RT Journal Article
T1 Statistical Agnostic Mapping: A framework in neuroimaging based on concentration inequalities.
A1 Górriz-Sáez, Juan Manuel
A1 Jiménez-Mesa, Carmen
A1 Romero-García, Raúl
A1 Segovia, Fermín
A1 Ramírez, Javier
A1 Castillo-Barnes, Diego
A1 Martínez-Murcia, Francisco Jesús
A1 Ortiz-García, Andrés
A1 Salas-González, Diego
A1 Álvarez-Illán, Ignacio
A1 Puntonet, Carlos
A1 López-García, Diego
A1 Gómez-Río, Manuel
A1 Suckling, John
K1 Medicina - Proceso de datos
K1 Estadística médica
K1 Complejidad computacional
K1 Comprobación de hipótesis (Estadística)
AB In the 1970s a novel branch of statistics emerged focusing its effort on the selection of a function for thepattern recognition problem that would fulfill a relationship between the quality of the approximation and itscomplexity. This theory is mainly devoted to problems of estimating dependencies in the case of limited samplesizes, and comprise all the empirical out-of sample generalization approaches; e.g. cross validation (CV). In thispaper a data-driven approach based on concentration inequalities is designed for testing competing hypothesisor comparing different models. In this sense we derive a Statistical Agnostic (non-parametric) Mapping (SAM)for neuroimages at voxel or regional levels which is able to: (i) relieve the problem of instability with limitedsample sizes when estimating the actual risk via CV; and (ii) provide an alternative way of Family-wiseerror (FWE) corrected 𝑝-value maps in inferential statistics for hypothesis testing. Using several neuroimagingdatasets (containing large and small effects) and random task group analyses to compute empirical familywiseerror rates, this novel framework resulted in a model validation method for small samples over dimensionratios, and a less-conservative procedure than FWE 𝑝-value correction to determine the significance mapsfrom the inferences made using small upper bounds of the actual risk.
PB Elsevier
YR 2020
FD 2020-09-26
LK https://hdl.handle.net/10630/28106
UL https://hdl.handle.net/10630/28106
LA eng
NO Gorriz, Juan & Group, SiPBA & neuroscience, CAM. (2019). Statistical Agnostic Mapping: a Framework in Neuroimaging based on Concentration Inequalities. https://doi.org/10.1016/j.inffus.2020.09.008
DS RIUMA. Repositorio Institucional de la Universidad de Málaga
RD 19 ene 2026