Skip to main content

Deep learning to differentiate parkinsonian disorders separately using single midsagittal MR imaging: a proof of concept study



To evaluate the diagnostic performance of deep learning with the convolutional neural networks (CNN) to distinguish each representative parkinsonian disorder using MRI.


This clinical retrospective study was approved by the institutional review board, and the requirement for written informed consent was waived. Midsagittal T1-weighted MRI of a total of 419 subjects (125 Parkinson’s disease (PD), 98 progressive supranuclear palsy (PSP), and 54 multiple system atrophy with predominant parkinsonian features (MSA-P) patients, and 142 normal subjects) between January 2012 and April 2016 was retrospectively assessed. To deal with the overfitting problem of deep learning, all subjects were randomly divided into training (85%) and validation (15%) data sets with the same proportions of each disease and normal subjects. We trained the CNN to distinguish each parkinsonian disorder using single midsagittal T1-weighted MRI with a training group to minimize the differences between predicted output probabilities and the clinical diagnoses; then, we adopted the trained CNN to the validation data set. Subjects were classified into each parkinsonian disorder or normal condition according to the final diagnosis of the trained CNN, and we assessed the diagnostic performance of the CNN.


The accuracies of diagnostic performances regarding PD, PSP, MSA-P, and normal subjects were 96.8, 93.7, 95.2, and 98.4%, respectively. The areas under the receiver operating characteristic curves for distinguishing each condition from others (PD, PSP, MSA-P, and normal subjects) were 0.995, 0.982, 0.990, and 1.000, respectively.


Deep learning with CNN enables highly accurate discrimination of parkinsonian disorders using MRI.

Key Points

Deep learning convolution neural network achieves differential diagnosis of PD, PSP, MSA-P, and normal controls with an accuracy of 96.8, 93.7, 95.2, and 98.4%, respectively.

The areas under the curves for distinguishing between PD, PSP, MSA-P, and normality were 0.995, 0.982, 0.990, and 1.000, respectively.

CNN may learn important features that humans not notice, and has a possibility to perform previously impossible diagnoses.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6



Convolutional neural network


Central nervous system


Diffusion-weighted imaging


Joint photographic experts group


Magnetic resonance


Multiple system atrophy with predominant parkinsonian features


Parkinson’s disease


Progressive supranuclear palsy


Rectified linear unit


Receiver operating characteristic


  1. 1.

    Seppi K, Yekhlef F, Diem A et al (2005) Progression of parkinsonism in multiple system atrophy. J Neurol 252:91–96

    Article  Google Scholar 

  2. 2.

    Dickson DW, Rademakers R, Hutton ML (2007) Progressive supranuclear palsy: pathology and genetics. Brain Pathol 17:74–82

    CAS  Article  Google Scholar 

  3. 3.

    Marras C, Lang A (2008) Invited article: changing concepts in Parkinson disease: moving beyond the decade of the brain. Neurology 70:1996–2003

    Article  Google Scholar 

  4. 4.

    Rizzo G, Copetti M, Arcuti S, Martino D, Fontana A, Logroscino G (2016) Accuracy of clinical diagnosis of Parkinson disease: a systematic review and meta-analysis. Neurology 86:566–576

    Article  Google Scholar 

  5. 5.

    Hotter A, Esterhammer R, Schocke MF, Seppi K (2009) Potential of advanced MR imaging techniques in the differential diagnosis of parkinsonism. Mov Disord 24:S711–S720

    Article  Google Scholar 

  6. 6.

    Cosottini M, Ceravolo R, Faggioni L et al (2007) Assessment of midbrain atrophy in patients with progressive supranuclear palsy with routine magnetic resonance imaging. Acta Neurol Scand 116:37–42

    CAS  Article  Google Scholar 

  7. 7.

    Oba H, Yagishita A, Terada H et al (2005) New and reliable MRI diagnosis for progressive supranuclear palsy. Neurology 64:2050–2055

    CAS  Article  Google Scholar 

  8. 8.

    Heim B, Krismer F, De Marzi R, Seppi K (2017) Magnetic resonance imaging for the diagnosis of Parkinson’s disease. J Neural Transm (Vienna) 124:915–964

    Article  Google Scholar 

  9. 9.

    Yasaka K, Akai H, Kunimatsu A, Kiryu S, Abe O (2018) Deep learning with convolutional neural network in radiology. Jpn J Radiol 36:257–272

    Article  Google Scholar 

  10. 10.

    Lee JG, Jun S, Cho YW et al (2017) Deep learning in medical imaging: general overview. Korean J Radiol 18:570–584

    Article  Google Scholar 

  11. 11.

    van der Burgh HK, Schmidt R, Westeneng HJ, de Reus MA, van den Berg LH, van den Heuvel MP (2016) Deep learning predictions of survival based on MRI in amyotrophic lateral sclerosis. Neuroimage Clin 13:361–369

    Article  Google Scholar 

  12. 12.

    Laukamp KR, Thiele F, Shakirin G et al (2019) Fully automated detection and segmentation of meningiomas using deep learning on routine multiparametric MRI. Eur Radiol 29:124–132

    Article  Google Scholar 

  13. 13.

    Lin W, Tong T, Gao Q et al (2018) Convolutional neural networks-based MRI image analysis for the Alzheimer’s disease prediction from mild cognitive impairment. Front Neurosci 12:777

    Article  Google Scholar 

  14. 14.

    Xin J, Zhang Y, Tang Y, Yang Y (2019) Brain differences between men and women: evidence from deep learning. Front Neurosci 13:185

    Article  Google Scholar 

  15. 15.

    Korfiatis P, Erickson B (2019) Deep learning can see the unseeable: predicting molecular markers from MRI of brain gliomas. Clin Radiol 74:367–373

    CAS  Article  Google Scholar 

  16. 16.

    Gong E, Pauly JM, Wintermark M, Zaharchuk G (2018) Deep learning enables reduced gadolinium dose for contrast-enhanced brain MRI. J Magn Reson Imaging 48:330–340

    Article  Google Scholar 

  17. 17.

    Litvan I, Agid Y, Calne D et al (1996) Clinical research criteria for the diagnosis of progressive supranuclear palsy (Steele-Richardson-Olszewski syndrome): report of the NINDS-SPSP International Workshop. Neurology 47:1–9

    CAS  Article  Google Scholar 

  18. 18.

    Hughes AJ, Daniel SE, Kilford L, Lees AJ (1992) Accuracy of clinical diagnosis of idiopathic Parkinson’s disease: a clinico-pathological study of 100 cases. J Neurol Neurosurg Psychiatry 55:181–184

    CAS  Article  Google Scholar 

  19. 19.

    Gilman S, Wenning GK, Low PA et al (2008) Second consensus statement on the diagnosis of multiple system atrophy. Neurology 71:670–676

    CAS  Article  Google Scholar 

  20. 20.

    Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 15:1929–1958

    Google Scholar 

  21. 21.

    Kanda Y (2013) Investigation of the freely available easy-to-use software “EZR” for medical statistics. Bone Marrow Transplant 48:452–458

    CAS  Article  Google Scholar 

  22. 22.

    Rizzo G, Zanigni S, De Blasi R et al (2016) Brain MR contribution to the differential diagnosis of parkinsonian syndromes: an update. Parkinsons Dis 2016:2983638

    PubMed  PubMed Central  Google Scholar 

  23. 23.

    Möller L, Kassubek J, Südmeyer M et al (2017) Manual MRI morphometry in parkinsonian syndromes. Mov Disord 32:778–782

    Article  Google Scholar 

  24. 24.

    Goldman JG, Bledsoe IO, Merkitch D, Dinh V, Bernard B, Stebbins GT (2017) Corpus callosal atrophy and associations with cognitive impairment in Parkinson disease. Neurology 88:1265–1272

    Article  Google Scholar 

  25. 25.

    Boelmans K, Bodammer NC, Suchorska B et al (2010) Diffusion tensor imaging of the corpus callosum differentiates corticobasal syndrome from Parkinson’s disease. Parkinsonism Relat Disord 16:498–502

    Article  Google Scholar 

  26. 26.

    Rosskopf J, Müller HP, Huppertz HJ, Ludolph AC, Pinkhardt EH, Kassubek J (2014) Frontal corpus callosum alterations in progressive supranuclear palsy but not in Parkinson’s disease. Neurodegener Dis 14:184–193

    Article  Google Scholar 

  27. 27.

    Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D (2017) Grad-CAM: visual explanations from deep networks via gradient-based localization. Available via Accessed 2 Feb 2019

  28. 28.

    Samek W, Binder A, Montavon G, Lapuschkin S, Muller KR (2017) Evaluating the visualization of what a deep neural network has learned. IEEE Trans Neural Netw Learn Syst 28:2660–2673

    Article  Google Scholar 

  29. 29.

    Philbrick KA, Yoshida K, Inoue D et al (2018) What does deep learning see? Insights from a classifier trained to predict contrast enhancement phase from CT images. AJR Am J Roentgenol 211:1184–1193

    Article  Google Scholar 

Download references


The authors state that this work has not received any funding.

Author information



Corresponding author

Correspondence to Shigeru Kiryu.

Ethics declarations


The scientific guarantor of this publication is Dr. Shigeru Kiryu.

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.


• retrospective

• diagnostic or prognostic 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.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Kiryu, S., Yasaka, K., Akai, H. et al. Deep learning to differentiate parkinsonian disorders separately using single midsagittal MR imaging: a proof of concept study. Eur Radiol 29, 6891–6899 (2019).

Download citation


  • Artificial intelligence
  • Parkinson disease
  • Magnetic resonance imaging
  • ROC curve
  • Deep learning