Voxel-Based Lesion-Symptom Mapping: A Nonparametric Bayesian Approach

Mame Diarra Fall,Emilie Lavau,Pascal Auzou

IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)（2018）CCF B

Cited 1|Views13

Abstract

The study of brain-injured patients (or lesion-based analysis) is a powerful paradigm for investigating structure-function relationships using neuroimaging techniques. Voxel-based Lesion-Symptom Mapping (VLSM) has been widely used to detect structure-function associations in neuroimaging studies. However this approach is based on Student t-test for which normality does not always hold. Our aim in the current study is twofold: 1) to confirm/refute the implication of the classical language areas using the Language Screening (LAST) test; and 2) to determine if it is possible to reduce the number of patients included in the VLSM study, using a different statistical approach. To achieve the second goal, we propose an alternative nonparametric and Bayesian test using Pólya trees. The approach is Bayesian, assigning prior distributions and computing the Bayes factor of H0 (null hypothesis) to H1 (alternative); and it is nonparametric since the priors are put on the unknown distribution functions under H0 and H1. Our results highlight that the Pólya tree prior provides a convenient and effective way for testing two sample differences in VLSM studies.

Translated text

Key words

Hypothesis testing,Nonparametric Bayes,Polya tree,VLSM,stroke,LAST test

求助PDF

上传PDF

Bibtex

AI Read Science

AI Summary

AI Summary is the key point extracted automatically understanding the full text of the paper, including the background, methods, results, conclusions, icons and other key content, so that you can get the outline of the paper at a glance.

Example

Background

Key content

Introduction

Methods

Results

Related work

Fund

Key content

Pretraining has recently greatly promoted the development of natural language processing (NLP)
We show that M6 outperforms the baselines in multimodal downstream tasks, and the large M6 with 10 parameters can reach a better performance
We propose a method called M6 that is able to process information of multiple modalities and perform both single-modal and cross-modal understanding and generation
The model is scaled to large model with 10 billion parameters with sophisticated deployment, and the 10 -parameter M6-large is the largest pretrained model in Chinese
Experimental results show that our proposed M6 outperforms the baseline in a number of downstream tasks concerning both single modality and multiple modalities We will continue the pretraining of extremely large models by increasing data to explore the limit of its performance

Upload PDF to Generate Summary

Must-Reading Tree

Example

Generate MRT to find the research sequence of this paper