New Statistical Methods for Medical Signals and Images

描述（由申请人提供）：医学和生物数据通常以数字化信号和图像的形式出现；例如，基因表达微阵列，质谱，流式细胞术细胞图。随着仪器数据采集成为常规，这些图像、信号或路径的序列通常与其他协变量测量一起被收集，导致数据集的基本测量单位或响应是一个非常高维的对象。基因微阵列是新技术如何导致大规模数据采集的一个主要例子。该项目继续专注于开发建模和理解这些自然适应高维的数据的技术。为了利用比较基因组杂交研究细菌菌株的基因组差异，我们提出了包含一种称为“融合套索”的统计方法的潜在变量模型，以联合模拟细菌的CGH测量。对于癌细胞的流式细胞术分析，我们提出了一种方法来识别刺激细胞后出现的新亚群。我们还建议开发和研究高维数据的预测和聚类技术。这项工作的大部分将在现有的和新的与医学和生物学研究人员的合作中进行，例如在癌症和自身免疫疾病方面。项目描述：这项工作有可能提高对癌症、心脏病和艾滋病等人类疾病的理解、诊断和预后，从而有助于提高美国公共卫生的整体质量

DESCRIPTION (provided by applicant): Medical and biological data often come in the form of digitized signals and images; for example, gene expression microarrays, mass spectrograms, and flow cytometry cell plots. As instrumental data acquisition becomes routine, sequences of such images, signals or paths are collected, often along with other covariate measurements, resulting in datasets where the basic unit of measurement, or response, is a very high-dimensional object. The gene microarray is a leading example of how new technology has led to data acquisition on a massive scale. The project continues to focus on developing techniques for modeling and understanding such data that naturally adapt to the high dimensionality. For studying genomic divergence of bacterial strains using comparative genomic hybridization, we propose latent variable models that incorporate a statistical method called the "fused lasso", to jointly model the CGH measurements from the bacteria. For flow cytometry analysis of cancer cells, we propose a method for identifying new sub-populations that have emerged after stimulation of the cells. We also propose to develop and study techniques for prediction and clustering for high-dimensional data. Much of this work will be carried out in existing and new collaborations with researchers in medicine and biology, working for example in cancer and auto-immune diseases. Project Narrative: This work can potentially improve the understanding, diagnosis and prognosis of human diseases such as cancer, heart disease and AIDS, and hence can help to improve the overall quality of public health of the U.S.