3090
Microstructural alterations in the corpus callosum of preschool autism spectrum disorder: a diffusion kurtosis imaging study1Department of Radiology, The Third Affiliated Hospital of Zhengzhou University, Zhengzhou, China, 2MR Research China, GE Healthcare, Beijing, China, 3Department of Radiology, Henan Provincial People's Hospital, Zhengzhou, China
Synopsis
Keywords: White Matter, White Matter, Autism spectrum disorder; Diffusion kurtosis imaging; White matter tract integrity; Corpus callosum; Machine learning.
Motivation: With the rising incidence of autism spectrum disorder (ASD) and the absence of clear diagnostic biomarkers, there's an urgent need to facilitate early diagnosis and intervention for affected children.
Goal(s): We aimed to investigate microstructural disparities within the corpus callosum of ASD.
Approach: We extracted diffusion parameters in the corpus callosum's genu, body, and splenium. Logistic Regression and Linear Discriminant Analysis models were constructed to assess the diagnostic potential of each parameter.
Results: Significant distinctions in diffusion and white matter tract integrity metrics were observed, and machine learning models revealed the effectiveness of metrics in ASD diagnosis.
Impact: DKI data can be used to evaluate the abnormalities in the microstructure of the corpus callosum in children with ASD and provide objective measurements to diagnose children with ASD.
Introduction
Autism spectrum disorder (ASD) is a neurodevelopmental condition that emerges in early childhood, marked by social impairments, repetitive behaviors, and narrow interests, with an estimated incidence of 1% to 2% [1, 2]. The corpus callosum, responsible for connecting the brain's two hemispheres, plays a crucial role in various cognitive functions, especially complex information integration [3]. Previous studies of white matter in individuals with ASD have mainly relied on diffusion tensor imaging (DTI) [4, 5]. However, the conventional DTI Gaussian diffusion model falls short in accurately capturing the complex microstructure of white matter. Instead, the non-Gaussian diffusion model of diffusion kurtosis imaging (DKI) proves to be a more suitable choice for depicting the intricate and heterogeneous nature of biological tissue structure [6]. Additionally, the white matter tract integrity (WMTI) index provides a comprehensive perspective on the diffusion patterns of water molecules within both intra-axonal and extra-axonal compartments [7]. Given the effectiveness of WMTI parameters in analyzing highly organized white matter fiber bundles, our study focused on the corpus callosum as the primary region of interest.Materials and methods
This research encompassed 58 children with ASD and 43 healthy control (HC) children, aged 20 to 60 months. All participants underwent comprehensive examinations using a high-field 3.0T MRI scanner (SIGNA PIONEER, GE Healthcare, Waukesha, WI). The preprocessing of diffusion kurtosis imaging (DKI) data, using the PyDesigner procedure [8], led to the derivation of 13 diffusion parameters. These parameters included diffusion tensor (DT) metrics like fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD), alongside diffusion kurtosis (DK) metrics such as axial kurtosis (AK), kurtosis fractional anisotropy (KFA), mean kurtosis (MK), and radial kurtosis (RK). Additionally, the study incorporated white matter tract integrity (WMTI) metrics, such as axonal water fraction (AWF), axonal diffusivity of the extra-axial space (EAS_AD), radial diffusivity of the extra-axial space (EAS_RD), tortuosity of the extra-axonal space (EAS_TORT), and diffusivity of the intra-axonal space (IAS_Da). Subsequently, diffusion parameters were normalized to the FMRIB58_FA template using the nonlinear normalization tool FNIRT in the FSL software [9]. Following normalization, we extracted the diffusion parameters within the three regions of interest: the genu, body, and splenium of the corpus callosum, in accordance with the ICBM-DTI-81 white-matter labels atlas (Figure 1).R (https://www.r-project.org/) was employed for statistical analysis. Independent sample T-test or Wilcoxon's test was conducted to compare 13 metrics in the genu, body, and splenium between ASD and HC group. The Bonferroni correction was conducted to correct multiple comparisons, and statistical significance was set at p < 0.05/3 = 0.0167. The logistic regression (LR) and linear discriminant analysis (LDA) models were established by individual indexes derived from the three subregions of the corpus callosum. Model validation employed a bootstrap-resampling approach with 1000 resampling iterations, evaluating key metrics, including the area under the curve (AUC), accuracy, sensitivity, and specificity.
Results
Regarding the DT indicators, a significant increase in AD was observed in the body and splenium of the corpus callosum (p = 0.010; p < 0.001). MD also showed a notable increase in the body and splenium of the corpus callosum (p = 0.004; p = 0.006). As for diffusion kurtosis parameters, substantial increases in MK, AK, and KFA (p < 0.001) were noted in all three subregions of the corpus callosum (Figure 2). Concerning WMTI parameters, AWF showed significant increases in all three subregions of the corpus callosum (all p < 0.001). Moreover, the body of the corpus callosum displayed a notable increase in EAS_TORT (p < 0.001) and a significant decrease in EAS_RD (p = 0.002) (Figure 2).Of the two machine learning models, the LR and LDA models, established using DT, DK, and WMTI parameters, demonstrated robust diagnostic performance. Notably, the LR model built with AWF achieved the highest AUC and accuracy values, with an AUC of 0.804 and an accuracy of 0.721, respectively (Figure 3).Discussion and Conclusion
The findings of this study unveil significant disparities in the diffusion parameters of the corpus callosum between children with ASD and the HC group. These differences imply alterations in the microstructure of the corpus callosum in individuals with ASD, potentially linked to overdevelopment or excessive maturation in young children with ASD. Furthermore, the LR and LDA models applied to the diffusion parameters demonstrate high AUC values, highlighting the robust diagnostic potential of DKI and its derived WMTI parameters. These results serve as a foundation for considering the DKI data and its derived WMTI metrics as potential biomarkers for ASD.Acknowledgements
No acknowledgmentReferences
1. Shaw, K.A., et al., Early Identification of Autism Spectrum Disorder Among Children Aged 4 Years - Autism and Developmental Disabilities Monitoring Network, 11 Sites, United States, 2020. MMWR Surveill Summ, 2023. 72(1): p. 1-15.
2. Xu, G., et al., Prevalence of Autism Spectrum Disorder Among US Children and Adolescents, 2014-2016. Jama, 2018. 319(1): p. 81-82.
3. Paul, L.K., et al., Agenesis of the corpus callosum: genetic, developmental and functional aspects of connectivity. Nat Rev Neurosci, 2007. 8(4): p. 287-99.
4. Zhao, Y., et al., Identify aberrant white matter microstructure in ASD, ADHD and other neurodevelopmental disorders: A meta-analysis of diffusion tensor imaging studies. Prog Neuropsychopharmacol Biol Psychiatry, 2022. 113: p. 110477.
5. Dimond, D., et al., Reduced White Matter Fiber Density in Autism Spectrum Disorder. Cereb Cortex, 2019. 29(4): p. 1778-1788.
6. Jensen, J.H., et al., Diffusional kurtosis imaging: the quantification of non-gaussian water diffusion by means of magnetic resonance imaging. Magn Reson Med, 2005. 53(6): p. 1432-40.
7. Fieremans, E., J.H. Jensen, and J.A. Helpern, White matter characterization with diffusional kurtosis imaging. Neuroimage, 2011. 58(1): p. 177-88.
8. Siddhartha, D., et al., PyDesigner: A Pythonic Implementation of the DESIGNER Pipeline for Diffusion Tensor and Diffusional Kurtosis Imaging. bioRxiv, 2021: p. 2021.10.20.465189.
9. Jenkinson, M., et al., FSL. Neuroimage, 2012. 62(2): p. 782-90.
Figures
