BundleMAP: Anatomically localized classification, regression, and hypothesis testing in diffusion MRI

PATTERN RECOGNITION

卷:63

Article

Khatami, Mohammad¹  Schmidt-Wilcke, Tobias²  Sundgren, Pia C.^3,4  Abbasloo, Amin¹  Schoelkopf, Bernhard⁵  Schultz, Thomas¹ 

[1] Univ Bonn, Dept Comp Sci, Friedrich Ebert Allee 144, D-53113 Bonn, Germany

[2] Univ Hosp Bergmannsheil, Dept Neurol, Buerkle De La Camp Pl 1, D-44789 Bochum, Germany

[3] Lund Univ, Inst Clin Sci Diagnost Radiol, SE-22185 Lund, Sweden

[4] Univ Michigan Hlth Syst, Dept Radiol, 1500 E Med Ctr Court, Ann Arbor, MI 48109 USA

[5] Max Planck Inst Intelligent Syst, Empir Inference Dept, Spemannstr 38, D-72026 Tubingen, Germany

关键词: Disease detection;    Manifold learning;    Support vector machines;    Classification;    Regression;    Fiber tracking;    Diffusion MRI;   

DOI  :  10.1016/j.patcog.2016.09.020

来源: Elsevier

【 摘 要 】

Diffusion MRI (dMRI) provides rich information on the white matter of the human brain, enabling insight into neurological disease, normal aging, and neuroplasticity. We present BundleMAP, an approach to extracting features from dMRI data that can be used for supervised classification, regression, and hypothesis testing. Our features are based on aggregating measurements along nerve fiber bundles, enabling visualization and anatomical interpretation. The main idea behind BundleMAP is to use the ISOMAP manifold learning technique to jointly parametrize nerve fiber bundles. We combine this idea with mechanisms for outlier removal and feature selection to obtain a practical machine learning pipeline. We demonstrate that it increases accuracy of disease detection and estimation of disease activity, and that it improves the power of statistical tests. (C) 2016 The Authors. Published by Elsevier Ltd.

【 授权许可】

Free