Search

In 2019, the world had been attacked with a severe situation by the new version of the SARSCOV- 2 virus, which is later called COVID-19. One can use artificial intelligence techniques to reduce time consumption and find safe solutions that have the ability to handle huge amounts of data. However, in this article, we investigated the classification performance of eight deep transfer learning methodologies involved (GoogleNet, AlexNet, VGG16, MobileNet-V2, ResNet50, DenseNet201, ResNet18, and Xception). For this purpose, we applied two types of radiographs (X-ray and CT scan) datasets with two different classes: non-COVID and COVID-19. The models are assessed by using seven types of evaluation metrics, including accuracy, sensitivity, specificity, negative predictive value (NPV), F1- score, and Matthew\"s correlation coefficient (MCC). The accuracy achieved by the X-ray was 99.3%, and the evaluation metrics that were measured above were (98.8%, 99.6%, 99.6%, 99.0%, 99.2%, and 98.5%), respectively. Meanwhile, the CT scan model classified the images without error. Our results showed a remarkable achievement compared with the most recent papers published in the literature. To conclude, throughout this study, it has been shown that the perfect classification of the radiographic lung images affected by COVID- 19.

Modern teknoloji ile birlikte insanların göz hareketlerini inceleyerek uyaranlara karşı vermiş oldukları tepkiler takip edilebilir hale gelmiştir. Bu takip yöntemlerden biri de Göz İzleme (Eye-Tracking) tekniğidir. Bu teknikteki gelişmeler sayesinde araştırmacılar, sağlık, savaş sanayi, sivil havacılık, web tasarımı, dijital medya vb. birçok alanda hayatı daha kolay hale getirilebilecek sistemler hakkında çalışmalar yapabilmektedir. Bu çalışma kapsamında, göz izleme teknolojisinin temel özelliklerinden faydalanılarak beyin bilgisayar arayüzü (BBA) uygulaması geliştirilmiştir. Katılımcıların göz sabitlenme bilgisi, tarafımızca hazırlanan deneysel paradigma yazılımları bünyesinde göz-izleme cihazı ile ölçülerek, kişilerden verilen ödevleri gerçekleştirmeleri istenmiştir. Bu kapsamda iki farklı uyaran yazılımı üretilmiştir. Birinci yazılımda, kişilerin ekranda çeşitli bölgelerde görülen butonlara odaklanarak, gözlerinin sabitlenmesi ile butonlara basmaları sağlanmıştır. İkinci yazılımda ise, katılımcının harflere odaklanması istenerek, kelimeler ve cümleleri yazdırmayı sağlayan sanal bir klavye uygulaması geliştirilmiştir. Ayrıca göz fiksasyonları ısı haritası ile görselleştirilmiştir. Tüm aşamalarda kullanılan yazılım ve analiz tarafımızca geliştirilmiştir. Sonuç olarak, hareket etmeden göz hareketleri ile bildirim yapmayı sağlayan hibrid bir sistem geliştirilmiştir. Göz hareketlerine dayalı önerilen BBA sistemi test edilmiş ve yüksek komut/dakika sonuçlarına ulaşılmıştır. Deneysel bulgular önerilen hibrid BBA’nın güçlü ve öne çıkacak bir teknoloji olduğunu göstermektedir. With the help of modern technology, people's reactions to stimuli by examining the eye movements have become traceable. One of these measurement methods is the Eye Tracking technique. The technical advancements of this technique enable researchers to carry out studies in the fields of health, war industry, civil aviation, web design, digital media etc. that can enhance to improve the quality of the systems that can make life easier. In the concept of this study, a brain computer interface (BCI) is developed by using the principal properties of the eye tracking technology. Eye fixation information of the subjects were measured by using the in-house developed experimental paradigm software with the eye tracker while they were performing the required tasks. Two different experimental paradigm software were implemented. In the former one, subjects were asked to fixate to the buttons that appeared on the screen and they were clicked when the subjects fixated on those buttons. In the latter one, a virtual keyboard was implemented where the subjects were asked to fixate on the characters in order to write words or sentences. Additionally, eye fixations were plotted with the use of heat maps. All of the methods and tools were developed by our team. As a result, a hybrid BCI has been produced using the eye movements of subjects without performing a movement. The developed software tools were applied and high values of instruction per minute was obtained. Experimental results showed that the proposed methodology can be a pioneering technology.

2019 Scientific Meeting on Electrical-Electronics and Biomedical Engineering and Computer Science, EBBT 2019 -- 24 April 2019 through 26 April 2019, nofulltext# --- Atasoy, Mehmet Berkay (Arel Author), Birankar, Eyüp (Arel Author), Arıca, Şafak Abdullah (Arel Author), Duru, Dilek Göksel (Arel Author), In this study, it was aimed to classify the electrical signals recorded from human brain during different season (summer-winter) videos as stimuli. Data have been recorded using 14 channels EEG from four male participants. The power of delta, theta, alpha, beta and gamma frequency bands have been recorded and used to classify the collected data. Decision tree pre-processing method have been used to select the attributes of frequency bands and electrodes. To classify the data, support vector machines (SVM), linear discriminant analysis (LDA) and logistic regression (LR) machine learning algorithms have been used. It was found that it was separated %82.25 with SVM, %81 with LDA and %80.75 with LR. The results of three algorithms have shown similar scores. © 2019 IEEE.

Duru, Dilek Göksel (Arel Author), Cognitive state of a person can be monitored by the use of brain electrical activity measurements (Electroencephalogram, EEG). In the concept of this study, it is aimed to classify EEG topographies using deep learning. Among the cognitive test paradigms, Stroop test with four colors is used to collect EEG from two participants. P300 and N400 components are selected as two classes. P300 topography is computed using the average of EEG from 280 to 320 ms after the stimuli while 380 to 420 time window is used for N400 topographies. After the EEG artefact rejection processes, 440 topograph images were used to train the deep network. Randomly selected 10 images that were excluded from training set were used for testing. All of the test images were correctly classified while 73% of the training set images were correctly classified.

Işıkçı Koca, Elif (Arel Author), Duru, Dilek Göksel (Arel Author) --- Conference: Electric Electronics, Computer Science, Biomedical Engineerings' Meeting (EBBT) : İstanbul, 26-27 April 2016., Bu çalışmada, beyin dinlenim durumu ve mental performans gerektiren durumlar arasındaki geçişlerde gözlenen merkezi ve otonom sinir sistemi yanıtlarındaki farklılıkların fiziksel performansa bağıl bulguları hedeflenmiştir. Mental aritmetik işlemi ile oluşturulan mental iş yükünün, kontrollü fiziksel performans sonrasında, en yüksek oksijen tüketim hızı ve kardiyorespiratuvar dayanıklılık kriteri olan aerobik kapasitenin ölçümü (VO2max) ve eşzamanlı EEG (elektroensefalografi) ölçümleri ile amatör sporcu ve sedanter gruplar arasında kalp atım hızı değişkenliği, oksijen tüketimi miktarı ve mental aktivite üzerine olan etkisi incelenmiştir. Kontrollü sportif aktivite ve yorgunluk yaratabilmek ve etkisini niceliksel ölçmek amacıyla gruplara mekik koşu testi uygulanmıştır. Yaşları 20-24 arasında değişen 7 amatör sporcu ve 7 sedanter toplam 14 kişilik gönüllü katılımcıdan oluşan iki grup oluşturulmuştur. Ölçülen kalp atım değişkenliği verilerine Mann-Whitney U testi uygulanmıştır. Test sonuçları yüksek frekans göz kapalı dinlenim durumunun istatistiksel anlamlı olduğunu göstermektedir (p=0.04; p

18th National Biomedical Engineering Meeting (BIYOMUT) --OCT 16-17, 2014 -- Istanbul, TURKEY, WOS: 000381577500013, Duru, Dilek Göksel (Arel Author), Eye movements are essential for natural vision. Eye tracking technology is being used in research in many disciplines to examine the differences in the visual attention of experts and novices. Eye tracking research in sports focuses in the performance of athletes and its relation with perceptual processes. The aim of these studies relies in training the visual behavior of the athletes and to arrange trainings to increase their performance. In this study, the elite group is the Karate-Do players, and the aim is to extract their visual behavioral characteristics and patterns, to explore their gaze strategies, and the differences in their perception, attention and judgement. The experimental paradigm is a Kata and Bunkai video montage from World championships, where the elite athletes are asked to follow the techniques of the players, and the controls, which are interested in Karate-Do but not ever tried it, are asked to follow the movements of the players. During this time the eye tracking is done, and average fixation times, and dwell time in each area of interest have been calculated, and a significant difference between the two groups has been detected. The elite athletes have less fixation counts but longer fixation time compared to controls. So the professionals are not only extracting information from the center, but also from the peripheral areas. This finding is in agreement with the eye tracking literature. Future works will rely in investigating Karate-Do players from varying Karate-Do branches in groupwise comparison.

Although food consumption is one of the most basic human behaviors, the factors underlying nutritional preferences are not yet clear. The use of classification algorithms can clarify the understanding of these factors. This study was aimed at measuring electrophysiological responses to food/nonfood stimuli and applying classification techniques to discriminate the responses using a single-sweep dataset. Twenty-one right-handed male athletes with body mass index (BMI) levels between 18.5% and 25% (mean age: 21.05 ± 2.5 ) participated in this study voluntarily. The participants were asked to focus on the food and nonfood images that were randomly presented on the monitor without performing any motor task, and EEG data have been collected using a 16-channel amplifier with a sampling rate of 1024 Hz. The SensoMotoric Instruments (SMI) iView XTM RED eye tracking technology was used simultaneously with the EEG to measure the participants’ attention to the presented stimuli. Three datasets were generated using the amplitude, time-frequency decomposition, and time-frequency connectivity metrics of P300 and LPP components to separate food and nonfood stimuli. We have implemented k -nearest neighbor (kNN), support vector machine (SVM), Linear Discriminant Analysis (LDA), Logistic Regression (LR), Bayesian classifier, decision tree (DT), and Multilayer Perceptron (MLP) classifiers on these datasets. Finally, the response to food-related stimuli in the hunger state is discriminated from nonfood with an accuracy value close to 78% for each dataset. The results obtained in this study motivate us to employ classifier algorithms using the features obtained from single-trial measurements in amplitude and time-frequency space instead of applying more complex ones like connectivity metrics.