Quantifying the Impact of Type 2 Diabetes on Brain Perfusion Using Deep Neural Networks

  • Behrouz Saghafi
  • Prabhat Garg
  • Benjamin C. Wagner
  • S. Carrie Smith
  • Jianzhao Xu
  • Ananth J. Madhuranthakam
  • Youngkyoo Jung
  • Jasmin Divers
  • Barry I. Freedman
  • Joseph A. Maldjian
  • Albert Montillo
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10553)

Abstract

The effect of Type 2 Diabetes (T2D) on brain health is poorly understood. This study aims to quantify the association between T2D and perfusion in the brain. T2D is a very common metabolic disorder that can cause long term damage to the renal and cardiovascular systems. Previous research has discovered the shape, volume and white matter microstructures in the brain to be significantly impacted by T2D. We propose a fully-connected deep neural network to classify the regional Cerebral Blood Flow into low or high levels, given 16 clinical measures as predictors. The clinical measures include diabetes, renal, cardiovascular and demographics measures. Our model enables us to discover any nonlinear association which might exist between the input features and target. Moreover, our end-to-end architecture automatically learns the most relevant features and combines them without the need for applying a feature selection method. We achieved promising classification performance. Furthermore, in comparison with six (6) classical machine learning algorithms and six (6) alternative deep neural networks similarly tuned for the task, our proposed model outperformed all of them.

References

  1. 1.
    Ashburner, J.: A fast diffeomorphic image registration algorithm. Neuroimage 38(1), 95–113 (2007)CrossRefGoogle Scholar
  2. 2.
    CDC: Estimates of diabetes and its burden in the united states. National diabetes statistics report (2014)Google Scholar
  3. 3.
    Freedman, B.I., Divers, J., Whitlow, C.T., Bowden, D.W., Palmer, N.D., et al.: Subclinical atherosclerosis is inversely associated with gray matter volume in African Americans with type 2 diabetes. Diab. Care 38(11), 2158–2165 (2015)CrossRefGoogle Scholar
  4. 4.
    Hsu, J.L., Chen, Y.L., Leu, J.G., Jaw, F.S., Lee, C.H., Tsai, Y.F., Hsu, C.Y., Bai, C.H., Leemans, A.: Microstructural white matter abnormalities in type 2 diabetes mellitus: a diffusion tensor imaging study. NeuroImage 59(2), 1098–1105 (2012)CrossRefGoogle Scholar
  5. 5.
    Kingma, D., Ba, J.: Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)
  6. 6.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: NIPS (2012)Google Scholar
  7. 7.
    Maldjian, J.A., Laurienti, P.J., Kraft, R.A., Burdette, J.H.: An automated method for neuroanatomic and cytoarchitectonic atlas-based interrogation of FMRI data sets. Neuroimage 19(3), 1233–1239 (2003)CrossRefGoogle Scholar
  8. 8.
    Malenka, R.C., Nestler, E., Hyman, S., Sydor, A., Brown, R., et al.: Molecular Neuropharmacology: A Foundation for Clinical Neuroscience. McGrawHill Medical, New York (2009)Google Scholar
  9. 9.
    Murea, M., Hsu, F.C., Cox, A.J., Hugenschmidt, C.E., Xu, J., Adams, J.N., et al.: Structural and functional assessment of the brain in European Americans with mild-to-moderate kidney disease: diabetes heart study-mind. Nephrol. Dial. Transplant. 30(8), 1322–1329 (2015)CrossRefGoogle Scholar
  10. 10.
    Peng, B., Chen, Z., Ma, L., Dai, Y.: Cerebral alterations of type 2 diabetes mellitus on MRI: a pilot study. Neurosci. Lett. 606, 100–105 (2015)CrossRefGoogle Scholar
  11. 11.
    Raffield, L.M., Cox, A.J., Freedman, B.I., Hugenschmidt, C.E., Hsu, F.C., et al.: Analysis of the relationships between type 2 diabetes status, glycemic control, and neuroimaging measures in the diabetes heart study mind. Acta Diabetologica 53(3), 439–447 (2016)CrossRefGoogle Scholar
  12. 12.
    Rofey, D.L., Arslanian, S.A., El Nokali, N.E., Verstynen, T., Watt, J.C., Black, J.J., et al.: Brain volume and white matter in youth with type 2 diabetes compared to obese and normal weight, non-diabetic peers: a pilot study. Int. J. Dev. Neurosci. 46, 88–91 (2015)CrossRefGoogle Scholar
  13. 13.
    Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. In: ICLR (2015)Google Scholar
  14. 14.
    Sink, K.M., Divers, J., Whitlow, C.T., Palmer, N.D., Smith, S.C., Xu, J., et al.: Cerebral structural changes in diabetic kidney disease: African American-diabetes heart study mind. Diab. Care 38(2), 206–212 (2015)CrossRefGoogle Scholar
  15. 15.
    Tzourio-Mazoyer, N., Landeau, B., Papathanassiou, D., Crivello, F., Etard, O., et al.: Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain. Neuroimage 15(1), 273–289 (2002)CrossRefGoogle Scholar
  16. 16.
    Yager, L., Garcia, A., Wunsch, A., Ferguson, S.: The ins and outs of the striatum: role in drug addiction. Neuroscience 301, 529–541 (2015)CrossRefGoogle Scholar
  17. 17.
    Zhang, T., Shaw, M., Humphries, J., Sachdev, P., Anstey, K.J., Cherbuin, N.: Higher fasting plasma glucose is associated with striatal and hippocampal shape differences: the 2sweet project. BMJ Open Diab. Res. Care 4(1), 1–8 (2016)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • Behrouz Saghafi
    • 1
  • Prabhat Garg
    • 1
  • Benjamin C. Wagner
    • 1
  • S. Carrie Smith
    • 2
  • Jianzhao Xu
    • 2
  • Ananth J. Madhuranthakam
    • 1
  • Youngkyoo Jung
    • 2
  • Jasmin Divers
    • 2
  • Barry I. Freedman
    • 2
  • Joseph A. Maldjian
    • 1
  • Albert Montillo
    • 1
  1. 1.University of Texas Southwestern Medical CenterDallasUSA
  2. 2.Wake Forest School of MedicineWinston-SalemUSA

Personalised recommendations