Deep Learning of Mass Spectrometry Imaging

质谱成像的深度学习

• 批准号: 10743626
• 负责人: Drew R Jones
• 金额: $ 43.58万
• 依托单位: NEW YORK UNIVERSITY SCHOOL OF MEDICINE
• 依托单位国家: 美国
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-09-07 至 2025-08-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10743626
• 关键词: Address; Algorithms; Antibody Specificity; Architecture; Awareness; Back; Cells; Classification; Color; Complex; Computer software; Cryoultramicrotomy; Data; Data Set; Detection; Dimensions; Discrimination; Elasticity; Endometrial Carcinoma; Equilibrium; Freezing; Goals; Histologic; Histopathology; Image; Image Analysis; Immunohistochemistry; Individual; Ions; Laboratories; Learning; Lipids; Lung Adenocarcinoma; Machine Learning; Malignant Neoplasms; Mass Spectrum Analysis; Measures; Medicine; Modality; Modeling; Molecular; Names; Neural Network Simulation; Optics; Pathologist; Pathology; Pattern; Peptides; Performance; Polysaccharides; Preparation; Publishing; Reporting; Research Personnel; Resolution; Sampling; Shapes; Stains; Structure; Techniques; Technology; Testing; Time; Tissue Stains; Tissues; Training; Vendor; Work; anticancer research; cancer diagnosis; cancer subtypes; convolutional neural network; data acquisition; data structure; deep learning; design; digital imaging; high dimensionality; human imaging; imaging detection; imaging modality; improved; informatics tool; interest; ion mobility; learning strategy; machine learning model; mass spectrometric imaging; molecular subtypes; multidimensional data; n-dimensional; neural network architecture; novel; optical imaging; people of color; protein biomarkers; tool; translational cancer research; two-dimensional

PROJECT SUMMARY Mass spectrometry imaging (MSI) is a rapidly developing technology which gives pathologists many new types of targets (e.g., metabolites and lipids) to assess for translational cancer research. However, the resulting data are even more complex than traditional images because they are highly-dimensional, and large (~100GB per tissue section). Each “pixel” in the resulting data structure contains a 2-dimensional mass spectrum made of both measured ion mass and ion mobility (m/z, 1/K0), and each spectrum typically contains hundreds to thousands of individual ions (metabolites and lipids). Deep-learning methods (machine learning) have been successfully applied to histopathology data by several laboratories including Dr. David Fenyo, Co-Investigator of the current proposal, with such models being able to discriminate between different cancer subtypes and grades for example. However, most machine learning models of image-data are designed around 3-data channels (Red, Green, Blue) for analysis of digital images. Therefore, the n-dimensional data structure of mass spectrometry imaging datasets is not easily amenable to these proven machine learning workflows. We will make MSI data accessible to these approaches by expanding to n-dimension “color-channels”, with each unique metabolite or lipid image serving as an individual data input. For the deep learning component, we will retain the same overall architecture and workflow of the Panoptes tool, published by Fenyo et. al., (Cell Reports, Medicine, 2021) but we will apply an n-dimensional approach and test the data structure on existing data which has parallel H&E stain information annotated by pathologists. These challenges are addressed in Aim1 of the current proposal, while Aim 2 addresses a closely related challenge of detecting image correlations both within and between these data structures and other imaging modalities. Image correlations within such data are more trivial, but these analyses are not well supported by existing academic or vendor software due to the amount of computation needed for hundreds of data dimensions. We further propose and test an approach for converting these multi- dimensional data into centroided single ion images, followed by linearization of the image to enable a simple Pearson correlation metric, thereby making a complete correlation matrix accessible by a scaling factor of n2 to the number of detected ions. Secondly, to deal with spatial correlations between MSI datasets and images from other modalities, or adjacent tissue sections which may be different in size and shape, we propose to implement a spatially aware elastic transform of the centroided image data prior to correlation analysis and machine learning.

项目摘要 质谱成像（MSI）是一种快速发展的技术，它为病理学家提供了许多新的类型 目标（例如，代谢物和脂质）以评估转化癌症研究。然而，由此产生的数据 甚至比传统图像更复杂，因为它们是高维的，而且很大（每个大约100 GB 组织切片）。结果数据结构中的每个“像素”都包含一个二维质谱， 测量的离子质量和离子迁移率（m/z，1/K 0），并且每个光谱通常包含数百至 成千上万的单个离子（代谢物和脂质）。深度学习方法（机器学习）已经被 成功应用于组织病理学数据的几个实验室，包括大卫Fenyo博士，合作研究者 目前的提议，这种模型能够区分不同的癌症亚型和等级， 比如然而，大多数图像数据的机器学习模型都是围绕3个数据通道设计的（Red， 绿色、蓝色）用于分析数字图像。因此，质谱的n维数据结构 成像数据集不容易服从这些经验证的机器学习工作流。我们将使MSI数据 通过扩展到n维“颜色通道”，这些方法可以访问每个独特的代谢物或 脂质图像用作个体数据输入。对于深度学习组件，我们将保留相同的整体 Panoptes工具的架构和工作流程，由Fenyo et.例如，(Cell报告，医学，2021），但 我们将应用n维方法，并在具有并行H&E的现有数据上测试数据结构 病理学家注释的染色信息。这些挑战在本提案的目标1中得到解决， 而目标2解决了检测这些图像内部和之间的图像相关性的密切相关的挑战， 数据结构和其他成像模式。这些数据中的图像相关性更微不足道，但这些 由于计算量大，现有的学术或供应商软件不能很好地支持分析 需要数百个数据维度。我们进一步提出并测试了一种方法，用于转换这些多- 将三维数据转换为质心单离子图像，然后对图像进行线性化，以实现简单的 Pearson相关度量，从而使完整的相关矩阵可通过比例因子n2访问， 检测到的离子数。其次，为了处理MSI数据集和来自 我们建议实施其它模态或尺寸和形状可能不同的相邻组织切片， 在相关性分析和机器处理之前对质心图像数据进行空间感知弹性变换 学习