Sürdürülebilir Kalkınma Amaçları

• 5
• 3
• 1

• MakaleAçık Erişim2025Scopus SJRQ1/r/munzur/2291
  Seismic Foresight: A Novel Multi-Input 1D Convolutional Mixer Model for Earthquake Prediction Using Ionospheric Signals

  Uyanik, Hakan; Kokum, Mehmet; Senturk, Erman; Freeshah, Mohamed; Ozcelik, Salih T. A.; Halil Akpinar, Muhammed; Celik, Serenay; Sengur, Abdulkadir

  This study proposes a novel deep learning approach for predicting significant earthquakes (Mw ≥ 5.0 ) in Turkey using ionospheric Total Electron Content (TEC) data and space weather indices. ConvMixer is a lightweight CNN architecture that blends spatial and channel information using depthwise convolutions and pointwise layers. Inspired by vision transformers, it offers efficient image classification with fewer parameters and high performance. We developed a multi-input one-dimensional convolutional mixer (MI-1D-ConvMixer) model to classify TEC data from the preceding five consecutive days as either precursory to an earthquake on the 6th day or normal. The model incorporates six inputs: five 1D TEC signals and one 1D space-weather index array, including the global geomagnetic index (Kp), storm duration distribution (Dst), sunspot number (R), geomagnetic storm index (Ap-index), solar wind speed (Vsw), and solar activity index (F10.7) are also utilized to reveal non-seismic related pre-earthquake ionospheric variations. Our methodology involves two stages: 1) a preprocessing stage to enhance TEC signals and 2) an end-to-end training of the MI-1D-ConvMixer model. The model architecture features depth-wise and point-wise convolutions with patch embedding, utilizing uses four tunable hyperparameters: network depth, hidden dimension size, kernel size, and patch size. We used 196 earthquakes data from Turkey from 2010-2023, and TEC data from the TNPGN-Active GNSS stations. The dataset was split into 75% for training and 25% for testing. Performance metrics, including classification accuracy, sensitivity, specificity, and F1-score, are used for evaluation. Our model achieved a classification accuracy of 97.49%, demonstrating its potential for earthquake prediction systems. This research contributes to the field by introducing a novel deep learning architecture specifically designed for integrating TEC and space weather data for earthquake prediction. Future work should focus on validating the model’s performance in different geographical regions and investigating its limitations.

  • 
  AtıfScopus7·OpenAlex5

  İncele
• BildiriAçık Erişim2025/r/munzur/2172
  Toplam Elektron Içeriǧi ve Uzay Iklim Verileri Kullanilarak Transfer Öǧrenme ile Deprem Öncüllerinin Analizi

  Uyanik, Hakan; Şentürk, Erman; Köküm, Mehmet; Özçelik, Salih Taha Alperen; Akpinar, Muhammed Halil; Şengür, Abdulkadir

  The stress changes occurring in the Earth's crust before earthquakes may cause some anomalies in the ionosphere and cause significant changes in the Total Electron Content (TEC) and Space Climate (SCC) parameters. The analysis of these changes offers a significant potential in the detection of earthquake precursors. In this study, the analysis of earthquake precursors was performed using TEC and UC data and the effectiveness of transfer learning and machine learning techniques was investigated. Features were extracted from five-day TEC and UC data using a pre-trained Multi-Input One-Dimensional Convolutional Mixer Network (ÇG-TB-EMA) and these features were classified with different machine learning algorithms. Within the scope of the study, earthquakes with a magnitude of Mw≥6.0 that occurred during the low and high solar activity periods of the 24th solar cycle were analyzed. The earthquakes were selected from the Far East region between 10°S-45°N latitudes and 120°E-160°E longitudes, where intense seismic activity is observed. Experimental studies have shown that during the low solar activity period, support vector machines (SVM) and artificial neural networks (ANN) models exhibited the highest performance with 96.04% accuracy and 96.08% F1 score, respectively. During the high solar activity period, the classification success decreased slightly, and SVM and ANN models achieved the best results with 90.28% accuracy and 89.86% F1 score. These findings reveal that the effect of solar activity on ionospheric variables should be taken into account in earthquake precursor detection. For future studies, long-term analyses are recommended to better understand the effects of solar activity.

  • 
  İncele
• Derleme MakaleAçık Erişim2025Scopus SJRQ1/r/munzur/1517
  Automated Detection of Neurological and Mental Health Disorders Using EEG Signals and Artificial Intelligence: A Systematic Review

  Uyanik, Hakan; Sengur, Abdulkadir; Salvi, Massimo; Tan, Ru-San; Tan, Jen Hong; Acharya, U. Rajendra

  Mental and neurological disorders significantly impact global health. This systematic review examines the use of artificial intelligence (AI) techniques to automatically detect these conditions using electroencephalography (EEG) signals. Guided by Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA), we reviewed 74 carefully selected studies published between 2013 and August 2024 that used machine learning (ML), deep learning (DL), or both of these two methods to detect neurological and mental health disorders automatically using EEG signals. The most common and most prevalent neurological and mental health disorder types were sourced from major databases, including Scopus, Web of Science, Science Direct, PubMed, and IEEE Xplore. Epilepsy, depression, and Alzheimer's disease are the most studied conditions that meet our evaluation criteria, 32, 12, and 10 studies were identified on these topics, respectively. Conversely, the number of studies meeting our criteria regarding stress, schizophrenia, Parkinson's disease, and autism spectrum disorders was relatively more average: 6, 4, 3, and 3, respectively. The diseases that least met our evaluation conditions were one study each of seizure, stroke, anxiety diseases, and one study examining Alzheimer's disease and epilepsy together. Support Vector Machines (SVM) were most widely used in ML methods, while Convolutional Neural Networks (CNNs) dominated DL approaches. DL methods generally outperformed traditional ML, as they yielded higher performance using huge EEG data. We observed that the complex decision process during feature extraction from EEG signals in ML-based models significantly impacted results, while DL-based models handled this more efficiently. AI-based EEG analysis shows promise for automated detection of neurological and mental health conditions. Future research should focus on multi-disease studies, standardizing datasets, improving model interpretability, and developing clinical decision support systems to assist in the diagnosis and treatment of these disorders.

  • 
  AtıfScopus58·OpenAlex46

  İncele
• MakaleAçık Erişim2025Scopus SJRQ1/r/munzur/821
  A new lightweight convolutional neural network model for detecting drivable road regions

  Doğan, Gürkan; Uyanık, Hakan; Ergen, Burhan

  Nowadays, due to the rapid increase in the number of autonomous vehicles on the market, the safe navigation of these vehicles in drivable road areas has become extremely important. One of the most crucial factors in ensuring safe navigation is addressing the detection of drivable road areas as a task of semantic segmentation. Considering that autonomous vehicles are modular, the algorithm to perform this task must have the optimum trade-off in terms of lightweight, computational complexity, and segmentation accuracy. In this study, RoNet, a new model based on convolutional neural networks that provides an optimum trade-off for the detection of drivable road regions, was designed and proposed. The standard convolution types for the encoder and decoder bottleneck module of the RoNet model, as well as the spatial edge attention mechanism, have been optimized by developing asymmetric convolution types using asymmetric atrous convolution, asymmetric convolution types using Prewitt and Sobel kernels. Spatial edge attention mechanism is designed to reduce the loss of detailed information in small-resolution feature maps. In experimental tests performed with CamVid and FUVid datasets, RoNet achieved a better trade-off in terms of segmentation accuracy, number of parameters, and computational complexity compared to other state-of-the-art methods.

  • 
  • 
  AtıfScopus3·WoS3·OpenAlex6

  İncele
• MakaleAçık Erişim2024/r/munzur/6355
  Convmixer ve SDD Kullanılarak DEHB Hastalığının EEG Sinyalleri ile Otomatik Olarak Tespit Edilmesi

  Sengur, Abdulkadir; Üzen, Hüseyin; Özçelik, Salih Taha Alperen; Uyanik, Hakan; Karakaş, Buğra

  DEHB, çocuklarda dikkat eksikliği, davranış problemleri, eğitimle ilgili sorunlar ve düşük özgüven gibi problemler oluşturabilir. Bu çalışma, Dikkat Eksikliği Hiperaktivite Bozukluğu (DEHB) teşhisini elektroensefalografi (EEG) sinyalleriyle değerlendirmeyi hedefleyen bir araştırmayı özetlemektedir. Araştırma, 30 DEHB tanısı almış çocuk ve 30 sağlıklı kontrol grubunun EEG verilerini kullanmıştır. EEG verileri öncelikle gürültü azaltma amacıyla işlenmiş ve ardından ConvMixer, ResNet50 ve ResNet18 gibi derin öğrenme modelleri kullanılarak sınıflandırılmıştır. Bulgular, ConvMixer'in düşük hesaplama kaynaklarına ihtiyaç duyarak yüksek sınıflandırma başarısı elde ettiğini göstermektedir. Ayrıca, EEG sinyallerinin DEHB teşhisinde kullanılabilirliği konusunda farklı kanalların etkileri incelenmiş ve T8 kanalının özellikle etkili olduğu tespit edilmiştir. Bu çalışma, EEG tabanlı DEHB teşhisi için daha hafif modellerin kullanılabilirliğini ve EEG kanallarının önemini vurgulamaktadır.

  • 
  AtıfOpenAlex1

  İncele
• MakaleAçık Erişim2023/r/munzur/6456
  Balistokardiyografi sinyalleri ile derin öğrenme tabanlı hipertansiyon tespiti

  Sengur, Abdulkadir; Özçelik, Salih Taha Alperen; Uyanik, Hakan

  Kan basıncı, damarlardaki kanın damar duvarlarına uyguladığı basınçtır. Bu basınç değerinin normal kabul edilen seviyelerin üzerinde seyir etmesi yüksek tansiyon (YT) veya hipertansiyon (HPT) olarak bilinir. Hayat kalitesini negatif yönde etkileyen, çoğu zaman organlarda çeşitli tahribatlara sebep olan ve ölümlere yol açabilen bu sağlık probleminin teşhisi oldukça önemlidir. Bu çalışmada, balistokardiyografi (BKG) sinyalleri kullanılarak HPT'nin otomatik teşhisine yönelik bir yöntem önerilmiştir. Bunun için BKG sinyalleri, sürekli dalgacık dönüşümü filtre bankası (SDDFB) yöntemi kullanılarak zaman-frekans domenine taşınmıştır. Bu işlemler yapılırken kullanılan dönüşüm yönteminde bazı parametre ayarları gerçekleştirilerek dönüşümün kalitesi arttırılmıştır. Daha sonra elde edilen görüntüler ResNet18, ResNet50, VGG16 ve AlexNet evrişimsel sinir ağlarıyla sınıflandırılmış ve elde edilen sonuçlar karşılaştırılmıştır. Önerilen yöntem ile ResNet18, ResNet50, VGG16 ve AlexNet mimarileri için sırasıyla %98,92, %99,34 ve %99,22 ve %98,07 sınıflandırma doğruluğu elde etmiştir. Elde edilen bu yüksek sınıflandırma sonuçları önerilen yöntemin hipertansiyon teşhisi için kullanılabileceğini ispatlar niteliktedir.

  • 
  İncele
• MakaleAçık Erişim2023/r/munzur/6376
  Bir Boyutlu Evrişimsel Sinir Ağı Yardımıyla Faz Kilitleme Değeri ve Diferansiyel Entropi Özellikleri Kullanılarak EEG Sinyallerinde Duygu Tanınması

  Sengur, Abdulkadir; Özçelik, Salih Taha Alperen; Uyanik, Hakan

  Duygu analizi günümüz bilim dünyasında üzerinde en çok araştırma yapılan alanların başında gelmektedir. Özellikle insan-bilgisayar etkileşimi gibi günlük hayatımıza her geçen gün daha çok dahil olan alanların yanı sıra nörobilim ve psikoloji gibi bilim dallarının da yakından incelediği bir konudur. Duygu analizi için konuşma sinyalleri, mimikler, vücut dili, yüz ifadeleri gibi yöntemler kullanılsa da bu yöntemler manipülasyona açık oldukları için biyolojik sinyaller kadar güvenilir sonuçlar vermezler. Bu çalışmada sanal gerçeklik (SG) teknolojisi yardımıyla hazırlanmış, biyoelektriksel bir sinyal olan elektroansefalografi (EEG) sinyalleri ile duygu tanıma için yeni bir yöntem önerilmiştir. Bu yöntemde EEG sinyallerinin alt bantlarının diferansiyel entropi (DE) ve faz kilitleme değeri (FKD) özellikleri, tasarlanan bir boyutlu evrişimsel sinir ağı (1B-ESA) yardımı ile pozitif ve negatif duyguların tanınması için kullanılmıştır. Her iki özellik yardımıyla elde edilen özellik matrisleri on defa teste tâbi tutularak ortalama başarı değerleri elde edilmiştir. Bu testler sonucunda DE ve FKD özellikleri ile en yüksek ortalama başarı puanları, tüm alt bant özellik matrislerinin birleştirilmesi ile sırasıyla %74,0611,41 ve %63,75901,72 olarak elde edilmiştir. Ayrıca çalışmada elde edilen yüksek frekanstaki sinyal bileşenlerine ait testlerin başarı oranlarının düşük frekans bantlarına göre daha yüksek elde edilmesi daha önce bu alanda yapılan benzer çalışmaların sonuçlarını destekler nitelikte olmuştur.

  • 
  AtıfOpenAlex2·TR Dizin1

  İncele
• MakaleAçık Erişim2023Scopus SJRQ2/r/munzur/2818
  Automated Hypertension Detection Using ConvMixer and Spectrogram Techniques with Ballistocardiograph Signals

  Ozcelik, Salih T. A.; Uyanık, Hakan; Deniz, Erkan; Sengur, Abdulkadir

  Blood pressure is the pressure exerted by the blood in the veins against the walls of the veins. If this value is above normal levels, it is known as high blood pressure (HBP) or hypertension (HPT). This health problem which often referred to as the “silent killer” reduces the quality of life and causes severe damage to many body parts in various ways. Besides, its mortality rate is very high. Hence, rapid and effective diagnosis of this health problem is crucial. In this study, an automatic diagnosis of HPT has been proposed using ballistocardiography (BCG) signals. The BCG signals were transformed to the time-frequency domain using the spectrogram method. While creating the spectrogram images, parameters such as window type, window length, overlapping rate, and fast Fourier transform size were adjusted. Then, these images were classified using ConvMixer architecture, similar to vision transformers (ViT) and multi-layer perceptron (MLP)-mixer structures, which have attracted a lot of attention. Its performance was compared with classical architectures such as ResNet18 and ResNet50. The results obtained showed that the ConvMixer structure gave very successful results and a very short operation time. Our proposed model has obtained an accuracy of 98.14%, 98.79%, and 97.69% for the ResNet18, ResNet50, and ConvMixer architectures, respectively. In addition, it has been observed that the processing time of the ConvMixer architecture is relatively short compared to these two architectures.

  • 
  AtıfScopus21·WoS16·OpenAlex21

  İncele
• MakaleAçık Erişim2023Scopus SJRQ1/r/munzur/1761
  A Multi-Input Convolutional Neural Networks Model for Earthquake Precursor Detection Based on Ionospheric Total Electron Content

  Uyanık, Hakan; Şentürk, Erman; Akpınar, Muhammed Halil; Ozcelik, Salih T. A.; Kokum, Mehmet; Freeshah, Mohamed; Sengur, Abdulkadir

  Earthquakes occur all around the world, causing varying degrees of damage and destruction. Earthquakes are by their very nature a sudden phenomenon and predicting them with a precise time range is difficult. Some phenomena may be indicators of physical conditions favorable for large earthquakes (e.g., the ionospheric Total Electron Content (TEC)). The TEC is an important parameter used to detect pre-earthquake changes by measuring ionospheric disturbances and space weather indices, such as the global geomagnetic index (Kp), the storm duration distribution (Dst), the sunspot number (R), the geomagnetic storm index (Ap-index), the solar wind speed (Vsw), and the solar activity index (F10.7), have also been used to detect pre-earthquake ionospheric changes. In this study, the feasibility of the 6th-day earthquake prediction by the deep neural network technique using the previous five consecutive days is investigated. For this purpose, a two-staged approach is developed. In the first stage, various preprocessing steps, namely TEC signal improvement and time-frequency representation-based TEC image construction, are performed. In the second stage, a multi-input convolutional neural network (CNN) model is designed and trained in an end-to-end fashion. This multi-input CNN model has a total of six inputs, and five of the inputs are designed as 2D and the sixth is a 1D vector. The 2D inputs to the multi-input CNN model are TEC images and the vector input is concatenated space weather indices. The network branches with the 2D inputs contain convolution, batch normalization, and Rectified Linear Unit (ReLU) activation layers, and the branch with the 1D input contains a ReLU activation layer. The ReLU activation outputs of all the branches are flattened and then concatenated. And the classification is performed via fully connected, softmax, and classification layers, respectively. In the experimental work, earthquakes with a magnitude of Mw5.0 and above that occurred in Turkey between 2012 and 2019 are used as the dataset. The TEC data were recorded by the Turkey National Permanent GNSS Network-Active (TNPGN-Active) Global Navigation Satellite System (GNSS) stations. The TEC data five days before the earthquake were marked as “precursor days” and the TEC data five days after the earthquake were marked as “normal days”. In total, 75% of the dataset is used to train the proposed method and 25% of the dataset is used for testing. The classification accuracy, sensitivity, specificity, and F1-score values are obtained for performance evaluations. The results are promising, and an 89.31% classification accuracy is obtained.

  AtıfScopus27·OpenAlex19

  İncele
• MakaleAçık Erişim2023Scopus SJRQ3/r/munzur/1635
  Effect of Spectrogram Parameters and Noise Types on The Performance of Spectro-temporal Peaks Based Audio Search Method

  Koseoglu, Murat; Uyanik, Hakan

  Audio search algorithms are used to detect the queried file in large databases, especially in multimedia applications. These algorithms are expected to perform the detection in a reliable and robust way within the shortest time. In this study, based on spectral peaks method, an audio fingerprint algorithm with a few minor modifications was developed to detect the matching audio file in target database. This method has two stages as the audio fingerprint extraction and matching. In the first stage, fingerprint features are extracted from spectral peaks on the spectrograms of audio files by hash functions. This state-of-art technique reduces the processing load and time considerably compared to traditional methods. In the second stage, fingerprint data of the queried file are compared with the data created in the first stage in the database. The algorithm was demonstrated, and the effect of spectrogram parameters (window size, overlap, number of FFT) was investigated by considering reliability and robustness under different noise sources. Also, it was aimed to contribute to new audio retrieval studies based on spectral peaks method. It was observed that the variation in the spectrogram parameters significantly affected the number of matchings, reliability and robustness. Under high noise conditions, the optimal spectrogram parameters were determined as 512 (window size), 50% (overlap), 512 (number of FFT). It was seen in general that the algorithm successfully detected the queried file in the database even in high noise conditions for these parameters. No significant effect of music genre was observed.

  AtıfScopus9·WoS6·OpenAlex11·TR Dizin1

  İncele