文章2：全文IdentifyingandPredictingAutismSpectrumDisorderBasedonMulti-SiteStructuralMRIWithMachineLearning.pdf资源-CSDN文库

需积分: 5 153 浏览量 2022-04-23 09:48:59 上传评论收藏 1.99MB PDF 举报

资源详情

资源评论

资源推荐

ORIGINAL RESEARCH

published: 22 February 2022

doi: 10.3389/fnhum.2021.765517

Frontiers in Human Neuroscience | www.frontiersin.org 1 February 2022 | Volume 1 5 | Article 765517

Edited by:

Miseon Shim,

Korea University, South Korea

Reviewed by:

Xun Yang,

Chongqing University, China

Weihao Zheng,

Lanzhou University, China

*Correspondence:

Chang Liu

liuchang@cdu.edu.cn

DeZhong Yao

dyao@uestc.edu.cn

†

These authors have contributed

equally to this work

Specialty section:

This article was submitted to

Brain Imaging and Stimulation,

a section of the journal

Frontiers in Human Neuroscience

Received: 27 August 2021

Accepted: 13 December 2021

Published: 22 February 2022

Citation:

Duan Y, Zhao W, Luo C, Liu X,

Jiang H, Tang Y, Liu C and Yao D

(2022) Identifying and Predicting

Autism Spectrum Disorder Based on

Multi-Site Structural MRI With

Machine Learning.

Front. Hum. Neurosci. 15:765517.

doi: 10.3389/fnhum.2021.765517

Identifying and Predicting Autism

Spectrum Disorder B ased on

Multi-Site Structural MRI With

Machine Learning

YuMei Duan

1†

, WeiDong Zhao

2†

, Cheng Luo

3†

, XiaoJu Liu

, Hong Jiang

5†

, YiQian Tang

Chang Liu

2,3

and DeZhong Yao

Department of Computer and Software, Chengdu Jincheng College, Chengdu, China,

College of Computer, Chengdu

University, Chengdu, China,

The Key Laboratory for Neuro Information of Ministry of Education, Center for Information in Bio

Medicine, High-Field Magnetic Resonance Brain Imaging Key Laboratory of Sichuan Province, School of Life Science and

Technology, University of Electronic Science and Technology of China, Chengdu, China,

Department of Abdominal

Oncology, Cancer Center, West China Hospital, Sichuan University, Chengdu, China,

Department of Neurosurgery, Rui-Jin

Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China

Although emerging evidence has implicated structural/functional abnormalities of

patients with Autism Spectrum Disorder(ASD), deﬁnitive neuroimaging mar kers remain

obscured due to inconsistent or incompatible ﬁndings, especially for structural imaging.

Furthermore, brain differences deﬁned by statistical analy sis are difﬁcult to implement

individual prediction. The present study has employed the machine learning techniques

under the uniﬁed framework in neuroimaging to identify the neuroimaging markers of

patients with ASD and distinguish them from typically developing controls(TDC). To

enhance the interpretability of the machine learning model, the study has processed three

levels of assessments including model-level assessment, feature-level assessment, and

biology-level assessment. According to these three levels assessment, the study has

identiﬁed neuroimaging markers of ASD including the opercular part of bilateral inferior

frontal gyrus, the orbital part of right inferior frontal gyrus, right rolandic operculum,

right olfactory cortex, right gyrus rectus, right insula, left inferior parietal gyrus, bilateral

supramarginal gyrus, bilateral angular gy rus, bilateral superior temporal gyrus, bilateral

middle temporal gyrus, and left inferior temporal gyrus. In addition, negative correlations

between the communication skill score in the Autism Diagnostic Observation Schedule

(ADOS_G) and regional gray matter (GM) volume in the gyrus rectus, left middle

temporal gyrus, and inferior temporal gyrus have been detected. A signiﬁcant negative

correlation has been found between the communication skill score in ADOS_G and the

orbital part of the left inferior frontal gyrus. A negative correlation between verbal skill

score and right angular gyrus and a signiﬁcant negative correlation between non-verbal

communication skill and right angular gyrus have been found. These ﬁndings in the study

have suggested the GM alte ra tion of ASD and correlated with the c linica l severity of

ASD disease symptoms. The interpretable machine learning framework gives sight to

the pathophysiological mechanism of ASD but can also be extended to other diseases.

Keywords: autism spectrum disorder, structural MRI, multi-site data, machine learning, searchlight technique

Duan et al. ASD With Machine Learning

1. INTRODUCTION

Autism Spectrum Disorder, known as ASD, is a complex

neuro-developmental disorder and has been characterized by a

series of symptoms including early-onset diﬃculties in social

communication as well as restricted, repetitive behaviors and

interests (

Pagnozzi et al., 2018). The symptoms of ASD generally

occur within the ﬁrst 3 years of life and tend to last even

one’s whole life (

Hazlett et al., 2017). ASD brings signiﬁcant

impairments on an individual’s language, emotions, behavior,

self-control, learning, and memory and also is accompanied

by intellectual disability. Moreover, it is reported that patients

with ASD are far more likely to encounter premature death

than healthy controls (

Hirvikoski et al., 2015). According to

the Morbidity and Mortality Weekly Report (MMWR) Series

published by the Centers for Disease Controls and Prevention

(CDC) in the United States, the prevalence of ASD among

children has increased from 1 in 150 to 1 in 54 over 16 years (from

2000 to 2016) and the incidence rate of ASD was 4.3 times higher

in boys t han girls (Maenner et al., 2020) in 2016. For each patient

with ASD, the average lifetime social cost is approximately $3.6

million (Cakir et al., 2020).

Actually, if ASD is unable to be detected and intervened

at an earlier age, the impairments are irreversible. Therefore,

early and accurate identiﬁcation and diagnosis are crucial to

improving the life quality of ASD patients and their families.

Unfortunately, it is notoriously diﬃcult to diagnose, especially

in children, since the cause of ASD is a result of combined

factors, including genetics, the structure and function of the

brain, as well as environmental inﬂuences (

Raki

c et al., 2020).

Until now, there are still no eﬀective medical treatments for ASD.

For the current practice guidelines to assess, diagnose and treat

ASD, it is recommended to use the behavioral observation of

symptomology following the Diagnostic and Statistical Manual

(Fifth Edition) (DSM-5) (American Psychiatric Association,

2013) symptom criteria and the Internat ional Classiﬁcation of

Mental and Behavioral Disorders (Tenth Edition) (ICD-10)

(Organization, 1993). However, uniformity is lacking while using

these practice guidelines, so it is probably prone to misdiagnosis

(

Eslami et al., 2021). Furthermore, these guidelines cannot point

out the biological bases related to behavioral symptoms due to

unclear neuroanatomy. Finally, these limitations have resulted in

calls for more optimal diagnostic approaches for ASD.

In the last few decades, advances in non-invasive

neuroimaging techniques and analysis have provided crucial

knowledge to uncover patterns of brain structure and function

that would be symptomatic for the autism spectrum. The vast

majority of statistical methods on Structural MRI have intended

to explore the common patterns between patients with ASD

and healthy groups, but previous volumetric and morphological

analysis on structural MRI often has derived contradicted

results. For example, some research work reported decreased

volumes of the amygdala for ASD (van Rooij Daan et al., 2017)

while others did not ﬁnd signiﬁcant alterations (Maier et al.,

2015

). Focusing on the hippocampus volumes, some reported

its reduction, others reported its enlargement or no changes

(

Barnea-Goraly et al., 2014; Maier et al., 2015). Xiao et al.

(2014) has found that both gray and white matter (WM) has a

signiﬁcant increment with ASD, and Hazlett e t al. (201 7) has

pointed out brain volume overgrowth is related to the emergence

and severity of ASD. While Palmen et al. (2005) and Jou et al.

(2011)

have noted that t here is no diﬀerence or decreased

WM volume between ASD and healthy controls, a nd Riddle

et al. (2016) conducted voxel-based morphometry analysis and

found that the total brain volume and t he left anterior superior

temporal gyrus increased for children aged 2–4 with ASD. But

these brain structural abnormalities are subtle at later ages

(Riedel et al., 2014). These inconsistent ﬁndings are most likely

due to diﬀerent collecting approaches and limited sample size

with heterogeneous characteristics of subjects (Riddle et al.,

2016). Moreover, traditional statistical analysis is based on mass

univariate techniques which process a single voxel independently

and ignore the relat ionship between voxels (Bonnici et a l., 2012;

Samartsidis et al., 2016

). Furthermore, it deﬁnes the common

pattern at t he level of groups and is unable to predict t he

unknown sample at the level of individuals (Zhutovsky et al.,

2019; Hu et al., 2021).

Most recently, the rapid advance of machine learning has

made it becomes possible to explore the underlying neural

mechanisms and provide accurate predictions and convincing

explanations for ASD from various aspects (Khodatars et al.,

2020; Eslami et al., 2021). Knutson (2013) has pointed out that

machine learning can detect diﬀerences in neuroimaging data

that might not be detected with traditional univariate analysis.

In previ ous studies, typica l statistical machine learning and

deep le arning have been utilized to identify ASD from NC in

terms of structural and functional alt erations. Statistical machine

learning requires the design of handmade features (feature

extraction/feature selection) and implement the identiﬁcation of

patients with ASD based on these features (feature classiﬁcation).

Ecker et al. (2010) has applied SVM to investigate the whole-brain

diﬀerences of GM and WM volume on 44 subjects and obtained

signiﬁcant predictive power. Additionally, it has been found that

these brain diﬀerences are related to symptom severity. Ecker

et al. (2010) and Wee et al. (2014) have extracted morphological

features based on structural images and used SVM or multi-

kernel technique to achieve satisfactory results. Furthermore,

Zheng et al. (2018) have constructed a multi-feature-based

network based on morphological features to explore the cortico-

cortical similarities of ASD. Bilgen et al. (2020) have modeled

the morphological relationship between pairs of ROIs with a

cortical networks and veriﬁed the classiﬁcation performance

of diﬀerent machine learning methods. Concerning female

children, Calderoni et al. (2012) have detected the abnormality

of t h e gray matter volume based on SVM-RFE (Leung et al.,

2006; ChenZhiHong et al., 2020) and found the increased cortical

volume in some brain regions involving the left superior frontal

gyrus (SFG). In addition, bilateral SFG and right temporoparietal

junction ( TPJ) resulted in the appearance of some atypical

symptoms of ASD and might be relevant to the pathophysiology

of female children in ASD. These ﬁndings are helpful to reveal

the important inﬂuence of the structural alterations and the

relationship between the brain structure and the pathophysiology

of ASD.

Frontiers in Human Neuroscience | www.frontiersin.org 2 February 2022 | Volume 15 | Article 765517

Duan et al. ASD With Machine Learning

However, the lack of a suﬃciently large sample at a single

site probably leads to poor generalizability that is notably

serious for neuroimaging due to limited participants and super-

high dimensionality of data. Consequently, the investigation

of large sample data from multi-site has attracted increasing

attention. Some studies (

Spera et al., 2019; Mwiza et al.,

2020) have ﬁgured out the superiority of machine learning

for the classiﬁcation of multi-site data based on fMRI or

the combining structural and MRI in the Autism Imaging

Data Exchange database (Di Martino et al., 2017). Due to

the excellent performance of deep learning in the ﬁeld of

artiﬁcial intelligence on large sample data, some researchers have

begun to dete ct abnormalities of functional conne ctiv ity based

on deep learning. Deep learning has combined dimensionality

reduction and feature classiﬁcation a nd implemented the end-

to-end classiﬁcation model automatically, which has achieved

satisfaction per formance (

Eslami et al., 2019; Sherkatghanad

et al., 2020). Furthermore, some attempts have been done to fuse

structural and functional features with the model of deep learning

to improve t he classiﬁcation performance (Raki

c et al., 2020).

But it cannot be denied that deep learning handles data with

the mechanism of a black box and it is so hard to identify the

abnormal brain regions and connect the classiﬁcation accuracy

with t h e underlying mechanism of ASD. Furthermore, multi-

site data also has brought the issue of data heterogeneity due

to diﬀerent scanning parameters and participant populations.

The direct way to address the heterogeneity issue is to apply

dimensionality reduction to transform source data into features

in the ﬁeld of machine learning (

Wang et al., 2020). Furth ermore,

these studies also utilized leave-one-site-out cross-validation to

evaluate the classiﬁcation performance in the expectation of

reducing the impact of heterogeneity simultaneously (Raki

et al., 2020; Eslami et al., 2021). In order to make the results

robust, Ashourvan et al. (2016) have further proposed intra-site

cross-validation a nd inter-site cross-validation and achieved 65%

accuracy with functional connections (FC) to identify ASD from

normal control.

In fact, detecting the structural/functional brain alterations

is vital to reveal the pathological mechanism of ASD. In

particular, these brain regions with obvious diﬀerences can

be recognized as the neuro-imaging biomarkers related to

the disease. Based on this kind of neuro-imaging biomarkers

(brain regions), achieving excellent classiﬁcation performance

even from diﬀerent multi-sites would be the most desirable

and helpful in the clinical diagnosis. Meanwhile, aiming at a

few investigations on volumetric changes based on machine

learning, this study has applied machine learning techniques

followed the uniﬁed framework to implement model-level,

feature-level, and biology-level assessment successively. First

of all, a searchlight-based classiﬁcation method has been used

to detect the volumetric changes locally and some c andid a te

brain regions have been deﬁned based on the areas of the

volumetric changes at the model-level assessment; Regarding

distinguished brain regions, this study has processed the “visual

lesion” analysis at the feature-level assessment. Stability based on

nested cross-validation and multi-site validation of each region

has been evaluated. The candidate regions with good stability

performance have been preserved and considered as candidate

biomarkers related to ASD. Finally, this study investigated

the relationship between candidate biomarkers and symptom

severity and analyzed our results with previous ﬁndings.

The main contributions of the study are discussed as follows:

(1) Previous machine learning st udies on ASD mainly focus

on the classiﬁcation performance or the important features.

Furthermore, this study paid attention to the interpretability of

the machine learning model to explore abnormal brain regions

related to ASD and conducted model level and feature level

assessment to ensure the robustness and stability of the results.

(2) The correlation analysis between abnormal brain regions and

clinical severity in our study has further proved the relationship

between the volume changes of some speciﬁc brain regions

and the c linical symptom. (3) The ﬁndings in our study are

partly consistent with previous research work. The abnormal

gray matter (GM) volume in the temporal lobe, Broca and

Wernicke area probably provides the support for the social brain

hypothesis and the broken mirror theory of ASD, which is helpful

to understand the neuroanatomy of ASD.

The structure of th is study is as follows: First, in section 2,

we provide a brief introduction to the pre-processing procedure

and statistical analysis of sMRI data . In section 3, we describe

the machine learning workﬂow in detail. Experimental results

and discussion are provided in section 4. Finally, in section 4, we

conclude t h e study and discuss the future direction.

2. MATERIAL

2.1. Participants

All data carried in the present study c ame from the Autism

Imaging Data Exchange (ABIDE II) (http://fcon_1000.projects.

nitrc.org/indi/abide/abide_II.html). Brieﬂy, ABIDE with ABIDE

I and ABIDE II is a public repository that provides structural MRI

and resting-state fMRI acquired on ASD and matched control

subjects for the purpose of data sharing and scienti ﬁc research

(

Martino et al., 2013). The ABIDE II includes 1,114 data sets

from 19 independent sites which comprise 521 participants with

ASD and 593 typically developing controls (TDC) with the age

from 5 to 64. All participants in ABIDE have received approval

from the Institutional Review Board (IRB) of e ach site. In

the present study, we have selected three independent datasets

from Georgetown University (GU), Oregon Health and Science

University (OHSU), and University of California Los Angeles

(UCLA) which are collected by t he same sc anner (Siemens) and

all participants are children with the age from 7 to 15 to reduce

the variability of multi-site neuroimaging data. Since GU has the

greatest participants, machine learning methods were conducted

on GU with nested cross-validation. Furthermore, we also trained

machine learning models for GU and tested them on OHSU and

UCLA to verify their robustness. Demographics information of

participants is summarized in Table 1. The s canning parameters

of th e th ree sites are listed in Table 2.

2.2. MRI Data Pre-processing

All structural images were processed using the SPM8 package

(Welcome Trust Center for Neuroimaging, L ondon, UK,

Frontiers in Human Neuroscience | www.frontiersin.org 3 February 2022 | Volume 15 | Article 765517

剩余15页未读，继续阅读

评论收藏

内容反馈

m0_53360345

粉丝: 5
资源: 1

文章2：全文Identifying and Predicting Autism Spectrum Disorder Based ...

评论0

最新资源

文章2：全文Identifying and Predicting Autism Spectrum Disorder Based ...

评论0

人在回路机器学习Human-in-the-Loop_Machine_Learning.pdf

Identifying-Medical-Diagnoses-and-Treatable-Diseases-by-Image-Based_2018_Cel.pdf

eta-nonfab-deploy-guide-2019oct.pdf

Ensemble-based Multi-Filter Feature Selection Method

Machine.Learning.Core.Swift.Beginner.Friendly.Ebook.8

MATLAB for Machine Learning

Beginning.Object-Oriented.Programming.with.C#

Wrox.Press.-.Professional..Programming.pdf

Artificial.Intelligence.and.Soft.Computing.17th.International.Conference.Part.I

R.Deep.Learning.Essentials.1785280589

Migrating.Large-Scale.Services.to.the.Cloud.1484218728

论文研究-Fuzzy Identification Based on Wavelet-Transformed Feature Extraction.pdf

JEDEC JEP158：2009 3D Chip Stack with Through-Silicon Vias (TSVS)：Identifying, Evaluating and Understanding Reliability Interactions - 完整英文电子版（21页）.pdf

Service Support-英文原版

Springer.The.Developer’s.Guide.to.Debugging.2008.pdf

Automatic Tuning of Compilers Using Machine Learning-Springer(2018).pdf

译文_Identifying-Encrypted-Malware-Traffic-with-Contex.docx

Unsupervised.Learning.with.R

Research Advances in Cloud Computing-Springer(2017).pdf

相关实用应用程序（Windows可用）

免费可用的ChatGPT网页版.zip

ChatGPT使用总结：150个ChatGPT提示词模板（完整版）

chromedriver-win64.zip

全国计算机二级WPSoffice精选350道选择题题库（含答案）.pdf

哈尔滨工业大学-ChatGPT调研报告-2023.3.6-94页.pdf

HAI-2024斯坦福AI指数报告（中文译版）.pdf

2023泛娱乐社交出海手册-ZEGO即构科技

4个亲测好用的ChatGPT4渠道

毕业设计的概要介绍与分析

最新资源