arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2410.18110q-bio.NC
收藏

从 fMRI 数据学习图像衍生的 PDE 表型

Learning Image Derived PDE-Phenotypes from fMRI Data

Ion Bica, Ryan Trang, Rui Hu, Wanhua Su, Zhichun Zhai, Qingrun Zhang

AI总结:

本文提出结合 CanICA、UMAP 与稀疏岭回归,从 fMRI 数据识别 PDE 并提取图像衍生表型,在 ADHD200 上实现高精度 ADHD 分类。

摘要:

偏微分方程(PDE)可对电磁场和流体力学等多种物理现象进行建模。Sparse Identification of Nonlinear Dynamics(SINDy,非线性动力学稀疏识别)和 PDE-Net 2.0 等方法已被开发出来,分别利用稀疏优化和深度神经网络,基于数据识别并建模 PDE。尽管 PDE 模型较少应用于 fMRI 数据,但它们具有揭示脑活动中隐藏联系和基本成分的潜力。研究使用 ADHD200 数据集,采用 Canonical Independent Component Analysis(CanICA,典型独立成分分析)和 Uniform Manifold Approximation(UMAP,统一流形逼近与投影)对 fMRI 数据进行降维。随后,研究使用 Sparse Ridge Regression(稀疏岭回归)从降维后的数据中识别 PDE,并在注意缺陷多动障碍(ADHD)分类中取得较高准确率。该研究展示了一种从 fMRI 数据中提取有意义特征以分析神经系统疾病的新方法,用于理解神经活动期间脑内氧转运(输送与消耗)的作用,这对研究颅内病变具有重要意义。

英文摘要:

Partial Differential Equations (PDEs) model various physical phenomena, such as electromagnetic fields and fluid mechanics. Methods like Sparse Identification of Nonlinear Dynamics (SINDy) and PDE-Net 2.0 have been developed to identify and model PDEs based on data using sparse optimization and deep neural networks, respectively. While PDE models are less commonly applied to fMRI data, they hold the potential for uncovering hidden connections and essential components in brain activity. Using the ADHD200 dataset, we applied Canonical Independent Component Analysis (CanICA) and Uniform Manifold Approximation (UMAP) for dimensionality reduction of fMRI data. We then used Sparse Ridge Regression to identify PDEs from the reduced data, achieving high accuracy in classifying attention deficit hyperactivity disorder (ADHD). The study demonstrates a novel approach to extracting meaningful features from fMRI data for neurological disorder analysis to understand the role of oxygen transport (delivery $\&$ consumption) in the brain during neural activity relevant for studying intracranial pathologies.

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。

↑