宫颈癌医学图像分析--宫颈图像与组织玻片
<h2 style="color: black; text-align: left; margin-bottom: 10px;">宫颈癌<span style="color: black;">关联</span>任务</h2>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">针对</span>宫颈癌，有<span style="color: black;">有些</span>基于宫颈图像的任务，<span style="color: black;">亦</span>有在宫颈切片上的<span style="color: black;">关联</span>工作。</p>
<h2 style="color: black; text-align: left; margin-bottom: 10px;">基于宫颈图像的任务</h2>
<h3 style="color: black; text-align: left; margin-bottom: 10px;">子宫颈非典型增生<span style="color: black;">归类</span></h3>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">文案</span>：Multimodal Deep Learning for Cervical Dysplasia Diagnosis, MICCAI 2016.</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">术语：</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">Cervical intraepithelial neoplasia (CIN) ：宫颈上皮内瘤变</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">Cervical Dysplasia：子宫颈非典型增生</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">Screening：筛查</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">综述：这是该<span style="color: black;">行业</span>深度学习<span style="color: black;">第1</span>篇<span style="color: black;">文案</span>，<span style="color: black;">科研</span>的问题是<span style="color: black;">运用</span>宫颈图像和额外信息（HPV和Pap检验结果等）进行多模态信息融合并<span style="color: black;">归类</span>。几种类别<span style="color: black;">包含</span>：Normal（正常）、CIN1 (mild，轻度)、CIN2 (moderate，中等)和CIN3 (severe，重度)。CIN2/3 (CIN2+)，即癌症，<span style="color: black;">必须</span>对应的治疗。<span style="color: black;">这儿</span><span style="color: black;">重点</span>的<span style="color: black;">办法</span>是两路特征分别处理，<span style="color: black;">最后</span>拼接，再采用全连接的方式完成信息融合。</p>
<div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://pic2.zhimg.com/80/v2-3896ad10a499bbc0f9f62995572a3959_720w.webp" style="width: 50%; margin-bottom: 20px;"></div>图1 多模态子宫颈非典型增生类型诊断<h3 style="color: black; text-align: left; margin-bottom: 10px;">基于碘和醋酸图像和<span style="color: black;">目的</span>检测的<span style="color: black;">办法</span></h3>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">文案</span>：Multi-modal Fusion Learning for Cervical Dysplasia Diagnosis. IEEE International Symposium on Biomedical Imaging (ISBI), 2019.</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">综述：这是浙大课题组的工作，<span style="color: black;">经过</span>前期特征融合模块融合两个网络（分别对应不同模态）在不同尺度的特征，最后将融合后的图像特征直接和非图像特征拼接，再进行<span style="color: black;">归类</span>。</p>
<div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://pic1.zhimg.com/80/v2-53cc9e5f91f5aa8fbcd877141ed55c34_720w.webp" style="width: 50%; margin-bottom: 20px;"></div>图2 基于<span style="color: black;">目的</span>检测丁文后的双路fusion<span style="color: black;">办法</span>
<h3 style="color: black; text-align: left; margin-bottom: 10px;">基于单向双向特征融合的<span style="color: black;">办法</span></h3>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">文案</span>：Multi-view Learning with Feature Level Fusion for Cervical Dysplasia Diagnosis, MICCAI 2019.</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">综述：这是吴健老师课题组的工作，仍然是基于碘和醋酸宫颈图像。<span style="color: black;">经过</span>类似于构建特征金字塔的方式，在每一个尺度对特征进行融合。二种融合方式<span style="color: black;">包含</span>一主一辅的单项融合，以及二个模态对等的双向融合。在<span style="color: black;">最后</span>的评测中两种<span style="color: black;">办法</span>各有千秋。</p>
你的话语如春风拂面，让我心生暖意。 一看到楼主的气势，我就觉得楼主同在社区里灌水。 外链发布论坛学习网络优化SEO。
页:
[1]