Skip to main content
SpringerLink
Log in

Diagnosing autism spectrum disorder in children using conventional MRI and apparent diffusion coefficient based deep learning algorithms

  • Imaging Informatics and Artificial Intelligence
  • Published:
European Radiology Aims and scope Submit manuscript

An Editorial Comment to this article was published on 25 November 2021

Abstract

Objective

To develop and validate deep learning (DL) methods for diagnosing autism spectrum disorder (ASD) based on conventional MRI (cMRI) and apparent diffusion coefficient (ADC) images.

Methods

A total of 151 ASD children and 151 age-matched typically developing (TD) controls were included in this study. The data from these subjects were assigned to training and validation datasets. An additional 20 ASD children and 25 TD controls were acquired, whose data were utilized in an independent test set. All subjects underwent cMRI and diffusion-weighted imaging examination of the brain. We developed a series of DL models to separate ASD from TD based on the cMRI and ADC data. The seven models used include five single-sequence models (SSMs), one dominant-sequence model (DSM), and one all-sequence model (ASM). To enhance the feature detection of the models, we embed an attention mechanism module.

Results

The highest AUC (0.824 ~ 0.850) was achieved when applying the SSM based on either FLAIR or ADC to the validation and independent test sets. A DSM using the combination of FLAIR and ADC showed an improved AUC in the validation (0.873) and independent test sets (0.876). The ASM also showed better diagnostic value in the validation (AUC = 0.838) and independent test sets (AUC = 0.836) compared to the SSMs. Among the models with attention mechanism, the DSM achieved the highest diagnostic performance with an AUC, accuracy, sensitivity, and specificity of 0.898, 84.4%, 85.0%, and 84.0% respectively.

Conclusions

This study established the potential of DL models to distinguish ASD cases from TD controls based on cMRI and ADC images.

Key Points

• Deep learning models based on conventional MRI and ADC can be used to diagnose ASD.

• The model (DSM) based on the FLAIR and ADC sequence achieved the best diagnostic performance with an AUC of 0.836 in the independent test sets.

• The attention mechanism further improved the diagnostic performance of the models.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Similar content being viewed by others

Abbreviations

ADC:

Apparent diffusion coefficient

ASD:

Autism spectrum disorder

ASM:

All-sequence model

cMRI:

Conventional MRI

DL:

Deep learning

DSM:

Dominant-sequence model

DWI:

Diffusion-weighted imaging

SSM:

Single-sequence model

TD:

Typically developing

References

  1. Global Research on Developmental Disabilities Collaborators (2018) Developmental disabilities among children younger than 5 years in 195 countries and territories, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Glob Health 6:e1100–e1121

    Google Scholar 

  2. Maenner MJ, Shaw KA, Baio J et al (2020) Prevalence of autism spectrum disorder among children aged 8 years - Autism and Developmental Disabilities Monitoring Network, 11 sites, United States, 2016. MMWR Surveill Summ 69:1–12

    PubMed  PubMed Central  Google Scholar 

  3. Lyall K, Croen L, Daniels J et al (2017) The changing epidemiology of autism spectrum disorders. Annu Rev Public Health 38:81–102

    PubMed  Google Scholar 

  4. Charman T, Baron-Cohen S, Swettenham J, Baird G, Drew A, Cox A (2003) Predicting language outcome in infants with autism and pervasive developmental disorder. Int J Lang Commun Disord 38:265–285

    PubMed  Google Scholar 

  5. Woods JJ, Wetherby AM (2003) Early identification of and intervention for infants and toddlers who are at risk for autism spectrum disorder. Lang Speech Hear Serv Sch 34:180–193

    PubMed  Google Scholar 

  6. (SIGN) SIGN (2016) Assessment, diagnosis and interventions for autism spectrum disorders: a national clinical guideline. Edinburgh: SIGN; 2016 (SIGN publication no 145)

  7. Zwaigenbaum L, Bauman ML, Choueiri R et al (2015) Early identification and interventions for autism spectrum disorder: executive summary. Pediatrics 136(Suppl 1):S1-9

    PubMed  Google Scholar 

  8. Nylander L, Holmqvist M, Gustafson L, Gillberg C (2013) Attention-deficit/hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) in adult psychiatry. A 20-year register study. Nord J Psychiatry 67:344–350

    PubMed  Google Scholar 

  9. Mandell DS, Ittenbach RF, Levy SE, Pinto-Martin JA (2007) Disparities in diagnoses received prior to a diagnosis of autism spectrum disorder. J Autism Dev Disord 37:1795–1802

    PubMed  Google Scholar 

  10. Ecker C, Bookheimer SY, Murphy DG (2015) Neuroimaging in autism spectrum disorder: brain structure and function across the lifespan. Lancet Neurol 14:1121–1134

    PubMed  Google Scholar 

  11. Zeglam AM, Al-Ogab MF, Al-Shaftery T (2015) MRI or not to MRI! Should brain MRI be a routine investigation in children with autistic spectrum disorders? Acta Neurol Belg 115:351–354

    PubMed  Google Scholar 

  12. Boddaert N, Zilbovicius M, Philipe A et al (2009) MRI findings in 77 children with non-syndromic autistic disorder. PLoS One 4:e4415

    PubMed  PubMed Central  Google Scholar 

  13. Tang S, Xu Y, Liu X et al (2020) Quantitative susceptibility mapping shows lower brain iron content in children with autism. Eur Radiol. https://doi.org/10.1007/s00330-020-07267-w

    Article  PubMed  PubMed Central  Google Scholar 

  14. Wegiel J, Flory M, Kaczmarski W et al (2017) Partial agenesis and hypoplasia of the corpus callosum in idiopathic autism. J Neuropathol Exp Neurol 76:225–237

    PubMed  PubMed Central  Google Scholar 

  15. Paul LK, Corsello C, Kennedy DP, Adolphs R (2014) Agenesis of the corpus callosum and autism: a comprehensive comparison. Brain 137:1813–1829

    PubMed  PubMed Central  Google Scholar 

  16. Numis AL, Major P, Montenegro MA, Muzykewicz DA, Pulsifer MB, Thiele EA (2011) Identification of risk factors for autism spectrum disorders in tuberous sclerosis complex. Neurology 76:981–987

    CAS  PubMed  PubMed Central  Google Scholar 

  17. Abdel Razek A, Mazroa J, Baz H (2014) Assessment of white matter integrity of autistic preschool children with diffusion weighted MR imaging. Brain Dev 36:28–34

    PubMed  Google Scholar 

  18. Mengotti P, D’Agostini S, Terlevic R et al (2011) Altered white matter integrity and development in children with autism: a combined voxel-based morphometry and diffusion imaging study. Brain Res Bull 84:189–195

    PubMed  Google Scholar 

  19. Ben Bashat D, Kronfeld-Duenias V, Zachor DA et al (2007) Accelerated maturation of white matter in young children with autism: a high b value DWI study. Neuroimage 37:40–47

    PubMed  Google Scholar 

  20. Schaefer GB, Mendelsohn NJ (2013) Clinical genetics evaluation in identifying the etiology of autism spectrum disorders: 2013 guideline revisions. Genet Med 15:399–407

    CAS  PubMed  Google Scholar 

  21. Pinaya WHL, Mechelli A, Sato JR (2019) Using deep autoencoders to identify abnormal brain structural patterns in neuropsychiatric disorders: a large-scale multi-sample study. Hum Brain Mapp 40:944–954

    PubMed  Google Scholar 

  22. Akhavan Aghdam M, Sharifi A (2018) Combination of rs-fMRI and sMRI data to discriminate autism spectrum disorders in young children using deep belief network. J Digit Imaging 31:895–903

    PubMed  PubMed Central  Google Scholar 

  23. Chen T, Chen Y, Yuan M et al (2020) The development of a practical artificial intelligence tool for diagnosing and evaluating autism spectrum disorder: multicenter study. JMIR Med Inform 8:e15767

    PubMed  PubMed Central  Google Scholar 

  24. Guo X, Dominick KC, Minai AA, Li H, Erickson CA, Lu LJ (2017) Diagnosing autism spectrum disorder from brain resting-state functional connectivity patterns using a deep neural network with a novel feature selection method. Front Neurosci 11:460

    PubMed  PubMed Central  Google Scholar 

  25. Hazlett HC, Gu H, Munsell BC et al (2017) Early brain development in infants at high risk for autism spectrum disorder. Nature 542:348–351

    CAS  PubMed  PubMed Central  Google Scholar 

  26. Barber AD, Srinivasan P, Joel SE, Caffo BS, Pekar JJ, Mostofsky SH (2012) Motor “dexterity”?: evidence that left hemisphere lateralization of motor circuit connectivity is associated with better motor performance in children. Cereb Cortex 22:51–59

    PubMed  Google Scholar 

  27. Nebel MB, Joel SE, Muschelli J et al (2014) Disruption of functional organization within the primary motor cortex in children with autism. Hum Brain Mapp 35:567–580

    PubMed  Google Scholar 

  28. Sanghyun Woo JP, Joon-Young Lee, In So Kweon (2018) CBAM: Convolutional Block Attention Module. arXiv:180706521v2

  29. Du Tran LB, Rob Fergus, Lorenzo Torresani, Manohar Paluri (2014) Learning spatiotemporal features with 3D convolutional networks. arXiv:14120767

  30. He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, pp 770–778. https://doi.org/10.1109/CVPR.2016.90.

  31. Diederik P, Kingma JB (2014) Adam: a method for stochastic optimization. arXiv:14126980 [cs.LG]

  32. Sujit SJ, Coronado I, Kamali A, Narayana PA, Gabr RE (2019) Automated image quality evaluation of structural brain MRI using an ensemble of deep learning networks. J Magn Reson Imaging 50:1260–1267

    PubMed  PubMed Central  Google Scholar 

  33. Heinsfeld AS, Franco AR, Craddock RC, Buchweitz A, Meneguzzi F (2018) Identification of autism spectrum disorder using deep learning and the ABIDE dataset. Neuroimage Clin 17:16–23

    Google Scholar 

  34. Zielinski BA, Prigge MB, Nielsen JA et al (2014) Longitudinal changes in cortical thickness in autism and typical development. Brain 137:1799–1812

    PubMed  PubMed Central  Google Scholar 

  35. Alexander AL, Lee JE, Lazar M et al (2007) Diffusion tensor imaging of the corpus callosum in autism. Neuroimage 34:61–73

    PubMed  Google Scholar 

  36. Aghdam MA, Sharifi A (2019) Diagnosis of autism spectrum disorders in young children based on resting-state functional magnetic resonance imaging data using convolutional neural networks. J Digit Imaging 32:899–918

    PubMed  PubMed Central  Google Scholar 

  37. Sundaram SK, Kumar A, Makki MI, Behen ME, Chugani HT, Chugani DC (2008) Diffusion tensor imaging of frontal lobe in autism spectrum disorder. Cereb Cortex 18:2659–2665

    PubMed  PubMed Central  Google Scholar 

  38. Ajay K, Sundaram Senthil K, Lalitha S et al (2010) Alterations in frontal lobe tracts and corpus callosum in young children with autism spectrum disorder. Cereb Cortex 20:2103–13

    Google Scholar 

  39. Pinto Gama HP, da Rocha AJ, Braga FT et al (2006) Comparative analysis of MR sequences to detect structural brain lesions in tuberous sclerosis. Pediatr Radiol 36:119–125

    PubMed  Google Scholar 

  40. Jurkiewicz E, Jozwiak S, Bekiesinska-Figatowska M, Pakula-Kosciesza I, Walecki J (2006) Cyst-like cortical tubers in patients with tuberous sclerosis complex: MR imaging with the FLAIR sequence. Pediatr Radiol 36:498–501

    PubMed  Google Scholar 

  41. Xu J, Wang C, Xu Z et al (2020) Specific functional connectivity patterns of middle temporal gyrus subregions in children and adults with autism spectrum disorder. Autism Res 13:410–422

    PubMed  Google Scholar 

  42. Postema MC, van Rooij D, Anagnostou E et al (2019) Altered structural brain asymmetry in autism spectrum disorder in a study of 54 datasets. Nat Commun 10:4958

    CAS  PubMed  PubMed Central  Google Scholar 

  43. Shukla DK, Keehn B, Lincoln AJ, Müller RA (2010) White matter compromise of callosal and subcortical fiber tracts in children with autism spectrum disorder: a diffusion tensor imaging study. J Am Acad Child Adolesc Psychiatry 49(1269–1278):1278.e1261–1262

    Google Scholar 

  44. Sui YV, Donaldson J, Miles L, Babb JS, Castellanos FX, Lazar M (2018) Diffusional kurtosis imaging of the corpus callosum in autism. Mol Autism 9:62

    PubMed  PubMed Central  Google Scholar 

  45. Adorjan I, Ahmed B, Feher V et al (2017) Calretinin interneuron density in the caudate nucleus is lower in autism spectrum disorder. Brain 140:2028–2040

    PubMed  PubMed Central  Google Scholar 

  46. Langen M, Durston S, Staal WG, Palmen SJ, van Engeland H (2007) Caudate nucleus is enlarged in high-functioning medication-naive subjects with autism. Biol Psychiatry 62:262–266

    PubMed  Google Scholar 

  47. Hau J, Aljawad S, Baggett N, Fishman I, Carper RA, Müller RA (2019) The cingulum and cingulate U-fibers in children and adolescents with autism spectrum disorders. Hum Brain Mapp 40:3153–3164

    PubMed  PubMed Central  Google Scholar 

Download references

Funding

The study was supported in part by the grants from Shandong Provincial Development Program of Medical Science and Technology (No. 2016WS0185), Shandong Province Graduate Education Quality Improvement Project (No. SDYKC19213), and Jining Key Research and Development Program (No. 2017SMNS012).

Author information

Authors and Affiliations

  1. Department of Radiology, the Affiliated Hospital of Jining Medical University, Jining, China

    Xiang Guo, Jiehuan Wang, Xiaoqiang Wang, Hao Yu, Li Xu, Hengyan Li & Yueqin Chen

  2. Children Rehabilitation Center, the Affiliated Hospital of Jining Medical University, Jining, China

    Wenjing Liu

  3. Infervision, Beijing, China

    Jiangfen Wu, Mengxing Dong & Weixiong Tan

  4. Department of Medical Imaging, the First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China

    Weijian Chen & Yunjun Yang

Authors
  1. Xiang Guo
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Jiehuan Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Xiaoqiang Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Wenjing Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Hao Yu
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Li Xu
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Hengyan Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Jiangfen Wu
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Mengxing Dong
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Weixiong Tan
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Weijian Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Yunjun Yang
    View author publications

    You can also search for this author in PubMed Google Scholar

  13. Yueqin Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Yueqin Chen.

Ethics declarations

Guarantor

The scientific guarantor of this publication is Yueqin Chen.

Conflict of interest

The authors of this manuscript declare no relationships with any companies, whose products or services may be related to the subject matter of the article.

Statistics and biometry

No complex statistical methods were necessary for this paper.

Informed consent

Written informed consent was waived by the Institutional Review Board.

Ethical approval

Institutional Review Board approval was obtained.

Methodology

• retrospective

• case-control study

• performed at one institution

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (DOCX 2096 KB)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Guo, X., Wang, J., Wang, X. et al. Diagnosing autism spectrum disorder in children using conventional MRI and apparent diffusion coefficient based deep learning algorithms. Eur Radiol 32, 761–770 (2022). https://doi.org/10.1007/s00330-021-08239-4

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00330-021-08239-4

Keywords

Search

Navigation