Regularized Hypothesis Testing in Random Fields with Applications to Neuroimaging.
The task of determining for which elements of a random field (e.g., pixels in an image) a certain null hypothesis may be rejected is a relevant problem in several scientific areas. In the current contribution, we introduce a new method for performing this task, the regularized hypothesis testing (RH...
| Publicado en: | Revista Mexicana de Ingeniería Biomédica Vol. 41; no. 2; pp. 22 - 40 |
|---|---|
| Autores principales: | , , |
| Formato: | Artículo |
| Publicado: |
Sociedad Mexicana de Ingenieria Biomedica, A.C.
May-Ago2020
|
| Materias: | |
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=lth&AN=145969656&site=ehost-live header: @attributes: shortDbName: lth uiTerm: 145969656 longDbName: MedicLatina uiTag: AN controlInfo: bkinfo: jinfo: jid: 01889532 7L3 jtl: Revista Mexicana de Ingeniería Biomédica issn: 01889532 maglogo: N pubinfo: dt: May-Ago2020 vid: 41 iid: 2 pid: 20850 pub: Sociedad Mexicana de Ingenieria Biomedica, A.C. artinfo: ui: 145969656 10.17488/RMIB.41.2.2 ppf: 22 ppct: 18 formats: fmt: @attributes: type: P size: 2.5MB tig: atl: Regularized Hypothesis Testing in Random Fields with Applications to Neuroimaging. aug: au: Dalmau-Cedeño, Oscar S. Alvarado-Carrillo, Dora E. Luis Marroquín, José affil: Centro de Investigación en Matemáticas, CIMAT A.C. su: Statistical hypothesis testing Functional magnetic resonance imaging Noise Brain imaging Bayesian analysis sug: subj: Statistical hypothesis testing Functional magnetic resonance imaging Noise Brain imaging Bayesian analysis keyword: Bayesian estimation campo aleatorio Markoviano estimación Bayesiana functional Magnetic resonance imaging Imágenes de Resonancia Magnética Funcional Markovian random fields Prueba de hipótesis regularizada Regularized hypothesis test campo aleatorio Markoviano estimación Bayesiana Imágenes de Resonancia Magnética Funcional Prueba de hipótesis regularizada ab: The task of determining for which elements of a random field (e.g., pixels in an image) a certain null hypothesis may be rejected is a relevant problem in several scientific areas. In the current contribution, we introduce a new method for performing this task, the regularized hypothesis testing (RHT) method, focusing on its use in neuroimaging research. RHT is based on the formulation of the hypothesis testing task as a Bayesian estimation problem, with the previous application of a Markovian random field. The latter allows for the incorporation of local spatial information and considers different noise models, including spatially correlated noise. In tests on synthetic data showing regular activation levels on uncorrelated noise fields, RHT furnished a true positive rate (TPR) of 0.97, overcoming the state-of-the-art morphology-based hypothesis testing (MBHT) method and the traditional family-wise error rate (FWER) method, which afforded 0.93 and 0.58, respectively. For fields with highly correlated noise, the TPR provided by RHT was 0.65, and by MBHT and FWER was 0.35 and 0.29, respectively. For tests utilizing real functional magnetic resonance imaging (fMRI) data, RHT managed to locate the activation regions when 60% of the original signal were removed, while MBHT located only one region and FWER located none. En varias áreas científicas aparece el problema de determinar los elementos de un campo aleatorio (por ejemplo, píxeles en una imagen) en los que se puede rechazar una cierta hipótesis nula. En este artículo presentamos un nuevo método para realizar esta tarea, centrado en aplicaciones para investigación de neuroimagen. Nuestra propuesta se basa en la formulación de pruebas de hipótesis como un problema de estimación Bayesiana, usando como a priori un campo aleatorio Markoviano, que permite incorporar información espacial local y considera diferentes modelos de ruido, incluido el ruido correlacionado espacialmente. Para pruebas en datos sintéticos con niveles de activación regulares sobre campos de ruido no correlacionado, nuestro método obtiene una tasa de verdaderos positivos (TPR) de 0.97, superando al método del estado del arte MBHT y al método de control FWER que obtienen 0.93 y 0.58 respectivamente; para campos con ruido altamente correlacionado, nuestro método obtiene un TPR de 0.65, mientras que MBHT y FWER obtienen 0.35 y 0.29 respectivamente. Para pruebas con datos reales de fMRI, nuestro método localiza las regiones de activación cuando removemos 60% de la señal original, mientras que MBHT no localiza región alguna y FWER localiza una de las dos regiones. pubtype: Academic Journal doctype: Article src: R language: English refInfo: copyright: @attributes: flag: Y custom: Copyright of Revista Mexicana de Ingeniería Biomédica is the property of Sociedad Mexicana de Ingenieria Biomedica, A.C. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. item: Revista Mexicana de Ingeniería Biomédica holder: Sociedad Mexicana de Ingenieria Biomedica, A.C. dt: @attributes: year: 2020 holdings: @attributes: islocal: N |
|---|