Scarcity of annotated images hampers the building of automated solution for reliable COVID-19 diagnosis and evaluation from CT. To alleviate the burden of data annotation, we herein present a label-free approach for segmenting COVID-19 lesions in CT via pixel-level anomaly modeling that mines out the relevant knowledge from normal CT lung scans. Our modeling is inspired by the observation that the parts of tracheae and vessels, which lay in the high-intensity range where lesions belong to, exhibit strong patterns. To facilitate the learning of such patterns at a pixel level, we synthesize `lesions' using a set of surprisingly simple operations and insert the synthesized `lesions' into normal CT lung scans to form training pairs, from which we learn a normalcy-converting network (NormNet) that turns an 'abnormal' image back to normal. Our experiments on three different datasets validate the effectiveness of NormNet, which conspicuously outperforms a variety of unsupervised anomaly detection (UAD) methods.
翻译:为减轻数据说明的负担,我们在此提出一种无标签的方法,通过像素级异常模型将CT的COVID-19损伤进行分解,这种模型将正常的CT肺扫描中的相关知识埋掉。我们的模型的灵感来自一种观察,即位于腐蚀所在高强度范围内的气管和船只部分呈现出强烈的形态。为了便利在像素水平上了解这种形态,我们用一套出乎意料的简单操作合成“遗漏”并将合成的“遗漏”插入正常的CT肺扫描以形成一对培训,从中我们学习一个正常的反转网络(NormNet),将“异常”图像恢复正常。我们在三个不同的数据集上进行的实验证实了NormNet的有效性,它明显地超越了各种未经监督的异常检测方法。