Yahya Taştan

Yahya Taştan

Auto-generated profile, based on publicly available information

Yahya Taştan

Marmara Üniversitesi

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
31 March 2021 in International Journal of Advances in Engineering and Pure Sciences
Verified
Journal Article
Closed access
PDF file
YTYahya Tastan
Yahya Tastan
HKHabib Kaymaz
Habib Kaymaz
Bu çalışmada, otonom araçların önünde bulunan zorluklar araştırılmıştır. Geçmişteki araç teknolojilerindeki deneyime dayanarak bu tür araçların ne kadar hızlı gelişip konuşlandırılacağı araştırılmıştır. Muhtemel zararları ve maliyetleri, sürüş faaliyetlerini nasıl etkileyecekleri, optimum yol mesafesi, park etme ve toplu taşıma arzı gibi planlama kararları üzerindeki etkileri üzerinde durulmuştur. Bu çalışmada Seviye 4 otonom araçlarının, yüksek maliyet ve sınırlı performansıyla birlikte 2020‘den sonra bazı ülkelerde yasal ve ticari araç olarak kullanılabileceği öngörülüyor. Seviye 5 otonom araçlar ise muhtemelen 2050'lerden 2060'lara kadar yaygın ve uygun fiyatlı hale geldiğinde önemli olacaktır. Bazı yollarda insan güdümlü araçların yasaklanmasını gerektirebilir. Birçok ulaşım ve şehir planlamacısı, araştırmacı ve siyasetçi araç otomasyonunun teknik yapısı ve faydasına aşina olsa da, sonuçta ortaya çıkacak etki veya sorumlu olanların eylemleri hakkında henüz fikir birliği görülmemektedir. Bu çalışma da, otonom araçlar, otonom araçlarda kullanılan sensör ve algoritmalar tanıtılmış, bu araçların gündelik yaşamımıza girmesine engel olan zorluklar anlatılmıştır.

30 July 2025 in IEEE Access
Verified
Journal Article
Open access
UBUlvi Baspinar
Ulvi Baspinar
YTYahya Taştan
Yahya Taştan
Brain-computer interfaces (BCIs) offer promising solutions for assisting individuals with disabilities, supporting neurorehabilitation, and enhancing human capabilities. However, the limited decoding accuracy of EEG-based motor imagery (MI) signals poses a major challenge for the practical deployment of BCI systems. A common approach involves using signals from opposite hemispheres to boost classification accuracy. Yet, to control devices such as robotic hands or prosthetics in a human-like manner, it is essential to accurately classify hand opening and closing tasks using EEG signals from the same motor cortex region. This study introduces TransformerNet, a novel deep learning architecture specifically designed to classify hand open-close MI tasks from the same brain region. The task is particularly challenging due to the high similarity and overlapping nature of the EEG signals. TransformerNet combines a convolutional module, inspired by EEGNet, to extract local spatial features, with a Transformer encoder that captures long-range temporal dependencies. Furthermore, a channel attention mechanism enhances the model’s ability to focus on the most informative features. In experimental evaluations, TransformerNet achieved an average classification accuracy of 85.97%, outperforming traditional deep learning methods. The model effectively captures high-level temporal-spectral patterns and uncovers hidden dependencies within the EEG signals. These results demonstrate the potential of integrating attention mechanisms with Transformer-based architectures to improve MI-based BCI performance. This advancement holds promise for real-world applications such as brain-controlled prosthetics, assistive devices, and human-computer interaction, moving BCI technologies closer to practical and reliable implementation.

2 May 2024
Verified
Preprint
Open access
PDF file
YTYahya Taştan
Yahya Taştan
UBUlvi Başpınar
Ulvi Başpınar
AHAhmet Hamurcu
Ahmet Hamurcu
ABAbdullah Bal
Abdullah Bal
BBBurcu Bulut
Burcu Bulut
GAGönül Acar
Gönül Acar
This study investigates the influence of immersive virtual reality environments and gamification on the classification of imaginary motor (MI) signals and the associated increase in energy in the motor cortex region for neurorehabilitation purposes. Two immersive virtual environments, indoor and outdoor, were selected, each with gamified and non-gamified scenarios. Event-Related Desynchronization (ERD) data underwent analyses to determine if there were significant differences in ERD levels between distinct age groups and whether Fully Immersive Virtual Reality (FIVR) environments induced notable energy increases. The initial analysis found no significant energy changes between age groups under constant environmental conditions. In the second analysis, FIVR environments did not lead to a statistically significant increase in cortical energy for the 21–24 age group (Group I). However, a notable difference in cortical energy increase was identified between gamified and non-gamified environments within the 32–43 age group (Group II). The study also explored the impact of environmental factors on MI signal classification using four deep learning algorithms. The Recurrent Neural Network (RNN) classifier exhibited the highest performance, with an average accuracy of 86.83%. Signals recorded indoors showed higher average classification performance, with a significant difference observed among age groups. Group I participants performed better in non-gamified environments (88.8%), while Group II achieved high performance indoors, especially in the gamified scenario (93.6%). Overall, the research underscores the potential of immersive virtual environments and gamification in enhancing MI signal classification and cortical energy increase, with age and environmental factors influencing the outcomes.