01. Araştırma Çıktıları | TR-Dizin | WoS | Scopus | PubMed

Permanent URI for this communityhttps://hdl.handle.net/20.500.14365/1

Browse

Now showing 1 - 20 of 4460

1,25-Dihydroxyvitamin D3 Induces N-Myc Downstream Regulated Gene-2 Expression in Papillary Thyroid Carcinoma Cells
(Dokuz Eylul Univ Inst Health Sciences, 2020) Sipahi, Murat; Bartik, Didem Keles; Doruk, Mehmet; Bayraktar, Firat; Oktay, Gulgun
Purpose: In addition to its role in serum calcium homeostasis, the anti-tumor function of 1,25-dihydroxyvitamin D-3 (calcitriol) in cancer development is well established. N-myc Downstream Regulated Gene 2 which functions as a tumor suppressor gene has recently been shown to be downregulated in various cancer leading to increased tumor incidence, progression and metastasis. The goal of this study was to investigate the possible effects of calcitriol treatment on NDRG2 expression in BCPAP papillary thyroid carcinoma cells. Methods: The experiments were carried on human primary thyroid follicular epithelial cells (Nthy-ori-3-1), and human papillary thyroid carcinoma cells (BCPAP). The half maximal inhibitory concentration (IC 50) of calcitriol on BCPAP cells was determined by WST-1 assay. BCPAP cells were treated with 15 and 30 mu M calcitriol for 24, 48, and 72 hours, respectively. Basal NDGR2 expression in Nthy-ori-3-1 and BCPAP cells as well as the alterations on NDRG2 expression in calcitriol treated BCPAP cells were evaluated with western blot. Results: A significant downregulation of NDRG2 was observed in BCPAP cells when compared to Nthy-ori-3-1 cells (p<0.01). IC50 dose of calcitriol was found to be 64, 54 and 43 mu M for 24, 48 and 72 hours, respectively. NDRG2 protein expression levels were significantly increased in 30 mu M calcitriol treated BCPAP cells after 48 hours (p<0.05). Conclusions: Calcitriol induced NDRG2 protein expression in BCPAP cells. We predict that calcitriol increased NDRG2 protein levels in BCPAP cells via c-Myc repression, which is upregulated by aberrant Wnt/beta-catenin signaling. Further investigation is required to enlighten the possible effect mechanisms of calcitriol in BCPAP cells.
Citation - WoS: 238
1-D Convolutional Neural Networks for Signal Processing Applications
(IEEE, 2019) Kiranyaz, Serkan; İnce, Türker; Abdeljaber, Osama; Avcı, Onur; Gabbouj, Moncef
1D Convolutional Neural Networks (CNNs) have recently become the state-of-the-art technique for crucial signal processing applications such as patient-specific ECG classification, structural health monitoring, anomaly detection in power electronics circuitry and motor-fault detection. This is an expected outcome as there are numerous advantages of using an adaptive and compact 1D CNN instead of a conventional (2D) deep counterparts. First of all, compact 1D CNNs can be efficiently trained with a limited dataset of 1D signals while the 2D deep CNNs, besides requiring 1D to 2D data transformation, usually need datasets with massive size, e.g., in the Big Data scale in order to prevent the well-known overfitting problem. 1D CNNs can directly be applied to the raw signal (e.g., current, voltage, vibration, etc.) without requiring any pre-or postprocessing such as feature extraction, selection, dimension reduction, denoising, etc. Furthermore, due to the simple and compact configuration of such adaptive 1D CNNs that perform only linear 1D convolutions (scalar multiplications and additions), a real-time and low-cost hardware implementation is feasible. This paper reviews the major signal processing applications of compact 1D CNNs with a brief theoretical background. We will present their state-of-the-art performances and conclude with focusing on some major properties.
Citation - Scopus: 318
1-D Convolutional Neural Networks for Signal Processing Applications
(Institute of Electrical and Electronics Engineers Inc., 2019) Kiranyaz, Serkan; İnce, Türker; Abdeljaber, O.; Avci, O.; Gabbouj, M.
1D Convolutional Neural Networks (CNNs) have recently become the state-of-the-art technique for crucial signal processing applications such as patient-specific ECG classification, structural health monitoring, anomaly detection in power electronics circuitry and motor-fault detection. This is an expected outcome as there are numerous advantages of using an adaptive and compact 1D CNN instead of a conventional (2D) deep counterparts. First of all, compact 1D CNNs can be efficiently trained with a limited dataset of 1D signals while the 2D deep CNNs, besides requiring 1D to 2D data transformation, usually need datasets with massive size, e.g., in the »Big Data» scale in order to prevent the well-known »overfitting» problem. 1D CNNs can directly be applied to the raw signal (e.g., current, voltage, vibration, etc.) without requiring any pre- or post-processing such as feature extraction, selection, dimension reduction, denoising, etc. Furthermore, due to the simple and compact configuration of such adaptive 1D CNNs that perform only linear 1D convolutions (scalar multiplications and additions), a real-time and low-cost hardware implementation is feasible. This paper reviews the major signal processing applications of compact 1D CNNs with a brief theoretical background. We will present their state-of-the-art performances and conclude with focusing on some major properties. Keywords - 1-D CNNs, Biomedical Signal Processing, SHM. © 2019 IEEE.
Citation - WoS: 1893
Citation - Scopus: 2230
1d Convolutional Neural Networks and Applications: a Survey
(Academic Press Ltd- Elsevier Science Ltd, 2021) Kiranyaz, Serkan; Avcı, Onur; Abdeljaber, Osama; İnce, Türker; Gabbouj, Moncef; Inman, Daniel J.
During the last decade, Convolutional Neural Networks (CNNs) have become the de facto standard for various Computer Vision and Machine Learning operations. CNNs are feed-forward Artificial Neural Networks (ANNs) with alternating convolutional and subsampling layers. Deep 2D CNNs with many hidden layers and millions of parameters have the ability to learn complex objects and patterns providing that they can be trained on a massive size visual database with ground-truth labels. With a proper training, this unique ability makes them the primary tool for various engineering applications for 2D signals such as images and video frames. Yet, this may not be a viable option in numerous applications over 1D signals especially when the training data is scarce or application specific. To address this issue, 1D CNNs have recently been proposed and immediately achieved the state-of-the-art performance levels in several applications such as personalized biomedical data classification and early diagnosis, structural health monitoring, anomaly detection and identification in power electronics and electrical motor fault detection. Another major advantage is that a real-time and low-cost hardware implementation is feasible due to the simple and compact configuration of 1D CNNs that perform only 1D convolutions (scalar multiplications and additions). This paper presents a comprehensive review of the general architecture and principals of 1D CNNs along with their major engineering applications, especially focused on the recent progress in this field. Their state-of-the-art performance is highlighted concluding with their unique properties. The benchmark datasets and the principal 1D CNN software used in those applications are also publicly shared in a dedicated website. While there has not been a paper on the review of 1D CNNs and its applications in the literature, this paper fulfills this gap. (C) 2020 The Author(s). Published by Elsevier Ltd.
A 2-Hop Coloring-Based Collision Free Infrastructure Design for Wireless Sensor Networks
(Institute of Electrical and Electronics Engineers Inc., 2016) Korkmaz I.; Dagdeviren O.; Dalkilic M.E.
This paper mainly proposes a design for a communication infrastructure for Wireless Sensor Networks. The proposed design prevents message collisions by arranging the time schedules to send, receive, forward and overhear packets of the nodes considering 2-hop graph coloring mechanism. The system aims to exclude the compromised nodes in the network using the overhearing mechanism, and copes with recovering the paths of the legitimate nodes using multipath redundancy. The proposed scheduling-based and overhearing supported infrastructure brings the advantage of providing the Sensor Networks with both reliable communication using backup paths and energy conservation by preventing the collisions. © 2016 IEEE.
2011-2013 Constitution-Making Process and Media Field in Turkey
(2019) Akın, Altuğ; Yaman Akyar, Burcu
This paper examines so-called “Constitutional Process” which took place between 2011 and 2013 in Turkey to eventually fail after more than two years of intense work and rise of expectations both in Turkish society and international community from the perspective of media and communications. How Turkish media influenced the constitution making process including workings of Constitutional Conciliation Commission is discussed, surveying relevant media content or performance with a particular focus on the political economy of Turkish media during that period. Traditional media outlets such as newspapers and television channels, and nowadays largely digital platforms, function as present-day agoras where political agenda is determined and contested. These online and offline channels have different levels of influence on discussions of the political elite, as presented extensively in media and communication studies. Together with the global media environment, national media foster the emergence of its own elites, while maintaining multi-layered relationships with high-level political decision-making processes. Media channels therefore bear symbolic powers that impact political policymaking by employing methods such as agenda setting and framing. In this line of thought, this article approaches Turkish media space as a specific field while its content and its political economy are examined in relation with the constitution making process to reveal the role played by media during highly politicized processes, such as constitution making, as well as eliciting the dynamics that herald such media performance.
Citation - WoS: 1
Citation - Scopus: 1
25-Hydroxyvitamin Levels in Sjögren’s Syndrome: Is It the Right Time to Dismiss the Case or Not
(Walter de Gruyter GmbH, 2024) Sımsır, Ilgın Yıldırım; Tanigor, Goksel; Karabulut, Gonca; Barutcuoglu, Burcu; Yılmaz, Zevcet
Objectives: This study aimed to investigate whether patients with primary Sjögren syndrome (SjS) have different levels of 25 OH-D3 (vitamin D) when compared to healthy populations and whether differences in 25 OH-D3 correlated with disease activity or markers. Methods: Eighty-eight female patients with SjS and 3,338 age-matched healthy female controls were included in this study. 25 OH-D3 levels were compared with healthy controls. Then the patients were stratified according to their 25 OH-D3 levels, either insufficient/deficient or normal (<50 nmol/L or ≥50 nmol/L). The disease activity was evaluated using The EULAR SjS disease activity index (ESSDAI) and its components. Correlation analyses were also performed for a possible correlation with disease characteristics and markers of activity. Results: No differences in 25 OH-D3 levels were found between SjS and healthy populations (p>0.05). No correla- tions were found between patient characteristics or labo- ratory values (p>0.05). Conclusions: This study did not find a link between disease characteristics and disease activity and 25 OH-D3 levels. Prospective studies with more patients should be conducted to reach a conclusion.
Citation - WoS: 1
Citation - Scopus: 2
2d Model of a Biomass Single Particle Pyrolysis-Analysis of the Influence of Fiber Orientation on the Thermal Decomposition Process
(Mdpi, 2025) Hercel, Paulina; Orhon, Atahan; Jozwik, Michal; Kardas, Dariusz
Understanding the influence of heat transfer on the pyrolysis process is crucial for optimizing industrial biofuel production processes. While numerous scientific studies focus on experimental investigations of pyrolysis using laboratory-scale devices, many neglect the essential role of thermal energy in initiating and controlling thermal decomposition processes. This study presents a transient two-dimensional numerical model of biomass single-particle pyrolysis, which includes the energy balance, mass conservation equations and pyrolysis gas pressure and velocity equations. The model employs explicit numerical methods to manage the high computational demands of 2D transient simulations, but is successfully validated with the use of experimental data found in the literature. The model reflects the heterogeneous structure of wood by using different thermal conductivity coefficients depending on the wooden fibers' orientation. The results demonstrate the impact of fiber orientation on the heat transfer and thermal decomposition processes. The anisotropic properties of wood led to varied temperature fields and pyrolysis decomposition stages, aligning well with experimental data, thus validating the model's accuracy. The proposed approach can provide a better understanding and lead to improvement in biofuel production processes, enabling more efficient and controlled conversion of biomass into fuel. By optimizing the pyrolysis process, it contributes to the development of sustainable energy preservation and regeneration methods, supporting a shift towards more sustainable fuel production patterns using renewable biomass resources like wood.
Citation - WoS: 6
Citation - Scopus: 7
2db: a Proteomics Database for Storage, Analysis, Presentation, and Retrieval of Information From Mass Spectrometric Experiments
(Biomed Central Ltd, 2008) Allmer, Jens; Kuhlgert, Sebastian; Hippler, Michael
Background: The amount of information stemming from proteomics experiments involving (multi dimensional) separation techniques, mass spectrometric analysis, and computational analysis is ever-increasing. Data from such an experimental workflow needs to be captured, related and analyzed. Biological experiments within this scope produce heterogenic data ranging from pictures of one or two-dimensional protein maps and spectra recorded by tandem mass spectrometry to text-based identifications made by algorithms which analyze these spectra. Additionally, peptide and corresponding protein information needs to be displayed. Results: In order to handle the large amount of data from computational processing of mass spectrometric experiments, automatic import scripts are available and the necessity for manual input to the database has been minimized. Information is in a generic format which abstracts from specific software tools typically used in such an experimental workflow. The software is therefore capable of storing and cross analysing results from many algorithms. A novel feature and a focus of this database is to facilitate protein identification by using peptides identified from mass spectrometry and link this information directly to respective protein maps. Additionally, our application employs spectral counting for quantitative presentation of the data. All information can be linked to hot spots on images to place the results into an experimental context. A summary of identified proteins, containing all relevant information per hot spot, is automatically generated, usually upon either a change in the underlying protein models or due to newly imported identifications. The supporting information for this report can be accessed in multiple ways using the user interface provided by the application. Conclusion: We present a proteomics database which aims to greatly reduce evaluation time of results from mass spectrometric experiments and enhance result quality by allowing consistent data handling. Import functionality, automatic protein detection, and summary creation act together to facilitate data analysis. In addition, supporting information for these findings is readily accessible via the graphical user interface provided. The database schema and the implementation, which can easily be installed on virtually any server, can be downloaded in the form of a compressed file from our project webpage.
Citation - WoS: 4
Citation - Scopus: 5
3d Helmholtz Coil System Setup for Thermal Conductivity Measurements of Magnetic Nanofluids
(Pergamon-Elsevier Science Ltd, 2023) Alsangur, Rahime; Dog, Serkan; Ateş, Ismet; Turgut, Alpaslan; Çetin, Levent
This study aims to design a mechatronic system that involves a 3D Helmholtz coil system implemented with the 3 omega; method to measure the thermal conductivity of magnetic nanofluids under uniform and rotating magnetic fields. For this purpose, a 3D Helmholtz coil system was designed and manufactured to generate a uniform and rotating magnetic field up to 400 G. First, the uniformity and rotation abilities of the magnetic field generated by the system were investigated numerically and experimentally. The investigations pointed out that the 3D Helmholtz coil system can generate a uniform magnetic field in 1D, 2D, and 3D with a maximum non-uniformity factor of 0.0016. After that, the thermal conductivity of Fe3O4 - water magnetic nanofluid samples with 1, 2, 3, 4, and 4.8 vol.% were measured under 1D, 2D, and 3D uniform magnetic field application. The magnetic field was applied at different direction angles between X, Y, and Z axes in the Cartesian coordinate system. The results pointed out that the thermal conductivity of the samples increases as the magnetic field and particle concentration increase. The maximum thermal conductivity enhancement was observed as similar to 9.1% and the minimum thermal conductivity was observed as similar to 1.9% when the magnetic field is applied in parallel and perpendicular directions, respectively. The measurement results also pointed out that under the external uniform magnetic field application at 2D and 3D, thermal conductivity enhancement is less affected by the particle concentration increment.
Citation - Scopus: 3
3d Printing With Bacterial Cellulose-Based Bioactive Composites for Design Applications
(Education and research in Computer Aided Architectural Design in Europe, 2022) Turhan G.D.; Afsar S.; Ozel B.; Doyuran A.; Varinlioglu G.; Bengisu, Murat
The bacterial cellulose (BC) biofilms are explored in design applications as replacements to petroleum-based materials in order to overcome the irreversible effects of the Anthropocene. Unlike biomaterials, designers as mediators could collaborate with bioactive polymers as a form of wetware to manufacture living design products with the aid of novel developments in biology and engineering. Past and ongoing experiments in the literature show that BC has a strong nanofibril structure that provides adhesion for attachment to plant cellulose-based networks and it could grow on the surfaces of the desired geometry thanks to its inherited, yet, controllable bio-intelligence. This research explores BC-based bioactive composites as wetware within the context of digital fabrication in which the methodology involves distinct, yet integrated, three main stages: Digital design and G-code generation (software stage); BC cultivation and printable bioactive composite formulation (wetware stage); digital fabrication with a customized 3D printer (hardware stage). The results have shown that the interaction of BC and plant-based cellulose fibers ofjute yarns has enhanced the structural load-bearing capacity of the form against compressive forces, while pure BC is known only by its tensile strength. Since the outcomes were fabricated with the use of a bioactive material, the degradation process also adds a fourth dimension: Time, by which the research findings could further establish a bio-upcycling process of wastes towards biosynthesis of valuable products. Moreover, developing a BC-based bioactive filament indicates potentially a feasible next step in the evolution of multiscale perspectives on the growth of habitable living structures that could reinforce the interaction between nature and architecture through collaboration with software, hardware, and wetware in innovative and sustainable ways. © 2022, Education and research in Computer Aided Architectural Design in Europe. All rights reserved.
3d Printing With Bacterial Cellulose-Based Bioactive Composites for Design Applications
(Ecaade-education & research computer aided architectural design europe, 2022) Turhan, Gözde Damla; Afşar, Seçil; Özel, Berfin; Doyuran, Aslıhan; Varinlioglu, Güzden; Bengisu, Murat
The bacterial cellulose (BC) biofilms are explored in design applications as replacements to petroleum-based materials in order to overcome the irreversible effects of the Anthropocene. Unlike biomaterials, designers as mediators could collaborate with bioactive polymers as a form of wetware to manufacture living design products with the aid of novel developments in biology and engineering. Past and ongoing experiments in the literature show that BC has a strong nanofibril structure that provides adhesion for attachment to plant cellulose-based networks and it could grow on the surfaces of the desired geometry thanks to its inherited, yet, controllable bio-intelligence. This research explores BC-based bioactive composites as wetware within the context of digital fabrication in which the methodology involves distinct, yet integrated, three main stages: Digital design and G-code generation (software stage); BC cultivation and printable bioactive composite formulation (wetware stage); digital fabrication with a customized 3D printer (hardware stage). The results have shown that the interaction of BC and plant-based cellulose fibers of jute yarns has enhanced the structural load-bearing capacity of the form against compressive forces, while pure BC is known only by its tensile strength. Since the outcomes were fabricated with the use of a bioactive material, the degradation process also adds a fourth dimension: Time, by which the research findings could further establish a bio-upcycling process of wastes towards biosynthesis of valuable products. Moreover, developing a BC-based bioactive filament indicates potentially a feasible next step in the evolution of multiscale perspectives on the growth of habitable living structures that could reinforce the interaction between nature and architecture through collaboration with software, hardware, and wetware in innovative and sustainable ways.
44th Febs Congress, From Molecules To Living Systems (2019) Posters (vol 9, Pg 65, 2019)
(Wiley, 2019) Said, Harun Muayad; Gul, Guler; Akdoğan, Gül; Soysal, Yasemin
[Abstract Not Available]
The 6-Minute Threshold: Tracking Three-Year Disability and Cognition in Multiple Sclerosis
(Sage Publications Ltd, 2025) Unal, Gozde Deniz; Caliskan, Can; Zengin, Ela; Ozakbas, Serkan
An Abbreviated History of Liver Transplantation
(Wolters Kluwer Medknow Publications, 2024) Schilsky, M.L.; Emre, S.H.
[No abstract available]
Abernethy Syndrome Presenting With Hepatopulmonary Syndrome, Hematuria and Multiple Fnhs
(Lippincott Williams & Wilkins, 2018) Farajov, R.; Iakobadze, Z.; Yilmaz, C.; Karaca, C.; Kilic, K.; Kul, G.; Tosun, A.
[Abstract Not Available]
Citation - WoS: 1
Ablation of Atrioventricular Nodal Reentrant Tachycardia With Focal Cryoablation, Compared With Radiofrequency Ablation: Single-Center Experience
(Arquivos Brasileiros Cardiologia, 2024) Topaloğlu, Caner; Fici, Francesco; Van de Borne, Philippe; Taşkın, Uğur; Doğduş, Mustafa; Saygı, Serkan; Tengiz, İstemihan
Background: The ablation of atrioventricular nodal reentrant tachycardia (AVNRT) with cryoablation is an alternative to radiofrequency (RF) ablation in patients due to the low risk of total atrioventricular block. An increase in early-late recurrences after cryoablation is reported as an important disadvantage. Objectives: In this study, we aimed to compare the acute procedural success and the long-term recurrence rates of patients, with AVNRT who underwent methods. Methods: 73 patients with AVNRT were included in the study: 32 with cryoablation and 41 with RF ablation. There was no statistically significant difference between acute procedural success in methods. The ablation procedure was performed by an operator experienced in arrhythmology. The choice of RF or cryoablation was made in the electrophysiology laboratory based on the material already available during the procedure. After the procedure, the patients were evaluated every 3 months for 2 years in polyclinic control. The significance level adopted in the statistical analysis was 5%. Results: The 2 groups of patients were homogeneous. The fluoroscopy time (p<0.001) was shorter, but atrium-his (p=0.004) and his-ventricular (p=0.015) times were longer in the cryoablation group. There was no significant difference, in terms of acute procedural success, post-procedure jump without a single echo, and presence of echo and jump. Conclusions: Cryoablation requires less fluoroscopy time and is a safe non-inferior alternative to RF ablation in patients with AVNRT. The risk of AV block is a significant problem with the use of RF energy, making it less suitable for use in young and physically active patients.
Citation - Scopus: 1
Ablation of Atrioventricular Nodal Reentrant Tachycardia With Focal Cryoablation, Compared With Radiofrequency Ablation: Single-Center Experience
(2024) Topaloğlu, C.; Fici, F.; Borne, P.V.; Taşkin, U.; Dogdus, M.; Saygi, S.; Tengiz, I.
BACKGROUND: The ablation of atrioventricular nodal reentrant tachycardia (AVNRT) with cryoablation is an alternative to radiofrequency (RF) ablation in patients due to the low risk of total atrioventricular block. An increase in early-late recurrences after cryoablation is reported as an important disadvantage. OBJECTIVES: In this study, we aimed to compare the acute procedural success and the long-term recurrence rates of patients, with AVNRT who underwent methods. METHODS: 73 patients with AVNRT were included in the study: 32 with cryoablation and 41 with RF ablation. There was no statistically significant difference between acute procedural success in methods. The ablation procedure was performed by an operator experienced in arrhythmology. The choice of RF or cryoablation was made in the electrophysiology laboratory based on the material already available during the procedure. After the procedure, the patients were evaluated every 3 months for 2 years in polyclinic control. The significance level adopted in the statistical analysis was 5%. RESULTS: The 2 groups of patients were homogeneous.
Citation - WoS: 9
Citation - Scopus: 8
Abnormal Cross Frequency Coupling of Brain Electroencephalographic Oscillations Related To Visual Oddball Task in Parkinson's Disease With Mild Cognitive Impairment
(Sage Publications Inc, 2023) Bayraktaroglu, Zubeyir; Akturk, Tuba; Yener, Görsev; de Graaf, Tom A.; Hanoglu, Lutfu; Yildirim, Ebru; Gunduz, Duygu Hunerli
Parkinson's disease (PD) is a movement disorder caused by degeneration in dopaminergic neurons. During the disease course, most of PD patients develop mild cognitive impairment (PDMCI) and dementia, especially affecting frontal executive functions. In this study, we tested the hypothesis that PDMCI patients may be characterized by abnormal neurophysiological oscillatory mechanisms coupling frontal and posterior cortical areas during cognitive information processing. To test this hypothesis, event-related EEG oscillations (EROs) during counting visual target (rare) stimuli in an oddball task were recorded in healthy controls (HC; N = 51), cognitively unimpaired PD patients (N = 48), and PDMCI patients (N = 53). Hilbert transform served to estimate instantaneous phase and amplitude of EROs from delta to gamma frequency bands, while modulation index computed ERO phase-amplitude coupling (PAC) at electrode pairs. As compared to the HC and PD groups, the PDMCI group was characterized by (1) more posterior topography of the delta-theta PAC and (2) reversed delta-low frequency alpha PAC direction, ie, posterior-to-anterior rather than anterior-to-posterior. These results suggest that during cognitive demands, PDMCI patients are characterized by abnormal neurophysiological oscillatory mechanisms mainly led by delta frequencies underpinning functional connectivity from frontal to parietal cortical areas.
Citation - WoS: 9
Citation - Scopus: 17
Abnormal Ecg Beat Detection Based on Convolutional Neural Networks
(Institute of Electrical and Electronics Engineers Inc., 2020) Ozdemir M.A.; Guren O.; Cura O.K.; Akan A.; Onan A.
The heart is the most critical organ for the sustainability of life. Arrhythmia is any irregularity of heart rate that causes an abnormality in your heart rhythm. Clinical analysis of Electrocardiogram (ECG) signals is not enough to quickly identify abnormalities in the heart rhythm. This paper proposes a deep learning method for the accurate detection of abnormal and normal heartbeats based on 2-D Convolutional Neural Network (CNN) architecture. Two channels of ECG signals were obtained from the MIT-BIH arrhythmia dataset. Each ECG signal is segmented into heartbeats, and each heartbeat is transformed into a 2-D grayscale heartbeat image as an input for CNN structure. Due to the success of image recognition, CNN architecture is utilized for binary classification of the 2-D image matrix. In this study, the effect of different CNN architectures is compared based on the classification rate. The accuracies of training and test data are found as 100.00% and 99.10%, respectively for the best CNN model. Experimental results demonstrate that CNN with ECG image representation yields the highest success rate for the binary classification of ECG beats compared to the traditional machine learning methods, and one-dimensional deep learning classifiers. © 2020 IEEE.

Browse

Browsing 01. Araştırma Çıktıları | TR-Dizin | WoS | Scopus | PubMed by Language "en"