AI-enabled in silico immunohistochemical characterization for Alzheimer's disease.
AI-enabled in silico immunohistochemical characterization for Alzheimer's disease.
复制标题
DOI：
10.1016/j.crmeth.2022.100191
复制
发表时间：
2022-04-25
期刊：
影响因子：
--
通讯作者：
中科院分区：
文献类型：
--
作者：
我们在硅免疫组织化学（IHC）中开发了一种深度学习方法，该方法将常规收集的组织化学染色样本作为输入，并通过计算生成虚拟的IHC幻灯片图像。我们将硅免疫组化应用于阿尔茨海默病样本，其中使用免疫组化染色在大脑的许多区域常规地识别出几个标志性的变化。在计算机上，IHC直接从组织化学图像中以高空间分辨率计算识别神经原纤维缠结、β-淀粉样斑块和神经性斑块，受体工作特征曲线下的面积在0.88到0.92之间。计算机免疫结构学会识别与这些病变相关的细微细胞形态，并可以生成捕捉实际免疫结构关键特征的计算机免疫结构幻灯片。基于深度学习的常规组织化学样本的计算处理常规需要免疫组织化学染色的特征识别应用于硅免疫组化识别阿尔茨海默病的标志性变化免疫组化（IHC）广泛应用于各种疾病，以识别感兴趣的蛋白质并帮助疾病诊断。然而，大规模制备IHC既耗时又昂贵。为了评估阿尔茨海默病（AD）的进展，必须通过免疫组化染色确定大脑许多区域的几个标志性变化。制备许多用免疫组化染色的阿尔茨海默病样品的成本一直是大型基因组-内表型研究的主要障碍。因此，阿尔茨海默病和其他疾病的研究和诊断将受益于无需免疫组化就能识别这些蛋白质的能力。他等人开发了一种基于硅IHC的深度学习方法，该方法使用常规组织化学染色来识别通常需要更昂贵的免疫组织化学染色的特征。该方法被应用于阿尔茨海默病，它确定了几种表征疾病进展的标志性变化。
We develop a deep learning approach, in silico immunohistochemistry (IHC), which takes routinely collected histochemical-stained samples as input and computationally generates virtual IHC slide images. We apply in silico IHC to Alzheimer's disease samples, where several hallmark changes are conventionally identified using IHC staining across many regions of the brain. In silico IHC computationally identifies neurofibrillary tangles, β-amyloid plaques, and neuritic plaques at a high spatial resolution directly from the histochemical images, with areas under the receiver operating characteristic curve of between 0.88 and 0.92. In silico IHC learns to identify subtle cellular morphologies associated with these lesions and can generate in silico IHC slides that capture key features of the actual IHC. Deep learning-based computational processing of routine histochemical samples Identification of features conventionally requiring immunohistochemical stains Apply in silico IHC to identify hallmark changes of Alzheimer's disease Immunohistochemistry (IHC) is used widely across diseases to identify proteins of interest and aid disease diagnosis. However, IHC is time consuming and expensive to prepare at a large scale. For evaluating the progression of Alzheimer's disease (AD), several hallmark changes must be identified via IHC staining across many regions of the brain. The cost of preparing many samples with IHC staining for AD has been a major barrier to large genomic-endophenotype studies. As a result, research and diagnosis of AD and other diseases would benefit from the ability to identify these proteins without the need for IHC. He et al. develop in silico IHC, a deep learning-based method that uses routine histochemical stains to identify features that conventionally require more expensive immunohistochemical stains. The method is applied to Alzheimer's disease, where it identifies several hallmark changes that characterize progression of the disease.