1. Proceedings
  2. I3M
  3. 2018
  4. IWISH
  5. 10.46354/i3m.2018.iwish.003

Evaluation of hardware-based evolutionary algorithms for the identification of motion-based action potentials in neural bundles

  • Volkhard Klinger  
  • Department of Embedded Systems University of Applied Science Hannover (FHDW) D.30173 Hannover, LS, Germany
Cite as
Klinger V. (2018). Evaluation of hardware-based evolutionary algorithms for the identification of motion-based action potentials in neural bundles. Proceedings of the 7th International Workshop on Innovative Simulation for Healthcare (IWISH 2018), pp. 14-21. DOI: https://doi.org/10.46354/i3m.2018.iwish.003

Abstract

The recording and interpretation of nerve signals for controlling exoprosthesis is a current challenge in medical technology. The aim of the higher-level project is both the measurement and identification of bio-signals. Based on evolutionary algorithms, using deep learning strategies, a model is created in a learning phase that describes the correlation between electroneurogram (ENG)-signals and motion sequences. In the following operation phase, this model will be used in daily life to control the prosthesis. In order to make the model adaptive in operation mode depending on various parameters, such as physiological parameters, body condition, change in position of the implanted cuff electrode and changes and further additions to the relationship between movement and ENG-signals, the objective is to provide continuous optimization of the model even in operation mode. This article presents an evaluation of a hardware implementation of an evolutionary algorithm with regard to speed and hardware-resources and compares it with a corresponding software solution. The results are evaluated both for use of a hardware solution as a local model optimizer and for use as a speed-up device within the learning mode.

References

  1. Alfke, P. (1996). Efficient shift registers, lfsr counters, and long pseudo-random sequence generators. Technical report, XILINX. Version 1.1.
  2. De Jong, K. A. (2006). Evolutionary Computation: A Unified Approach. MIT press.
  3. Gold, C., Henze, D. A., and Koch, C. (2007). Using extracellular action potential recordings to constrain compartmental models. Journal of Computational Neuroscience, 23(1):39–58.
  4. Goldberg, D. (1989). Genetic Algorithms in Search, Optimization and Machine Learning. Addison Wesley, MA.
  5. Hazan, L., Zugaro, M., and Buzs´aki, G. (2006). Klusters, NeuroScope, NDManager: A free software suite for neurophysiological data processing and visualization. J Neurosci Methods,
    155(2):207–216.
  6. Holland, J. (1975). Adaptation in natural and artificial systems. University of Michigan Press, Ann Arbor.
  7. Klinger, V. (2015). Biosignal acquisition system for prosthesis control and rehabilitation monitoring. In Bruzzone, A., Frascio, M., Novak, V., Longo, F., Merkuryev, Y., and Novak, V., editors, 4nd International Workshop on Innovative Simulation for Health Care (IWISH 2015).
  8. Klinger, V. (2017). SMoBAICS: The Smart Modular Biosignal Acquisition and Identification System for Prosthesis Control and Rehabilitation Monitoring. International Journal of Privacy and Health Information Management (IJPHIM), 5(2).
  9. Klinger, V. (2018). An IoT-Based Platform for Rehabilitation Monitoring and Biosignal Identification. International Journal of Privacy and Health Information Management (IJPHIM), 6(1).
  10. Klinger, V. and Bohlmann, S. (2017). Closed-Loop Verification Approach for the Identification of Motion-Based Action Potentials in Neural Bundles using a Continuous Symbiotic System. In Bruzzone, A., Frascio, M., Novak, V., Longo, F., Merkuryev, Y., and Novak, V., editors, 6nd International Workshop on Innovative Simulation for Health Care (IWISH 2017).
  11. Klinger, V. and Klauke, A. (2013). Identification of motion-based action potentials in neural bundles using an algorithm with multiagent technology. In Backfrieder, W., Frascio, M., Novak, V., Bruzzone, A., and Longo, F., editors, 2nd International Workshop on Innovative Simulation for Health Care (IWISH 2013).
  12. Klinger, V. and Klauke, A. (2015). Identification of motion-based action potentials in neural bundles. International Journal of Privacy and Health Information Management (IJPHIM), 3(1).
  13. Kok, J., Gonzalez, L. F., Walker, R. A., Gurnett, T., and Kelson, N. A. (2010). A synthesizable hardware evolutionary algorithm design for unmanned aerial system realtime path planning. In Wyeth, G. and Upcroft, B., editors, Australasian Conference on Robotics and Automation 2010 (ACRA 2010), Brisbane, QLD.
  14. Koza, J. R. (1992). Genetic programming: On the programming of computers by means of natural selection. MIT Press, Cambridge, MA.
  15. Meyer-Baese, U. (2013). Digital Signal Processing with Field Programmable Gate Arrays. Signals and Communication Technology. Springer Berlin Heidelberg.
  16. Navarro, X., Krueger, T. B., Lago, N., Micera, S., Stieglitz, T., and Dario, P. (2005). A critical review of interfaces with the peripheral nervous system for the control of neuroprostheses and hybrid bionic systems. Journal of the Peripheral Nervous System, 10(3):229–258.
  17. Neymotin, S., Lytton, W., Olypher, A., and Fenton, A. (2011). Measuring the quality of neuronal identification in ensemble recordings. J Neurosci, 31(45):16398–409.
  18. Rashid, H. (2017). Evaluation of hardware implementation of evolutionary algorithm. Project work, Technical University Hamburg-Harburg, Institute for Nano and Medical Electronics.
  19. Schmidt, M. and Lipson, H. (2009). Distilling free-form natural laws from experimental data. Science, 324(5923):81–85.