Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/de-gruyter/evaluation-of-emotiv-eeg-neuroheadset-8aJla3ycp0
References
References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.
- Publisher
- de Gruyter
- Copyright
- Copyright © 2015 by the
- ISSN
- 1895-9091
- eISSN
- 1896-530X
- DOI
- 10.1515/bams-2015-0026
- Publisher site
- See Article on Publisher Site
Abstract
Electroencephalography (EEG) has become more popular, and as a result, the market grows with new EEG products. The new EEG solutions offer higher mobility, easier application, and lower price. One of such devices that recently became popular is Emotiv EEG. It has been already tested in various applications concerning brain-computer interfaces, neuromarketing, language processing, and detection of the P-300 component, with a general result that it is capable of recording satisfying research data. However, no one has tested and described its usefulness in long-term research. This article presents experience from using Emotiv EEG in two research projects that involved 39 subjects for 22 sessions. Emotiv EEG has significant technical issues concerning the quality of its screw threads. Two complete and successful solutions to this problem are described. Keywords: computer-aided design (CAD) modelling; electroencephalography (EEG); Emotiv; three-dimensional (3D) printing. Introduction Emotiv EEG (Emotiv EPOC/EEG, Kwun Tong, Hong Kong) is a low-cost, multichannel, wireless electroencephalogram (EEG) designed for research. It internally samples EEG signals at a frequency of 2048 Hz, which are then downsampled to 128 Hz, and sends the data to a computer via Bluetooth. The headset utilizes 14 gold-plated sensors positioned in accordance to the 10-20 system [1]: AF3, AF4, F7, F3, F4, F8, FC5, FC6, T7, T8, P7, P3, P4, P8, O1, O2. Moistening of polyester felt pads and applying headset to patient's head take up to 10 min and do not require any exterior assistance. Emotiv has been already used in various researches concerning brain-computer interfaces (BCIs) [26], neuromarketing [7], language processing [8], detection of the P-300 component [9, 10], and virtual reality [11]. *Corresponding author: Grzegorz M. Wójcik, University of Economics and Innovation, Projektowa 4, 20-209 Lublin, Poland, E-mail: gmwojcik@wsei.lublin.pl Piotr Wierzgala and Anna Gajos: University of Economics and Innovation, Lublin, Poland In comparison with other medical/research devices such as the ANT system (Advanced Neuro Technology, ANT, Enschede, The Netherlands), ActiCap system (Brain Products, Munich, Germany), and traditional silver/silver chloride disc electrodes, Emotiv EEG performs worse but satisfies for non-clinical applications [2, 9, 12]. The plastic-based screw threads in the Emotiv headset may easily break if not used carefully [9]. From our experience, the problem is much more serious because screw threads break even when they are handled with care (Figures 1 and 2). Our series of EEG experiments utilized four Emotiv EEG headsets. Each of the 24 students participated in 10 sessions, and one session lasted about 45 min. Toward the end of the project, all (n=64) screw threads broke and were no longer suitable for use. We reported the problem to an Emotiv employee, who replied that the sensors are made of polycarbonate (with a small percentage of acrylonitrile butadiene styrene [ABS]), a high-quality plastic that is also used in bulletproof glass. The Emotiv employee added that very few plastics are better suited for this application and that screw threads break when handled roughly. We were advised to put as little force as needed to feel a slight click when the screw thread is locked in place. We have handled our screw threads exactly the way the company advised, but they still had a tendency to break. The most straightforward way to solve this problem is to order a new sensor pack. Unfortunately, the company's shipping policy is hardly acceptable to many customers. In case of accessories, orders with a minimum of two items for shipping to other countries, except the USA, is separately charged with transportation costs per item regardless of size and weight. Ordering a new set of sensors for every couple of research sessions is unacceptable for a device whose one of the main advantages is its low price. Sensors are consumable items and need replacement after some time, but such a fast consumption was unexpected. This might be a serious problem for many researchers because damaged screw threads keep falling out of the headset, which makes its application very inconvenient or even impossible. The goal of this article is to share with the community some techniques to extend the lifespan of the Emotiv equipment. With the growing popularity of Emotiv EEGs, it is important for other researchers to know issues inherent in their device of choice. The presented case is not the sole 212Wójcik et al.: Evaluation of Emotiv EEG These screw threads may be used as a replacement parts in an emergency. Our first approach was to create a screw thread with lathe and milling machines. A technical drawing was prepared and given to an engineer operating the machines. The lathe was used to cut the shape of a screw thread, whereas the milling machine was used to create the details. The second approach was to make a three-dimensional (3D) screw thread model and print it with a 3D printer. The model was created in a computer-aided design (CAD) system and printed with a BFB-3000 3D Touch printer. Researchers willing to reproduce the results presented in this article can use the wireframe projections in the appendix. The projections provide all the necessary measurements and were successfully used in both approaches presented. Figure 1:A set of damaged screw threads. Results Two types of materials were tested during the production process: polyamide PA-6 (Figure 3) and polioksymetylen POM-C (Figure 4). In the end, polyamide was used to make all the screw threads because it was more convenient to cut precisely. Two milling cutters with 0.6- and 1.0-mm diameters were used. Most parts of the screw thread were made with the 1.0-mm cutter. The 0.6-mm cutter was necessary to cut a small bump in a latch. The engineer needed approximately 30 min to create one screw thread. The quality of the printed screw thread was lower compared with the milled one. A single line of melted plastic was set to 0.125 mm, which is the highest precision for the BFB-3000 printer. Despite this, the printed screw thread (Figures 5 and 6) had noticeable irregularities in the area of the latch. The material used for printing was ABS. The printed screw thread was usable but tends to fall Figure 2:Undamaged latch (left, front only) and damaged latch (right). instance because there are other researchers [9] and users [13] who either noticed similar problems or experienced them. Although it is possible to keep ordering new screw threads from the manufacturer, it may be unacceptable for some projects because of cumulative costs and time needed for shipment. This article provides the best practices and proven solutions to handle Emotiv EEG screw threads issues. Materials and methods To ensure a longer life for the screw threads, they can be left fitted to a headset. It will decrease the possibility of damaging the screw thread. In such case, only the felt pads should be pulled out and stored in the hydrator pack. This will additionally reduce the amount of time the saline is in contact with the sensor plate, which causes oxidization. Each headset has two rubber comfort pads, which are assembled with the same screw threads as the electrodes. Figure 3:Screw thread made from polyamide PA-6. Wójcik et al.: Evaluation of Emotiv EEG213 Figure 4:Screw thread made from polioksymetylen POM-C. Figure 5:Screw thread 3D CAD model. Payette, ID, USA), ActiveTwo Analog Input Box (BioSemi, Amsterdam, the Netherlands), Easy II EEG PSG Machine (Cadwell, Kennewick, WA, USA), 2EB Clinical System (BrainMaster Technologies, Bedford, OH, USA), and selected TrueScan32 as the most attractive option with respect to their criteria. Other interesting options are Atlantis I, Atlantis II, Discovery 24E (BrainMaster Technologies), Mitsar EEG 201 System, Mitsar 202 24, Mitsar 202 32 (Mitsar, St. Petersburg, Russia), and OpenBCI. Owing to better properties, the screw threads created by the engineer were used in the second EEG project involving 15 subjects for 12 sessions, 30 min each. In contrast to the original screw threads used in the first project, none of the self-made screw threads were damaged or had signs of wear at the end. Although the 3D printing method was not chosen, it seems to be very promising, especially with increasing precision of the printing process. Within 2 months, we have managed to produce significantly better and cheaper screw threads then the original ones, which might indicate that Emotiv sensors are meant to have limited useful life. Planned obsolescence puts consumer under pressure to purchase again in order to continue use of the product, and it might be especially beneficial for a company if there is no competing manufacturer on the market. The advantages of Emotiv EEG are its low price, mobility, and short setup time. Unfortunately, the increasing popularity of Emotiv EEG seems to be undermined by its low quality. Although Emotiv claims to use high-quality plastic for their screw threads, our experience shows otherwise. Similar observations have been made by other researchers [9] and Emotiv forum users [13]. The signal from the Emotiv headset was also noted to degrade significantly between the second and third hours of use and that the headset may cause discomfort after 1 h of use [12]. Researchers who have chosen Emotiv and are willing to extend the lifespan of the original sensors should be careful not to screw them out from the headset between experiments. Because it is possible to damage a couple of screw threads during one experiment session, it may be a good idea to order Emotiv headset with at least one extra hydrator pack or to self-produce a set of replacement screw threads before starting a research. Figure 6:Screw thread printed with BFB-3000 3D Touch. out of a socket because of insufficient printing precision in the area of the latch. There are several alternative low-cost EEGs available on the market. Portelli et al. [14] compared the cost per channel of four EEGs, TrueScan32 (DeyMed Diagnostics, Discussion and conclusions BCIs may become more popular with many prospective applications [1517], thanks to the cheap EEG systems appearing on the market [18]. Despite the fact that Emotiv EEG was handled with care, toward the end of 214Wójcik et al.: Evaluation of Emotiv EEG the neurofeedback experiment, all the screw threads used to fit sensors in a headset were damaged and could not used. Two solutions concerning self-made replacement parts of defective elements have been proposed. Screw threads made with lathe and milling machines proved to be very robust and cost-effective. They were tested in similar conditions as the screw threads provided with the headsets. After the project ended, all the self-made screw threads were still usable and had no signs of wear. Acknowledgments: This research has been supported by University of Economics and Innovation in Lublin, Poland (grant no. 12/OP/4/2013/). The authors would like to thank Marcin Smolira, Krzysztof Dmitruk, and Ewa Boguciska for their support and assistance. Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission. Research funding: None declared. Employment or leadership: None declared. Honorarium: None declared. Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication. Appendix 16.90 mm 14.50 mm AA 21° 11.00 mm 34° 40° 124° 124° 1.00 mm 80° 2.20 mm 2.40 mm 5.80 mm R0 mm R0 mm 3.99 mm 2.26 mm 3.53 mm 12.50 mm B BB 16.90 mm R 0. 57 4.67 mm 11.00 mm 0.07 mm 1.18 mm 6.00 mm 4.40 mm 2.26 mm 5.80 mm 3.44 mm 0.29 mm 2.34 mm Wireframe projections for CAD 3D printout solutions. 0.60 mm 0.75 mm 34° 21° 0.88 mm 40 ° R0 .57 mm Wójcik et al.: Evaluation of Emotiv EEG215
Journal
Bio-Algorithms and Med-Systems – de Gruyter
Published: Dec 1, 2015