3792 research outputs found
Sort by
Geopolitical parallax: beyond Walter Lippmann just after large language models
Objectivity in journalism has long been contested, oscillating between ideals of neutral, fact-based reporting and the inevitability of subjective framing. With the advent of large language models (LLMs), these tensions are now mediated by algorithmic systems whose training data and design choices may themselves embed cultural or ideological biases. This study investigates geopolitical parallax—systematic divergence in news quality and subjectivity assessments—by comparing articlelevel embeddings from Chinese-origin (Qwen, BGE, Jina) and Western-origin (Snowflake, Granite) model families. We evaluate both on a human-annotated news quality benchmark spanning fifteen stylistic, informational, and affective dimensions, and on parallel corpora covering politically sensitive topics, including Palestine and reciprocal China–United States coverage. Using logistic regression probes and matched-topic evaluation, we quantify per-metric differences in predicted positive-class probabilities between model families. Our findings reveal consistent, nonrandom divergences aligned with model origin. In Palestinerelated coverage, Western models assign higher subjectivity and positive emotion scores, while Chinese models emphasize novelty and descriptiveness. Cross-topic analysis shows asymmetries in structural quality metrics—Chinese-on-US scoring notably lower in fluency, conciseness, technicality, and overall quality—contrasted by higher negative emotion scores. These patterns align with media bias theory and our distinction between semantic, emotional, and relational subjectivity, and extend LLM bias literature by showing that geopolitical framing effects persist in downstream quality assessment tasks. We conclude that LLMbased media evaluation pipelines require cultural calibration to avoid conflating content differences with model-induced bias.Preprint's Versio
Self-supervised learning of 3D structure from 2D OCT slices for retinal disease diagnosis on UK biobank scans
UK Biobank data handling and experiments was performed by Yasemin Turkan as part of her PhD thesis, using the UK Biobank Resource under Application Number 82266. Serdar Nazli developed the model and conducted experiments on the OCTA-500 dataset. Computational resources were provided by the Turkish National High-Performance Computing Center (UHEM) under Grant Number 1017802024. This study was also supported by the Scientific and Technological Research Council of Turkey (T\u00FC BI?TAK) under Grant Number 122E509.This study presents a self-supervised learning framework for retinal disease classification using Optical Coherence Tomography (OCT) scans. To balance the contextual richness of 3D volumes with the computational efficiency of 2D architectures, we introduce a quasi-3D input generation strategy. Each input is constructed by stacking three OCT slices, sampled from channel-specific Gaussian distributions centered on the volume midplane, and arranged in a standard three-channel 2D format compatible with existing pre-trained models. These quasi-3D images are used to pre-train a Vision Transformer (ViT-Base) via a Masked Autoencoder (MAE) with a shared masking pattern, encouraging the model to reconstruct masked regions by encoding anatomical continuity across slices. Pre-training is conducted on 10,000 unlabeled OCT volumes from the UK Biobank. The encoder is then fine-tuned on the OCTA-500 dataset for three-class and four-class retinal disease classification tasks, including macular degeneration and diabetic retinopathy. The model achieves 92.57% accuracy on the three-class task, matching the performance of RETFound while using over 150 times less pre-training data and a smaller backbone.Türkiye Bilimsel ve Teknolojik Araştırma KurumuPublisher's Versio
Renewable electricity consumption and economic growth: a cross-income panel ARDL analysis with implications for energy transition
The transition to renewable energy and the corresponding use of electricity generated from renewable sources is an inevitable solution that must be adopted to mitigate the effects of the climate crisis. The extant literature on the energy-economic growth nexus present mixed findings – some studies suggest the existence of a relationship while others find no significant relationship. Nevertheless, a growing number of recent studies provide evidence of an existing relationship. This study employs the panel ARDL techniques PMG, MG and DFE to investigate the short-run and long-run dynamics between renewable electricty consumption (RELC) and economic growth over the period 2000-2022 across 48 countries classified by income level. The results clearly indicate that the use of green electricity has a positive effect on economic growth across all income levels, albeit with varying magnitudes. Findings of the study provide particularly encouraging empirical evidence for a green transition in developing countries, underscoring the need for more attention to the unique challenges and opportunities faced by these economies.Yenilenebilir enerjiye geçiş ve buna bağlı olarak yenilenebilir enerji kaynaklı elektrik kullanımı, iklim krizinin etkilerini hafifletmek amacıyla kaçınılmaz olarak benimsenmesi gereken bir çözümdür. Enerji ile ekonomik büyüme bağlantısına ilişkin mevcut literatür farklı bulgulara ulaşmış; kimi çalışmalar bir ilişkinin varlığını öne sürerken, kimileri de önemli bir bağlantı tespit etmemiştir. Bununla birlikte, giderek artan sayıda yeni çalışma mevcut ilişkiye dair kanıt sunmaktadır. Bu çalışma, 2000-2022 döneminde gelir düzeyine göre sınıflandırılmış 48 ülkede yenilenebilir elektrik tüketimi (YET) ile ekonomik büyüme arasındaki kısa ve uzun vadeli dinamikleri araştırmak için panel ARDL teknikleri PMG, MG ve DFE'yi kullanmaktadır.. Sonuçlar, yeşil elektrik kullanımının, değişen büyüklüklerde de olsa, tüm gelir düzeylerinde ekonomik büyüme üzerinde olumlu etkiye sahip olduğunu açıkça göstermektedir. Çalışmanın bulguları, özellikle gelişmekte olan ülkelerde yeşil dönüşüme ilişkin cesaretlendirici ampirik kanıtlar sunmakta olup, söz konusu ekonomilere özgü zorluk ve fırsatlara daha fazla dikkat edilmesi gerektiğinin altını çizmektedir.Publisher's Versio
Retinal disease classification from bimodal OCT and OCTA using a CNN-ViT hybrid architecture
This study was supported by the Scientific and Technological Research Council of Turkey (TUBITAK) under Grant Number 122E509.Retinal diseases are the leading cause of vision impairment and blindness worldwide. Early and accurate diagnosis is critical for effective treatment, and recent advances in imaging technologies such as Optical Coherence Tomography (OCT) and OCT Angiography (OCTA), have enabled detailed visualization of the retinal structure and vasculature. By leveraging these modalities, this study proposes an advanced deep learning architecture called MultiModalNet for automated multi-class retinal disease classification. MultiModalNet employs a dual-branch design, where OCTA projection maps are processed through a ResNet101 encoder, and cross-sectional slices from the OCT volume (B-scans) are analyzed using a Vision Transformer (ViT-Large). The extracted features from both branches were fused and passed through the fully connected layers for the final classification. Evaluated on the 3-class OCTA-500 dataset, which includes Age-related Macular Degeneration (AMD), Diabetic Retinopathy (DR), and Normal cases, the proposed model achieved state-of-the-art classification accuracy of 94.59 percent, significantly o utperforming single-modality baselines. This result highlights the effectiveness of integrating vascular and structural information to improve the diagnostic performance. The findings suggest that hybrid multi-modal deep learning approaches can play a transformative role in computer-aided ophthalmology, enhancing both clinical decision-making and screening workflows.Türkiye Bilimsel ve Teknolojik Araştırma KurumuPublisher's Versio
Intelligent health monitoring in 6G networks: machine learning-enhanced VLC-based medical body sensor networks
Recent advances in Artificial Intelligence (AI)-driven wireless communication are driving the adoption of Sixth Generation (6G) technologies in crucial environments such as hospitals. Visible Light Communication (VLC) leverages existing lighting infrastructure to deliver high data rates while mitigating electromagnetic interference (EMI); however, patient movement induces fluctuating signal strength and dynamic channel conditions. In this paper, we present a novel integration of site-specific ray tracing and machine learning (ML) for VLC-enabled Medical Body Sensor Networks (MBSNs) channel modeling in distinct hospital settings. First, we introduce a Q-learning-based adaptive modulation scheme that meets target symbol error rates (SERs) in real time without prior environmental information. Second, we develop a Long Short-Term Memory (LSTM)-based estimator for path loss and Root Mean Square (RMS) delay spread under dynamic hospital conditions. To our knowledge, this is the first study combining ray-traced channel impulse response modeling (CIR) with ML techniques in hospital scenarios. The simulation results demonstrate that the Q-learning method consistently achieves SERs with a spectral efficiency (SE) lower than optimal near the threshold. Furthermore, LSTM estimation shows that D1 has the highest Root Mean Square Error (RMSE) for path loss (1.6797 dB) and RMS delay spread (1.0567 ns) in the Intensive Care Unit (ICU) ward, whereas D3 exhibits the highest RMSE for path loss (1.0652 dB) and RMS delay spread (0.7657 ns) in the Family-Type Patient Rooms (FTPRs) scenario, demonstrating high estimation accuracy under realistic conditions.Publisher's Versio
Investigation of the mediating role of risktaking behavior, reward sensitivity, and cyclothymic/irritable temperament in the relationship between bipolar disorder symptoms and chronotype
Text in Turkish ; Abstract: Turkish and Englishxii, 102 leavesIncludes bibliographical references (leaves 66-87)Bu araştırmada, sağlıklı popülasyonda bipolar bozukluk (BB) riski ile kronotip arasındaki ilişkide, ödül duyarlılığı, risk alma davranışı ve siklotimik/irritabl mizacın aracılık rolleri incelenmiştir. Çalışmaya, BB veya başka herhangi bir psikiyatrik tanısı bulunmayan, Türkiye’de üniversite öğrenimi gören ve çalışmaya gönüllü olarak katılmayı kabul eden, 18-24 yaş aralığındaki 244 birey (61.1% (n=149) kadın) dahil edilmiştir. Veriler, yüz yüze ve çevrim içi olmak üzere iki aşamada toplanmıştır. Katılımcılara bilgisayar ortamında Ödül için Efor Sarf Etme Görevi ve Modifiye Balon Analog Risk Testi uygulanmış, ardından Sabahçıl-Akşamcıl Testi, Duygudurum Bozuklukları Ölçeği (DBÖ), TEMPS-A Mizaç Ölçeği’nin Siklotimik ve İrritabl alt boyutları, Davranışsal İnhibisyon Sistemi/Davranışsal Aktivasyon Sistemi Ölçeği ve BARRATT Dürtüsellik Ölçeği Kısa Formu doldurtulmuştur. Katılımcıların yaş ortalaması 21.23±1.59’dur. Kronotip dağılımı incelendiğinde, sabahçıl tipte 27 kişi (%11.1), ara tipte 152 kişi (%62.3) ve akşamcıl tipte 65 kişi (%26.6) yer almaktadır. Kronotip ve DBÖ ile ölçülen BB riski arasında anlamlı bir ilişki saptanmamıştır (r=-0.025, p>0.05). Kronotip gruplarına göre DBÖ ile ölçülen BB riski açısından anlamlı bir fark bulunmamıştır (F=0.373, p=0.689). Dürtüsellik düzeyleri arttıkça DBÖ ile ölçülen BB riski de artış göstermektedir (r=0.350, p0.05). No significant difference was found in BD risk as measured by the MDQ between chronotype groups (F=0.373, p=0.689). As impulsivity levels increased, BD risk as measured by the MDQ also increased (r=0.350, p<0.001). Additionally, a significant and positive correlation was found between irritable temperament and BD risk as measured by the MDQ (r=0.149, p=0.018). In conclusion, no relationship was identified between chronotype and BD risk as measured by the MDQ. However, factors such as irritable temperament and impulsivity were observed to be significantly associated with BD risk as measured by the MDQ. This finding may be attributed to the fact that the sample consisted of healthy individuals who did not exhibit clinical levels of BD symptoms. These findings indicate the need for further research to better understand the relationship between chronotype and BD risk, as well as to identify psychological and behavioral factors associated with BD vulnerability.BİPOLAR BOZUKLUKBB ProdromuBB İçin Risk ve Klinik Olarak Çok Yüksek RiskGenel Popülasyonda BB BelirtileriKRONOTİPSabahçıl KronotipAkşamcıl KronotipSİKLOTİMİK/İRRİTABL MİZAÇRİSK ALMA DAVRANIŞIÖDÜL DUYARLILIĞIDAS ModeliDEĞİŞKENLER ARASINDAKİ İLİŞKİLERLE İLGİLİ YAPILMIŞ ARAŞTIRMALARKronotip ve BB BelirtileriKronotip ve Siklotimik/İrritabl MizaçKronotip ve Risk Alma DavranışıKronotip ve Ödül DuyarlılığıBB Belirtileri ve Ödül DuyarlılığıBB Belirtileri ve Risk Alma DavranışıBB Belirtileri ve Siklotimik/İrritabl MizaçBB Belirtileri, Kronotip ve Siklotimik/İrritabl Miza
The mediating role of self-efficacy in prediction of attachment styles and cognitive flexibility
Text in Turkish ; Abstract: Turkish and EnglishIncludes bibliographical references (leaves 66-82)xii, 90 leavesAmaç: Bu çalışmanın amacı yetişkinlerde bağlanma stilleri ile bilişsel esneklik arasındaki ilişkiyi incelemek ve bu ilişkide öz yeterliğin aracı rolünü incelemektir. Değişkenlerin birbiri ile olan ilişkisine bakılmasına ek olarak söz edilen değişkenlerin farklı sosyodemografik değişkenler açısından incelenmesi de amaçlanmıştır. Yöntem: Araştırmanın örneklemi, yaş ortalaması 42.36±14.05 olan, yaşları 18 ile 65 arasında değişen, ilişkisi en az 2 aydır devam eden, 395’i kadın ve 115’i erkek olmak üzere toplam 510 katılımcıdan oluşmaktadır. Araştırmada ölçüm araçları olarak Yakın İlişkilerde Yaşantılar Envanteri-II, Bilişsel Esneklik Envanteri ve Genel Öz Yeterlik Ölçeği kullanılmıştır. Veri analizinde, frekans analizi, hiyerarşik regresyon analizi, bağımsız gruplar t-testi, Pearson korelasyon analizi, hiyerarşik regresyon analizi ve PROCESS aracı etki analizi uygulanmıştır. Bulgular: Kaygılı ve kaçıngan bağlanma stilleri bilişsel esnekliği ve öz yeterliliği negatif düzeyde yordamaktadır. Öz yeterlilik ise bilişsel esnekliği pozitif yönde yordamaktadır. Öz yeterliliğin, bağlanma stilleri ve bilişsel esneklik arasındaki ilişkide kısmi aracılık rolü bulunmaktadır. Kaygılı ve kaçıngan bağlanma düzeyleri arttıkça öz yeterlilik ve bilişsel esneklik düzeyleri düşüş göstermektedir. Öte yandan, öz yeterlilik arttıkça bilişsel esnekliğin de arttığı sonucuna ulaşılmıştır. Sonuç: Bireyin ebeveynleri ile ilişkisinde oluşturduğu bağlanma stillerinin yetişkin hayatında romantik ilişkilerinde gözlemlendiği görülmüştür. İlişkilerinde sahip olduğu bağlanma stilleri bilişsel esneklik ve öz yeterlilik düzeyini etkilemektedir. Bilişsel esnekliğin bağlanma stilleri ile olan ilişkisinde de öz yeterliliğin önemli bir role sahip olduğu görülmüştür.Purpose: The purpose of this study is to examine the relationship between attachment styles and cognitive flexibility in adults and to examine the mediating role of self-efficacy in this relationship. In addition, it is aimed to examine the mentioned variables in terms of different sociodemographic variables. Method: The sample of the study consists of a total of 510 participants, 395 female and 115 males, with an average age of 42.36±14.05, ages ranging from 18 to 65, and a relationship that has been going on for at least 2 months. The Experiences in Close Relationships Inventory-II, Cognitive Flexibility Inventory, General Self-Efficacy Scale were used in the study. Frequency analysis, hierarchical regression analysis, independent groups t-test, Pearson correlation analysis, regression analysis and PROCESS mediation analysis were applied in the data analysis. Results: Anxious and avoidant attachment styles negatively predicted cognitive flexibility and self-efficacy. In addition, self-efficacy positively predicted cognitive flexibility. Self-efficacy had a partial mediating role in the relationship between attachment styles and cognitive flexibility. As anxious and avoidant attachment levels increased, self-efficacy and cognitive flexibility levels decreased. On the other hand, cognitive flexibility increased as selfefficacy increased. Conclusion: It has been observed that the attachment styles that an individual forms in their relationship with their parents are observed in their romantic relationships in adult life. The attachment styles they have in their relationships affect their cognitive flexibility and self-efficacy levels. It has been observed that self-efficacy plays an important role in the relationship between cognitive flexibility and attachment styles.BAĞLANMABağlanma KavramıBağlanma StilleriBağlanma Zemininde Bilişsel GelişimBİLİŞSEL ESNEKLİKBilişsel Esneklik KavramıBilişsel Esneklik KuramlarıBilişsel Esnekliğe Etki Eden ÖzelliklerÖZ YETERLİKÖzyeterlilik KavramıÖz Yeterlilik Gelişimini Etkileyen FaktörlerÖz Yeterliliğin KaynaklarıARAŞTIRMA DEĞİŞKENLERİ ARASINDAKİ İLİŞKİLERE DAİR ÇALIŞMALA
Organizational-level maturity in artificial intelligence, cybersecurity, and digitalization: a survey-based assessment
Text in Turkish ; Abstract: Turkish and EnglishIncludes bibliographical references (leaves 46-51)xii, 52 leavesDijital teknolojilerin sektörler genelinde ivmelenen gelişimi, örgütlerin rekabet gücünü sürdürebilmeleri ve çevik biçimde dönüşüme ayak uydurabilmeleri için Yapay Zekâ (YZ), Siber Güvenlik (SG) ve Dijital Dönüşüm (DD) alanlarında daha derinlemesine yetkinliklere sahip olmalarını zorunlu kılmıştır. Bu üç alan, dijital çağda sadece teknik kapasite olarak değil; aynı zamanda yönetsel strateji, risk yönetimi, veri bütünlüğü ve sürdürülebilir inovasyon açısından da hayati rol oynamaktadır. Literatürde her bir alan için ayrı ayrı önemli çalışmalar bulunmakla birlikte, bu alanların birbirleriyle olan etkileşimleri ve bütünleşik bir çerçevede organizasyonel olgunluk üzerindeki bileşik etkileri yeterince derinlemesine analiz edilmemiştir. Bu bağlamda sunulan çalışma, YZ, SG ve DD olgunluk düzeylerini çok boyutlu bir yapıda ele alarak aralarındaki nedensel ilişkileri Yapısal Eşitlik Modellemesi (SEM) ile ortaya koymayı amaçlamaktadır. Ayrıca, karma yöntemli bir metodoloji benimsenmiş; nicel anket bulguları sentetik modelleme teknikleriyle desteklenerek kapsamlı bir değerlendirme gerçekleştirilmiştir. Araştırma bulguları, YZ, SG ve DD arasında istatistiksel olarak anlamlı ve çift yönlü korelasyonlar bulunduğunu göstermekte; özellikle teknoloji ve finans sektörlerinde bu olgunluk düzeylerinin kamu ve eğitim sektörlerine kıyasla daha ileri düzeyde olduğunu ortaya koymaktadır. Çalışma, bu alanlar arasında stratejik entegrasyonun sağlanmasının dijital dayanıklılık açısından kritik olduğunu savunmakta ve entegre bir YZ-SG stratejisinin uygulanmasına yönelik yol gösterici ampirik veriler sunmaktadır. Böylece, sunulan model hem kuramsal katkı sağlamakta hem de ileride yapılacak ampirik saha araştırmaları için stratejik ve metodolojik bir temel oluşturmaktadır.The accelerating advancement of digital technologies across industries has made it imperative for organizations to develop deeper competencies in Artificial Intelligence (AI), Cybersecurity (CS), and Digital Transformation (DT) in order to maintain competitive advantage and adapt to change with agility. These three domains play a critical role in the digital age not only in terms of technical capacity, but also in managerial strategy, risk management, data integrity, and sustainable innovation. While significant research has been conducted on each domain individually, the interactions among them and their combined impact on organizational maturity within an integrated framework have not been sufficiently analyzed in depth. This study aims to address that gap by examining the maturity levels of AI, CS, and DT in a multidimensional structure, and revealing the causal relationships among them using Structural Equation Modeling (SEM). A mixedmethods approach has been adopted, combining quantitative survey data with synthetic modeling techniques for a comprehensive evaluation. The research findings indicate statistically significant and bidirectional correlations among AI, CS, and DT. Particularly, the technology and finance sectors exhibit higher maturity levels compared to the public and education sectors. The study argues that strategic integration among these areas is critical for digital resilience and provides empirical insights to guide the implementation of an integrated AI-CS strategy. Thus, the proposed model contributes both theoretically and methodologically, laying a foundation for future empirical field research
Examination of the relationship between codependency, individuation, self-silencing, and self-objectification
Mevcut çalışmanın temel amacı, eş bağımlılık ile bireyleşme, kendini susturma ve kendini nesneleştirme ilişkisinin incelenmesidir. Aynı zamanda eş bağımlılığın bazı sosyodemografik faktörlere göre farklılaşmasının incelenmesi de araştırmanın bir diğer amacıdır. Bu doğrultuda gerçekleştirilen çalışmanın örneklemini 18-29 yaş aralığındaki 313 birey oluşturmaktadır. Çalışmanın veri toplama araçları katılımcılara sırasıyla Sosyodemografik Bilgi Formu, Beliren Yetişkinler için Bireyleşme Testi - Kısa Form (BYBT - KF), Eş Bağımlılık Belirleme Ölçeği (EBBÖ), Nesneleştirilmiş Beden Bilinci Ölçeği (NBBÖ) ve Kendini Susturma Ölçeği (KSÖ) şeklinde sunulmuştur. Çalışmanın hipotezlerini test etmek amacıyla eş bağımlılığın sosyodemografik faktörlere göre farklılaşıp farklılaşmadığını değerlendirebilmek için t-test vet ek yönlü varyans analizi (ANOVA) gerçekleştirilmiştir. Ardından mevcut değişkenlerin birbirleri ile ilişkilerinin görülebilmesi için Pearson ve Spearman Korelasyon Analizi uygulanmıştır. Son olarak sosyodemografik değişkenler, bireyleşme, kendini nesneleştirme ve kendini susturma hiyerarşik regresyon analizine dahil edilmiştir. Yapılan istatistiksel analizlerin sonucunda sosyodemografik değişkenlerden daha küçük yaş ve daha düşük eğitim düzeyine sahip katılımcıların eş bağımlılık seviyelerinin daha yüksek olduğu görülmüştür. Ayrıca yapılan korelasyon analizleri incelendiğinde eş bağımlılık ile nesneleştirilmiş beden bilinci ölçeğinin beden izleme ve beden utancı alt boyutları, kendini susturma ölçeğinin dışsallaştırılmış benlik algısı, fedakarlık, kendini susturma ve bölünmüş benlik alt boyutları arasında pozitif yönde istatistiksel olarak anlamlı ilişki bulunmuştur. Ek olarak eş bağımlılık ile eğitim düzeyi, yaş, baba ile bireyleşme düzeyi ve nesneleştirilmiş beden bilinci ölçeğinin kontrol inancı alt boyutu arasında negatif yönde istatistiksel olarak anlamlı ilişki bulunmuştur. Son olarak hiyerarşik regresyon analizine dahil edilen tüm değişkenlerin toplam varyansın %54’ünü anlamlı olarak açıkladığı görülmüştür. Mevcut bulgular ilgili literatür ışığında tartışılmış, çalışmanın sınırlılıklarına ve gelecek çalışmalar için önerilere yer verilmiştir.The primary aim of the present study is to examine the relationship between codependency and individuation, self-silencing, and selfobjectification. Additionally, the study investigates whether codependency varies based on certain sociodemographic factors. The sample consists of 313 individuals aged 18–29. Data were collected using the Sociodemographic Information Form, the Short Form of the Individuation Test for Emerging Adults (ITEA-SF), the Codependency Assessment Scale (CAS), the Objectified Body Consciousness Scale (OBCS), and the Self-Silencing Scale (SSS). To test the study’s hypotheses, t-tests and one-way analysis of variance (ANOVA) were conducted to assess whether codependency differs across sociodemographic factors. Pearson and Spearman correlation analyses were performed to explore the relationships between the variables. Finally, sociodemographic variables, individuation, self-objectification, and self-silencing were included in a hierarchical regression analysis. The statistical analyses revealed that participants who were younger and had lower education levels exhibited higher levels of codependency. Correlation analyses indicated statistically significant positive relationships between codependency and the body surveillance and body shame subscales of the OBCS, as well as the externalized self-perception, self-sacrifice, self-silencing, and divided self subscales of the SSS. Additionally, statistically significant negative relationships were found between codependency and education level, age, individuation with the father, and the control beliefs subscale of the OBCS. The hierarchical regression analysis showed that all included variables significantly explained 54% of the total variance. The findings are discussed in light of the relevant literature, and the study’s limitations and recommendations for future research are addressed.EŞ BAĞIMLILIKBİREYLEŞMEKENDİNİ SUSTURMAKENDİNİ NESNELEŞTİRMEBİREYLEŞME, KENDİNİ SUSTURMA, KENDİNİ NESNELEŞTİRME VE EŞ BAĞIMLILIK İLİŞKİSİEŞ BAĞIMLILIK VE SOSYODEMOGRAFİK FAKTÖRLEREŞ BAĞIMLILIĞIN SOSYODEMOGRAFİK ÖZELLİKLERE GÖRE İNCELENMESİEş Bağımlılığın Cinsiyete Göre İncelenmesiEş Bağımlılığın Yaşa Göre İncelenmesiEş Bağımlılığın Eğitim Düzeyine Göre İncelenmesiEş Bağımlılığın Medeni Duruma Göre İncelenmesiEş Bağımlılığın Psikoterapi Alma Durumuna Göre İncelenmes
5G ve 6G teknolojilerinin gelişimi uzaktan sağlık izleme sistemleri için milimetre dalga ve VLC entegrasyonu ile sağlanması
Text in English ; Abstract: English and TurkishIncludes bibliographical references (leaves 57-72)xiv, 73 leavesThis thesis examines the pivotal role of wireless networks in healthcare, emphasizing the need for high-performance technologies like 5G and emerging 6G to enable efficient data transfer between medical devices such as sensors and remote monitoring equipment. We delve into the current research landscape surrounding 5G mmWave technology in remote health monitoring systems, focusing on its applications, main challenges, and future trends. We explore the wireless connectivity requirements of reconfigurable hybrid optical-radio-based Medical Body Sensor Networks (MBSNs), proposing an extension of conventional MBSNs to more flexible and generic solutions. This thesis introduces a comprehensive literature review across diverse domains including antenna design, small implantable antennas, on-body wearable solutions, and adaptable detection and imaging systems. Our research further investigates methodological approaches in monitoring systems, analyzing channel characteristics, advancements in wireless capsule endoscopy, and sensing and imaging techniques. Additionally, we explore how 6G's framework integrates Visible Light Communication (VLC) in healthcare, demonstrating how VLC-enabled MBSNs can revolutionize remote patient monitoring and real-time health data transmission by accurately estimating VLC channel parameters, such as channel DC gain and RMS delay spread. We introduce a sophisticated ray tracing technique and ML-based algorithm to model channels and estimate path loss and RMS delay spread within different hospital settings such as ICU ward and family-type patient room. The detailed results of the hospital scenarios are listed using various machine learning algorithms such as LSTM, GRU, RNN, Linear Regression SVR, and KNN. The estimation was illustrated and detailed comprehensively by choosing the best-performing ML technique.Bu tez kablosuz ağların sağlık hizmetlerindeki önemli rolünü incelemekte olup sensörler ve uzaktan izleme ekipmanları gibi tıbbi cihazlar arasında verimli veri aktarımını sağlamak için 5G ve gelişmekte olan 6G gibi yüksek performanslı teknolojilere duyulan ihtiyacı vurgulamaktadır. Uzaktan sağlık izleme sistemlerinde 5G mmWave teknolojisinin mevcut araştırma alanını ele alarak, uygulamaları, ana zorlukları ve gelecekteki eğilimlere odaklanıyoruz. Yeniden yapılandırılabilir hibrit optik-radyo tabanlı Tıbbi Vücut Sensör Ağlarının (MBSN'ler) kablosuz bağlantı gereksinimlerini inceleyerek, geleneksel MBSN'lerin daha esnek ve genel çözümlere genişletilmesini öneriyoruz. Bu tez, anten tasarımı, küçük implante edilebilir antenler, vücut üzerinde giyilebilir çözümler ve uyarlanabilir algılama ve görüntüleme sistemleri gibi çeşitli alanlarda kapsamlı bir literatür taraması sunmaktadır. Araştırmamız ayrıca izleme sistemlerindeki metodolojik yaklaşımları, kanal karakteristiklerini analiz ederek kablosuz kapsül endoskopisindeki ilerlemeleri ve algılama ve görüntüleme tekniklerini incelemektedir. Ayrıca 6G çerçevesinin sağlık hizmetlerinde Görünür Işık İletişimini (VLC) nasıl entegre ettiğini araştırarak, VLC tabanlı MBSN'lerin uzaktan hasta izleme ve gerçek zamanlı sağlık verisi iletimini kanal DC kazancı ve RMS gecikme yayılımı gibi VLC kanal parametrelerini doğru bir şekilde tahmin ederek nasıl devrim yaratabileceğini göstermekteyiz. Farklı hastane ortamlarında (yoğun bakım ünitesi ve aile tipi hasta odası gibi) kanalları modellemek, yol kaybı ve RMS gecikme yayılımını tahmin etmek için gelişmiş bir ışın izleme tekniği ve ML tabanlı bir algoritma tanıtıyoruz. Hastane senaryolarının detaylı sonuçları, LSTM, GRU, RNN, Linear Regression SVR ve KNN gibi çeşitli makine öğrenimi algoritmaları kullanılarak listelenmiştir. Tahminler, en iyi performans gösteren ML tekniği seçilerek kapsamlı bir şekilde detaylandırılmış ve açıklanmıştır.ADVANCEMENTS IN REMOTE HEALTH DEVICESCURRENT METHODS AND APPROACHES IN REMOTE HEALTH MONITORINGMACHINE LEARNING CHANNEL PARAMETER ESTIMATION IN VLC-BASED MEDICAL BODY SENSOR NETWORKSOVERVIEW OF MACHINE LEARNING CHANNEL PARAMETER ESTIMATION IN VLC-BASED MBSNSMl Approaches For Channel Parameter EstimationMobile Channel Model For Vlc-Based MbsnsLstm-Based Channel Parameter Estimatio