利用深度卷积神经网络规避形态学特征描述对微流图像中不溶性蛋白质聚集体进行分类
Using Deep Convolutional Neural Networks to Circumvent Morphological Feature Specification when Classifying Subvisible Protein Aggregates from Micro-Flow Images
摘要:
流动成像显微镜(Flow Imaging Microscopy,简称FCM)通常用于学术界和工业界,以表征蛋白质治疗剂中的不溶性颗粒(尺寸为$\le 25 μm$的颗粒)。制药公司被要求记录关于蛋白质治疗产品的大量数据,但仅根据美国FDA法规(即,USP $\big \langle 788 \big \rangle$),以控制交付产品中超过$10$和$25 μm$的颗粒数量。因此,大量的数字图像可用于分析。目前最先进的方法依赖于一个相对低维的“形态特征”来表征颗粒,但这些方法忽略了大量的信息编码在现有的大型数字图像库。深度卷积神经网络(CNN或“ConvNets”)已经证明了从原始宏观图像数据中提取预测信息的能力,而不需要在各种任务中选择或指定“形态特征”。然而,蛋白质治疗剂的异质性、多分散性以及与亚可见光粒子测量相关的光学现象,给CNN在图像分析中的应用带来了新的挑战。在这篇文章中,我们展示了一种监督学习技术,利用CNN从原始图像中提取信息,以预测过程条件或应力状态(冻融,机械振动等)。产生了各种不同的蛋白质图像。我们证明,我们的新分类器(结合样本“图像池”策略)可以获得几乎完美的预测,使用少至20个图像从一个给定的蛋白质配方在各种情况下的相关蛋白质治疗质量控制和过程监控。
英文摘要:
Flow-Imaging Microscopy (FIM) is commonly used in both academia and industry to characterize subvisible particles (those $\le 25 μm$ in size) in protein therapeutics. Pharmaceutical companies are required to record vast volumes of FIM data on protein therapeutic products, but are only mandated under US FDA regulations (i.e., USP $\big \langle 788 \big \rangle$) to control the number of particles exceeding $10$ and $25 μm$ in delivered products. Hence, a vast amount of digital images are available to analyze. Current state-of-the-art methods rely on a relatively low-dimensional list of "morphological features" to characterize particles, but these methods ignore an enormous amount of information encoded in the existing large digital image repositories. Deep Convolutional Neural Networks (CNNs or "ConvNets") have demonstrated the ability to extract predictive information from raw macroscopic image data without requiring the selection or specification of "morphological features" in a variety of tasks. However, the heterogeneity, polydispersity of protein therapeutics, and optical phenomena associated with subvisible FIM particle measurements introduce new challenges regarding the application of CNNs to FIM image analysis. In this article, we demonstrate a supervised learning technique leveraging CNNs to extract information from raw images in order to predict the process conditions or stress states (freeze-thaw, mechanical shaking, etc.) that produced a variety of different protein images. We demonstrate that our new classifier (in combination with a sample "image pooling" strategy) can obtain nearly perfect predictions using as few as 20 FIM images from a given protein formulation in a variety of scenarios of relevance to protein therapeutics quality control and process monitoring.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。