Transport transforms for biomedical data modeling, estimation, and classification

用于生物医学数据建模、估计和分类的传输转换

• 批准号: 10672626
• 负责人: Gustavo Kunde Rohde
• 金额: $ 35.51万
• 依托单位: UNIVERSITY OF VIRGINIA
• 依托单位国家: 美国
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-03-01 至 2027-06-30
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10672626
• 关键词: Address; Biological; Brain imaging; COVID-19 patient; Cells; Classification; Clinical Sciences; Collaborations; Communities; Complement; Complex; Computer Models; Computer software; Cytometry; Data; Data Analyses; Data Set; Development; Discipline; Documentation; Electronic Health Record; Energy consumption; Engineering; Ensure; Funding; Future; Gene Expression; Genome; Goals; Image; Imaging technology; Knee Osteoarthritis; Learning; Letters; Magnetic Resonance; Magnetic Resonance Imaging; Mathematics; Measurement; Measures; Methodology; Methods; Modeling; Molecular; Optics; Organ; Outcome; Pathology; Pattern; Pattern Recognition; Phase; Problem Solving; Risk; Sampling; Scientist; Series; Signal Transduction; Specific qualifier value; Speed; System; Techniques; Technology; Training; Visualization; absorption; biomedical data science; breast cancer diagnosis; cancer risk; chemical reaction; data mining; data modeling; data space; deep learning; deep learning model; electric impedance; experimental study; high dimensionality; imaging Segmentation; improved; innovation; learning strategy; lectures; machine learning model; mathematical model; microscopic imaging; morphometry; multidimensional data; neural network; predictive modeling; radiological imaging; radiomics; reconstruction; software development; technology research and development; tool; tumor progression; virtual; voltage

The goal of the project is to develop a new mathematical and computational modeling framework for from biomedical data extracted from biomedical experiments such as voltages, spectra (e.g. mass, magnetic resonance, impedance, optical absorption, …), microscopy or radiology images, gene expression, and many others. Scientists who are looking to understand relationships between different molecular and cellular measurements are often faced with questions involving deciphering differences between different cell or organ measurements. Current approaches (e.g. feature engineering and classification, end-to-end neural networks) are often viewed as “black boxes,” given their lack of connection to any biological mechanistic effects. The approach we propose builds from the “ground up” an entirely new modeling framework build based on recently developed invertible transformation. As such, it allows for any machine learning model to be represented in original data space, allowing for not only increased accuracy in prediction, but also direct visualization and interpretation. As an outcome of the previous funding period, our current approach outperforms other mathematical modeling tools when processing segmented signals and images by a wide margin in terms of accuracy, computational complexity, amount of training data needed, interpretability and robustness to out of distribution samples. In this current phase we seek to generalize the method beyond segmented images and signals to virtually any dataset type. We will explore proof of concept applications in cytometry, pathology, and radiomics.

该项目的目标是开发一种新的数学和计算方法 从生物医学中提取的生物医学数据的建模框架 实验，例如电压、光谱（例如质量、磁共振、 阻抗、光吸收……）、显微镜或放射学图像、基因 表达，以及许多其他寻求理解的科学家。 不同分子和细胞测量之间的关系通常是 面临涉及破译不同细胞之间差异或 当前的方法（例如特征工程和 分类、端到端神经网络）通常被视为“黑匣子”， 鉴于它们与任何生物机械效应缺乏联系。 我们建议从“头开始”构建一个全新的建模框架 基于最近开发的可逆变换构建。 在原始数据空间中表示的任何机器学习模型，允许 不仅提高了预测的准确性，而且还可以直接可视化和 作为上一个资助期的结果，我们当前的解释。 该方法在处理时优于其他数学建模工具 在准确度方面大幅分割信号和图像， 计算复杂度、所需训练数据量、可解释性和 在当前阶段，我们寻求对分布样本的鲁棒性。 将该方法推广到分割图像和信号之外的几乎任何领域 我们将探索细胞计数中的概念验证应用， 病理学和放射组学。

项目成果

Massive image-based single-cell profiling reveals high levels of circulating platelet aggregates in patients with COVID-19.

基于大规模图像的单细胞分析揭示了 COVID-19 患者循环血小板聚集水平较高。

• 发表时间: 2021-12-09
• 影响因子: 16.6
• 作者: Nishikawa, Masako;Kanno, Hiroshi;Zhou, Yuqi;Xiao, Ting;Suzuki, Takuma;Ibayashi, Yuma;Harmon, Jeffrey;Takizawa, Shigekazu;Hiramatsu, Kotaro;Nitta, Nao;Kameyama, Risako;Peterson, Walker;Takiguchi, Jun;Shifat;Zhuang, Yan
• 通讯作者: Zhuang, Yan

Neural Networks, Hypersurfaces, and the Generalized Radon Transform.

神经网络、超曲面和广义氡变换。

• 发表时间: 2020-07
• 影响因子: 14.9
• 作者: Kolouri, Soheil;Yin, Xuwang;Rohde, Gustavo K
• 通讯作者: Rohde, Gustavo K

Data-driven Identification of Parametric Governing Equations of Dynamical Systems Using the Signed Cumulative Distribution Transform.

使用有符号累积分布变换对动力系统参数控制方程进行数据驱动识别。

• DOI: 10.1016/j.cma.2024.116822
• 发表时间: 2023-08-23
• 期刊: Computer methods in applied mechanics and engineering
• 影响因子: 7.2
• 作者: A. Rubaiyat;D. H. Thai;J. Nichols;M. Hutchinson;S. Wallen;Christina J. Naify;Nathan Geib;M. Haberman;G. Rohde
• 通讯作者: G. Rohde

Invariance encoding in sliced-Wasserstein space for image classification with limited training data.

切片 Wasserstein 空间中的不变性编码，用于训练数据有限的图像分类。

• 发表时间: 2023-05
• 影响因子: 8
• 作者: Shifat;Zhuang, Yan;Li, Shiying;Rubaiyat, Abu Hasnat Mohammad;Yin, Xuwang;Rohde, Gustavo K
• 通讯作者: Rohde, Gustavo K

PARTITIONING SIGNAL CLASSES USING TRANSPORT TRANSFORMS FOR DATA ANALYSIS AND MACHINE LEARNING.

使用传输变换对信号类别进行分区以进行数据分析和机器学习。

• 发表时间: 2021-06
• 作者: Aldroubi, Akram;Li, Shiying;Rohde, Gustavo K
• 通讯作者: Rohde, Gustavo K