gms | German Medical Science

Artificial Vision 2013

The International Symposium on Visual Prosthetics

08.11. - 09.11.2013, Aachen

Image processing and neural stimulation methods developed for epiretinal implant systems

Meeting Abstract

Search Medline for

  • Irfan Karagoz - Electrical and Electronics Engineering Department, Gazi University, Ankara, Turkey
  • G. Sobaci - Ophthalmology Department, GATA Hospital, Ankara, Turkey
  • M. Ozden - Electrical and Electronics Engineering, Kirikkale University, Kirikkale, Turkey

Artificial Vision 2013. Aachen, 08.-09.11.2013. Düsseldorf: German Medical Science GMS Publishing House; 2014. Doc13artvis10

doi: 10.3205/13artvis10, urn:nbn:de:0183-13artvis101

Published: February 13, 2014

© 2014 Karagoz et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by-nc-nd/3.0/deed.en). You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.


Outline

Text

Objective: To develop image processing and neural stimulation methods to improve visual performance of current retina implant systems.

Materials and Methods: To overcome low spatial and temporal resolution problem, in our study, artificial retina model which has 3D two stage local-Adaptive DOG filter based spatio-temporal filtering structure (3D-ADOG) and Spatio-Temporal Electrode Mapping and Local Interleaved Stimulation (STEMLIS) methods are developed. While 3D-ADOG encodes images as like a retina and produce artificial spike activity, STEMLIS method is responsible for interaction-free stimulation of micro-electrode matrix. Performance of the methods is evaluated by real time image reconstruction methods used in DaVinciTM digital video processor module with LCD glasses and phosphene based simulations. Beside simulation with normal seeing peoples, quantitative results are also obtained. For visual evaluation tests, 20 normal seeing peoples ages between 20 and 40 (27,9±5,64) were selected.

Results: For image reconstruction based, visually conducted contrast sensitivity, object counting, and pattern recognition tests, 3D-ADOG model respectively yielded higher scores than classical DOG based model, %11.4, %12.5, %9.8 and %73.7. For the same test set based on phosphene representation, STEMLIS method yielded higher scores than classical stimulation method as %10.5 , %16.1, %14.1 respectively.

Discussion: By considering these simulation results, it has been concluded that the 3D-ADOG model and the STEMLIS method can contribute to sight restoration studies.

Acknowledgement: This work was supported by a grant from TUBITAK (110E077)