Topographical Prominence Method for Suppressing Electrical Stimulus Artifact
-
1
Chungbuk National University School of Medicine, Department of Physiology, Korea
-
2
University of Ulsan, Department of Biomedical Engineering, Korea
Motivation: Retinal prosthesis aims at restoring vision by means of electrical stimulation of remaining inner retinal neurons. Electrical stimulation through the prosthesis elicits RGC spikes directly and/or indirectly via synapses, therefore transmits visual information to the brain. The directly-evoked RGC spikes which are fast in time, tend to coincide with electrical stimulus artifact, and then are obscured by stimulus artifact. Previously, we suggest moving average filter (MAF) and forward and reverse (FR) filter algorithms for artifact subtraction methods in addition to SALPA, the most well-known and generally used method [1]. Here, we apply topographic prominence concept which is used to evaluate relative heights of mountain in field of geography and geology to our newly-proposed MAF and FR filter algorithms [2].
Material and Methods:
1) Ex-vivo recording of retina and electrical stimulation with MEA
C3H/HeJ strains (rd1 mice) at postnatal week 10 and higher were used for the retinal degeneration model. Extracted mouse retina (¡3 ¡¿ 3 mm2) was placed RGC layer down on the surface of the multi-electrode array (MEA; 8 ¡¿ 8 grid layout, electrode diameters of 30 ¥ìm, inter-electrode distances of 200 ¥ìm). Electrical stimulus was applied via one electrode and all other electrodes were used for recording RGC responses. Fifty cathodic phase-1st biphasic current pulses (duration 500 ¥ìs, amplitude 30 ¥ìA) were applied at every 1 sec. Cadmium chloride (CdCl2; 20 ¥ìM), a synapse blocker-applied data were used for this paper to pool directly-evoked RGC spikes, exclusively.
2) Artifact subtraction algorithms using topographic prominence concept
Topographic prominence discerns whether a peak is the spike or not using the width at the half of height of prominence (Fig. 1). When a peak has width over 0.4 ms, it is regarded as stimulus artifact, since the depolarization phase in normal spikes is usually within the range of 0.4 ms. We applied topographic prominence concept to our MAF and FR filter algorithms [2]. SALPA algorithm was used as a reference to investigate the efficacy of MAF and FR filter algorithms.
3) Comparison of three algorithms: False positive error and False negative error
To compare three artifact subtraction algorithms, two parameters - false positive error and false negative error - were used. Spikes within 4 ms window after stimulus onset were regarded as false positive spikes because in our data, directly-evoked spikes usually appear at 4 ms after stimulus. False negative error was calculated by comparing the latency of first spikes acquired from each pair of different algorithms.
Results: The MAF and FR filter algorithms based on topographic prominence concept were optimized by controlling prominence parameters (Fig. 2). After optimization, statistical analysis was performed in 108 RGCs (n= 4 patches). Each false positive error rate was 3.3 ¡¾ 0.8 %, 5.3 ¡¾ 1.5 % and 1.2 ¡¾ 0.7 % in SALPA, MAF, and FR filter, respectively. There was a significant difference between FR filter and two other algorithms (p<0.05). Each pair of false negative error rate was 16.8 ¡¾ 1.4 %, 20.9 ¡¾ 1.6 % (SALPA & MAF), and 17.1 ¡¾ 1.5 %, 22.7 ¡¾ 1.8 % (SALPA & FR filter). There was a significant difference between SALPA and FR filter (p<0.05).
Discussion & Conclusion: In terms of false positive error, FR filter algorithm shows better performance than two other algorithms. In terms of false negative error, SALPA shows better performance than FR filter. Accordingly, we suggest that MAF and FR filter algorithms based on topographic prominence concept could play a complementary role for SALPA to detect directly-evoked RGC spikes.
References
[1] Wagenaar DA, Potter SM. Real-time multi-channel stimulus artifact suppression by local curve fitting. J Neurosci Methods. 2002; 120(2): 113-120.
[2] Choi M, Ahn J, Oh S, Pi K, Chee Y, Ko H et al. Comparison of the Three Filter Algorithms for Detection of Electrically-Evoked Short-Latency Responses in Retinal Ganglion Cells. IFMBE Proceedings. 2015; 51: 970-973.
Acknowledgements
Supported by grants of the MEST (NRF-2010-0020852, NRF-2013R1A1A3009574) of Korea to YSG.
Keywords:
Retinal Ganglion Cells,
Directly- evoked spike,
Electrical stimulus artifact,
Topographical prominence
Conference:
MEA Meeting 2016 |
10th International Meeting on Substrate-Integrated Electrode Arrays, Reutlingen, Germany, 28 Jun - 1 Jul, 2016.
Presentation Type:
Poster Presentation
Topic:
MEA Meeting 2016
Citation:
Ahn
JR,
Choi
M,
Park
DJ,
Koo
K and
Goo
YS
(2016). Topographical Prominence Method for Suppressing Electrical Stimulus Artifact.
Front. Neurosci.
Conference Abstract:
MEA Meeting 2016 |
10th International Meeting on Substrate-Integrated Electrode Arrays.
doi: 10.3389/conf.fnins.2016.93.00103
Copyright:
The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers.
They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.
The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.
Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.
For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.
Received:
22 Jun 2016;
Published Online:
24 Jun 2016.
*
Correspondence:
Dr. Yong S Goo, Chungbuk National University School of Medicine, Department of Physiology, Cheongju, Korea, ysgoo@chungbuk.ac.kr