Table 2 Summary of the behavioral and imaging findings in ASD.

From: Emerging behavioral and neuroimaging biomarkers for early and accurate characterization of autism spectrum disorders: a systematic review

Reference

Study type

Subjects

Age range

Method

Brain regions

Findings

Hazlett et al.64

A prospective and longitudinal study

High and low-risk ASD

6–24 months

Structural brain MRI

Cortical surface area

Hyperexpansion of cortical surface area precedes brain volume overgrowth

Nordahl et al.68

A prospective and longitudinal study

ASD and typically developing

2–4 years

Structural brain MRI

Amygdala

Increased total cerebral and amygdala volume

Pote et al.65

A prospective and longitudinal study

High and low-risk ASD

4–36 months

Structural brain MRI

Cerebellum and subcortical regions

Larger cerebellum and subcortical volumes

Schumann et al.71

A prospective and longitudinal study

ASD and typically developing

1.5–5 years

Structural brain MRI

Cerebral gray and white matter

Increase in cerebral gray and white matter specifically in frontal, temporal, and cingulate cortices

Shen et al.66,67

A prospective and longitudinal study

High-risk and low-risk ASD

6 months to 3 years

Structural brain MRI

Extra-axial cerebrospinal fluid

Increase in volume of extra-axial cerebrospinal fluid

Zhu et al.69

Prospective and cross-sectional

ASD and typically developing

2–5 years

Structural brain MRI

Amygdala

Larger bilateral amygdala volumes

Conti et al.97

Prospective and cross-sectional

ASD and other developmental delays

Below 36 months

DTI

White matter connectivity

The over-connectivity pattern in ASD occurred in networks involving the frontotemporal nodes and basal ganglia

Fingher et al.86

Prospective and cross-sectional

ASD and typically developing

<2.5 years

DTI

Corpus callosum

Fractional anisotropy and the cross-sectional area of the temporal corpus callosum tract were significantly larger

Solso et al.94

Prospective and longitudinal

ASD and typically developing

12–48 months

DTI

Frontal lobe

Frontal fiber tracts displayed deviant early development and age-related changes

Weinstein et al.84

Prospective and cross-sectional

ASD and typically developing

1.5–5.8 years

DTI

Corpus callosum left superior longitudinal fasciculus and right and left cingulum

Increase in fractional anisotropy was seen in corpus callosum as well as in the left cingulum

Wolff et al.88

A prospective and longitudinal study

High risk and ASD

6–24 months

DTI

White matter fiber tract

Aberrant development of WM pathways may precede the manifestation of autistic symptoms in the first year of life

Wolff et al.87

A prospective and longitudinal study

High risk and ASD

6–24 months

DTI

Cerebellum and corpus callosum

Aberrant structural properties of callosal and cerebellar white matter pathways were observed

Xiao et al.98

Prospective and cross-sectional

ASD and developmental delay

2–3 years

Structural imaging and DTI

Gray and white matter volumes and corpus callosum, cingulate cortex, and limbic lobes

An increase in global gray and white matter volume and higher fractional anisotropy value was also observed in the corpus callosum, posterior cingulate cortex, and limbic lobes

Ciarrusta et al.113

Prospective and cross-sectional

High risk and typically developing

Neonates

Resting-state fMRI

Resting-state networks

Significantly high regional homogeneity within multiple resting-state networks

Eggebrecht et al.103

Prospective and longitudinal

High risk and typically developing

12–24 months

fMRI

visual, dorsal attention and default mode network

Association between joint attention and these mentioned networks were found

Eyler et al.105

Prospective and longitudinal

ASD and typically developing

12–48 months

Resting-state fMRI

Temporal

Abnormally right-lateralized temporal cortex response to language

McKinnon et al.106

Prospective and longitudinal

High risk and typically developing

12–24 months

Resting-state fMRI

Dorsal attention, subcortical, and default mode network

Association between these networks and RRBs were found except that for self-injurious behaviors