医学图像分割：一种多尺度特征融合方法]]></description><category></category></item><item><title><![CDATA[论文笔记Dense-PSP-UNet: A neural network for fast inference liver ultrasound segmentation]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/129944426</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/129944426</guid><author>Stark_Z</author><pubDate>Tue, 04 Apr 2023 10:00:00 +0800</pubDate><description><![CDATA[提出了一种神经网络架构，即Dense-PSP-UNet，以实现实时肝脏超声分割]]></description><category></category></item><item><title><![CDATA[论文笔记A deep residual attention-based U-Net with a biplane joint method for liver segmentation from CT]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/129929456</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/129929456</guid><author>Stark_Z</author><pubDate>Mon, 03 Apr 2023 14:43:02 +0800</pubDate><description><![CDATA[基于深度剩余注意的双平面联合U-Net在CT图像肝脏分割中的应用]]></description><category></category></item><item><title><![CDATA[论文笔记Fusing enhanced Transformer and large kernel CNN for malignant thyroid nodule segmentation]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/129928465</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/129928465</guid><author>Stark_Z</author><pubDate>Mon, 03 Apr 2023 14:42:15 +0800</pubDate><description><![CDATA[由于超声图像的复杂性和变异性，通过超声快速准确地诊断恶性甲状腺结节是一项有价值但具有挑战性的任务。提出的TCNet包含一个大核CNN分支和一个增强的Transformer分支。增强的Transformer分支由一个线性嵌入层（LE）和三个面片合并层（PM）组成，其中包含一些跳过连接操作。然后，经过并行、分层的特征提取后，将特征映射图送到解码器，解码器将特征映射图恢复到原始大小，并恢复到高分辨率。它们包括一个超大的内核（高达31 × 31）深度方向的Conv层，以捕获更多的结构信息（形状）。]]></description><category></category></item><item><title><![CDATA[论文笔记LiM-Net: Lightweight multi-level multiscale network with deep residual learning]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/129160503</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/129160503</guid><author>Stark_Z</author><pubDate>Wed, 22 Feb 2023 13:07:51 +0800</pubDate><description><![CDATA[新颖的轻量级的多层次多尺度方法从CT图像自动分割肝脏]]></description><category></category></item><item><title><![CDATA[论文笔记Efficient two-step liver and tumour segmentation on abdominal CT]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/129140500</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/129140500</guid><author>Stark_Z</author><pubDate>Tue, 21 Feb 2023 12:47:10 +0800</pubDate><description><![CDATA[两步分割网络FRAUNet，依次分割肝脏和肿瘤。]]></description><category></category></item><item><title><![CDATA[论文笔记CPAD-Net: Contextual parallel attention and dilated network for liver tumor segmentation]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/129130586</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/129130586</guid><author>Stark_Z</author><pubDate>Mon, 20 Feb 2023 20:01:34 +0800</pubDate><description><![CDATA[本文设计了一种变异的U型结构，即CPAD-Net，用于肝脏肿瘤的分割。在跳跃连接处采用CPAM块以提取上下文高级和低级特征。]]></description><category></category></item><item><title><![CDATA[论文笔记Practical Window Setting Optimization for Medical Image Deep Learning]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/126865793</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/126865793</guid><author>Stark_Z</author><pubDate>Thu, 15 Sep 2022 10:10:41 +0800</pubDate><description><![CDATA[医学图像深度学习的实用窗口设置优化]]></description><category></category></item><item><title><![CDATA[论文笔记United adversarial learning for liver tumor segmentation]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/126778541</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/126778541</guid><author>Stark_Z</author><pubDate>Fri, 09 Sep 2022 10:47:07 +0800</pubDate><description><![CDATA[联合对抗学习用于肝肿瘤分割和多模态非对比MRI检测]]></description><category></category></item><item><title><![CDATA[论文笔记Automatic liver segmentation from abdominal CT volumes using improved convolution neural network]]></title><link>https://blog.youkuaiyun.com/Stark_Z/article/details/126763240</link><guid>https://blog.youkuaiyun.com/Stark_Z/article/details/126763240</guid><author>Stark_Z</author><pubDate>Thu, 08 Sep 2022 14:01:00 +0800</pubDate><description><![CDATA[使用改进的卷积神经网络从腹部 CT 体积自动分割肝脏]]></description><category></category></item></channel></rss>