Advertisement

Evaluation of Infants with Spinal Muscular Atrophy Type-I Using Convolutional Neural Networks

  • Bilge SoranEmail author
  • Linda Lowes
  • Katherine M. Steele
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9914)

Abstract

Spinal Muscular Atrophy is the most common genetic cause of infant death. Due to its severity, there is a need for methods for automated estimation of disease progression. In this paper we propose a Convolutional-Neural-Network (CNN) model to estimate disease progression during infants’ natural behavior. With the proposed methodology, we were able to predict each child’s score on current behavior-based clinical exams with an average per-subject error of 6.96 out of 72 points (<10 % difference), using 30-second videos in leave-one-subject-out-cross-validation setting. When simple statistics were used over 30-second video-segments to estimate a score for longer videos, we obtained an average error of 5.95 (\(\sim \)8 % error rate). By showing promising results on a small dataset (N \(=\) 70, 2-minute samples, which were handled as 1487, 30-second video segments), our methodology demonstrates that it is possible to benefit from CNNs on small datasets by proper design and data handling choices.

Keywords

Spinal muscular atrophy Longitudinal anlaysis Microsoft kinect Convolutional neural networks Regression 

References

  1. 1.
    http://cs231n.github.io/neural-networks 2. Accessed 25 June 2016
  2. 2.
    Adde, L., et al.: General movement assessment: predicting cerebral palsy in clinical practise. Early Hum. Dev. 83(1), 13–18 (2007)CrossRefGoogle Scholar
  3. 3.
    Benettazzo, F., Iarlori, S., Ferracuti, F., Giantomassi, A., Ortenzi, D., Freddi, A., Monteriù, A., Innocenzi, S., Capecci, M., Ceravolo, M.G., Longhi, S.: Low cost RGB-D vision based system to support motor disabilities rehabilitation at home. In: Andó, B., Siciliano, P., Marletta, V., Monteriú, A. (eds.) Ambient Assisted Living. Biosystems & Biorobotics, pp. 449–461. Springer, Cham (2015)CrossRefGoogle Scholar
  4. 4.
    Cano, S., et al.: Rasch analysis of clinical outcome measures in spinal muscular atrophy. Muscle Nerve 49(3), 422–430 (2014)CrossRefGoogle Scholar
  5. 5.
    Cioni, G., et al.: Which better predicts later outcome in fullterm infants: quality of general movements or neurological examination? Early Hum. Dev. 50(1), 71–85 (1997)CrossRefGoogle Scholar
  6. 6.
    Crawford, T.: Concerns about the design of clinical trials for spinal muscular atrophy. Neuromuscul. Disord. 14(8), 456–460 (2004)CrossRefGoogle Scholar
  7. 7.
    Finkel, R.: Electrophysiological and motor function scale association in a pre-symptomatic infant with spinal muscular atrophy type I. Neuromuscul. Disord. 23(2), 112–115 (2013)CrossRefGoogle Scholar
  8. 8.
    Finkel, R., et al.: Observational study of spinal muscular atrophy type I and implications for clinical trials. Neurology 83(9), 810–817 (2014)CrossRefGoogle Scholar
  9. 9.
    Geng, X., et al.: Automatic age estimation based on facial aging patterns. PAMI 29(12), 2234–2240 (2007)CrossRefGoogle Scholar
  10. 10.
    Glanzman, A., et al.: The Children’s hospital of philadelphia infant test of neuromuscular disorders (CHOP INTEND): test development and reliability. Neuromuscul. Disord. 20(3), 155–161 (2010)CrossRefGoogle Scholar
  11. 11.
    Glorot, X., Bengio, Y.: Understanding the difficulty of training deep feedforward neural networks. In: AISTATS (2010)Google Scholar
  12. 12.
    González-Ortega, D., et al.: A kinect-based system for cognitive rehabilitation exercises monitoring. Comput. Methods Programs Biomed. 113(2), 620–631 (2014)CrossRefGoogle Scholar
  13. 13.
    Hadders-Algra, M.: Evaluation of motor function in young infants by means of the assessment of general movements: a review. Ped. Phys. Ther. 13(1), 27–36 (2001)Google Scholar
  14. 14.
    Han, J., et al.: Reachable workspace in facioscapulohumeral muscular dystrophy (FSHD) by kinect. Muscle Nerve 51(2), 168–175 (2015)CrossRefGoogle Scholar
  15. 15.
    Hondori, H., et al.: A review on technical and clinical impact of microsoft kinect on physical therapy and rehabilitation. J. Med. Eng. (2014)Google Scholar
  16. 16.
    Iarlori, S., et al.: RGBD camera monitoring system for alzheimers disease assessment using recurrent neural networks with parametric bias action recognition. In: ICCHP (2014)Google Scholar
  17. 17.
    Kontschieder, P., et al.: Quantifying progression of multiple sclerosis via classification of depth videos. In: Golland, P., Hata, N., Barillot, C., Hornegger, J., Howe, R. (eds.) MICCAI 2014, Part II. LNCS, vol. 8674, pp. 429–437. Springer, Heidelberg (2014)Google Scholar
  18. 18.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: NIPS (2012)Google Scholar
  19. 19.
    Mu, G., et al.: Human age estimation using bio-inspired features. In: CVPR (2009)Google Scholar
  20. 20.
    Niu, Z., et al.: Ordinal regression with multiple output CNN for age estimation. In: CVPR (2016)Google Scholar
  21. 21.
    Oskarsson, B., et al.: Upper extremity 3-dimensional reachable workspace assessment in amyotrophic lateral sclerosis by kinect sensor. Muscle Nerve 53(2), 234–241 (2016)CrossRefGoogle Scholar
  22. 22.
    Phagava, H., et al.: General movements in infants with autism spectrum disorders. In: Georgian Medical News (2008)Google Scholar
  23. 23.
    Prechtl, H., Hopkins, B.: Developmental transformations of spontaneous movements in early infancy. Early Hum. Dev. 14(3), 233–238 (1986)CrossRefGoogle Scholar
  24. 24.
    Rahmati, H., et al.: Frequency-based features for early cerebral palsy prediction. In: EMBC (2015)Google Scholar
  25. 25.
    Russman, B.S.: Spinal muscular atrophy: clinical classification and disease heterogeneity. J. Child Neurol. 22(8), 946–951 (2007)CrossRefGoogle Scholar
  26. 26.
    Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. CoRR (2014)Google Scholar
  27. 27.
    Sooklal, S., et al.: Using the kinect for detecting tremors: challenges and opportunities. In: BHI (2014)Google Scholar
  28. 28.
    Spasojević, S., et al.: A vision-based system for movement analysis in medical applications: the example of parkinson disease. In: Computer Vision Systems (2015)Google Scholar
  29. 29.
    Srivastava, N., et al.: Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15(1), 1929–1958 (2014)MathSciNetzbMATHGoogle Scholar
  30. 30.
    Stahl, A., et al.: An optical flow-based method to predict infantile cerebral palsy. Neural Syst. Rehabil. Eng. 20(4), 605–614 (2012)CrossRefGoogle Scholar
  31. 31.
    Swoboda, K.J., et al.: Perspectives on clinical trials in spinal muscular atrophy. J. Child Neurol. 22(8), 957–966 (2007)CrossRefGoogle Scholar
  32. 32.
    Szegedy, C., et al.: Going deeper with convolutions. In: CVPR (2015)Google Scholar
  33. 33.
    Yi, D., Lei, Z., Li, S.Z.: Age estimation by multi-scale convolutional network. In: Cremers, D., Reid, I., Saito, H., Yang, M.-H. (eds.) ACCV 2014. LNCS, vol. 9005, pp. 144–158. Springer, Heidelberg (2015)Google Scholar
  34. 34.
    Zeiler, M.D., Fergus, R.: Visualizing and understanding convolutional networks. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014, Part I. LNCS, vol. 8689, pp. 818–833. Springer, Heidelberg (2014)Google Scholar
  35. 35.
    Zerres, K., Rudnik-Schoneborn, S.: Natural history in proximal spinal muscular atrophy: clinical analysis of 445 patients and suggestions for a modification of existing classifications. Arch. Neurol. 52(5), 518–523 (1995)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  • Bilge Soran
    • 1
    Email author
  • Linda Lowes
    • 2
  • Katherine M. Steele
    • 1
  1. 1.Mechanical Engineering DepartmentUniversity of WashingtonSeattleUSA
  2. 2.Clinical Therapies DepartmentNationwide Children’s HospitalColumbusUSA

Personalised recommendations