Recep Tayyip Erdoğan Üniversitesi Kurumsal Akademik Arşivi
DSpace@RTEÜ, Recep Tayyip Erdoğan Üniversitesi tarafından doğrudan ve dolaylı olarak yayınlanan; kitap, makale, tez, bildiri, rapor, araştırma verisi gibi tüm akademik kaynakları uluslararası standartlarda dijital ortamda depolar, Üniversitenin akademik performansını izlemeye aracılık eder, kaynakları uzun süreli saklar ve yayınların etkisini artırmak için telif haklarına uygun olarak Açık Erişime sunar.
Güncel Gönderiler
M ultimodal E motion D etection for E ducation and W ork E nvironment by U sing I mproved A rtificial I ntelligence M achine V ision S ystem
(Paradigm Publishing Services, 2026) Daud, Wan Mohd Bukhari Wan; Kıral, Adnan; Tokhi, Mohamed Osman; Yee, Lee Chung; Zawawi, Muhammad Muzhafar Mohammad
The use of artificial intelligence (AI) has significantly advanced emotion recognition within human-computer interaction (HCI). This paper aims to develop a multimodal emotion detection system for educational and work environments using an enhanced AI machine vision system. The primary focus is on training and testing a multimodal AI model in Python using convolutional neural networks (CNN). The results from the trained facial emotion AI model demonstrated substantial improvements. Training accuracy increased from 30.49% to 72.21%, while validation accuracy improved from 37.6% to 60.58%. Simultaneously, training loss decreased from 180.69% to 73.65%, and validation loss reduced from 172.97% to 107.53%. This CNN-based model can use OpenCV to detect seven emotions: happy, sad, neutral, angry, afraid, disgusted, and surprised. The ECG emotion AI model, also trained with CNN, also successfully recognized patterns for the same seven emotions. When these two models are combined into a multimodal AI system, they can detect facial and ECG-based emotions simultaneously. This comprehensive approach allows for the detection of both visible and hidden emotions, such as stress or anxiety, which may not be easily discernible through facial expressions alone. The integration of these models into a multimodal AI system provides a more accurate and holistic understanding of human emotions, enhancing applications in educational and work settings. The improved detection capabilities can lead to better user experiences and more effective responses to emotional states, ultimately contributing to advancements in HCI.
Letter to the editor: "Beyond pain relief: Preventing recurrence and preserving function after lumbar discectomy"
(ISASS, 2026) Kanat, Ayhan
...
Some notes on the fine spectrum of quintet band matrix operator over c0 and c
(Hacettepe University, 2026) Bişgin, Mustafa Cemil; Topal, Kübra
In this work, we determine the fine spectrum of quintet band matrix operator G(r, s, t, u, v) over c0 and c. The quintet band matrix G(r, s, t, u, v) is the general form of the matrices D(r, 0, s, 0, t), ∆4, Q(r, s, t, u), ∆3, D(r, 0, 0, s), B(r, s, t), ∆2, B(r, s), ∆, right shift and Zweier matrices, where ∆4, Q(r, s, t, u), ∆3, B(r, s, t), ∆2, B(r, s) and ∆ are called fourth order difference, quadruple band, third order difference, triple band, second order difference, double band(generalized difference) and difference matrix, respectively.
Prognostic value of fibrosis-4 index, hepatic biomarkers, and plasma acute-phase reactants in critical patients undergoing percutaneous endoscopic gastrostomy: a retrospective cohort study
(Multidisciplinary Digital Publishing Institute (MDPI), 2026) İşsever, Kubilay; Muhtaroğlu, Ali; Kuloğlu, Ersin; Aslan, Sefer; Acar, Berkan; Konur, Kamil; Dülger, Ahmet Cumhur
Background/Objectives: Identifying reliable predictors of all-cause mortality in intensive care unit (ICU) patients undergoing percutaneous endoscopic gastrostomy (PEG) remains clinically important to decide the route of nutrition. This study aimed to assess the prognostic significance of hepatic biomarkers, particularly the Fibrosis-4 (FIB-4) score. Methods: We conducted a retrospective cohort study in the ICUs of a tertiary care university hospital. Adult patients who underwent PEG between 1 January 2022 and 31 December 2024 were included. Pre-procedural demographic, clinical, and laboratory data were retrieved from electronic medical records. The primary outcome was all-cause mortality within three years of PEG placement. Survival analyses were performed using Kaplan–Meier curves and Cox proportional hazards regression to identify independent predictors. Results: Older age, elevated FIB-4, total and direct bilirubin, gamma glutamyl transferase (GGT), lactate dehydrogenase (LDH), ferritin, c-reactive protein (CRP), procalcitonin, INR, and prothrombin (PT) levels, and reduced calcium, platelet count, and albumin were significantly associated with all-cause mortality (p < 0.05). ROC analysis identified FIB-4, GGT, LDH, and albumin as significant predictors of mortality. Kaplan–Meier analysis confirmed that patients with higher FIB-4, GGT, and LDH, and lower albumin levels, had significantly shorter survival. Cox proportional hazards regression analysis revealed that while the FIB-4 score was a significant predictor of mortality in the univariate model, it lost its independent prognostic value when adjusted for confounders. Instead, lower albumin, elevated CRP, and increased GGT emerged as the independent risk factors for mortality. Conclusions: Pre-procedural albumin, CRP, and GGT levels are strong independent prognostic indicators of all-cause mortality in ICU patients undergoing PEG. While FIB-4 serves as a practical initial screening tool, adverse outcomes might be primarily driven by systemic inflammation and nutritional depletion rather than isolated hepatic fibrosis.
Physics-informed neural network and data-driven modeling of non-Fourier heat transfer in laser-irradiated semiconductor media using bi-Helmhotz nonlocal theory
(Springer, 2026) Sur, Abhik; Hussain, Syed Modassir; Craciun, Eduard-Marius; Singhal, Abhina; Ranjit, Nayan Kumar; Yaylacı, Murat; Das, Soumik
This study develops a hybrid analytical and physics-informed machine learning framework to analyze non-Fourier heat transfer in laser-irradiated semiconductor media. The model integrates bi-Helmholtz nonlocal thermoelasticity with dual length-scale parameters to capture size-dependent mechanical effects, alongside a modified Green–Naghdi heat conduction theory to describe finite-speed thermal wave propagation and relaxation phenomena. The coupled governing equations for displacement, temperature, carrier density, and stress are first solved analytically using a normal-mode approach to obtain benchmark solutions. A physics-informed neural network (PINN) is then constructed by embedding the governing equations and boundary conditions into the learning process, enabling efficient and accurate prediction of multiphysics responses. The proposed approach significantly reduces computational cost while preserving high accuracy. Parametric analysis highlights the strong influence of nonlocal parameters and thermal relaxation on wave propagation and carrier dynamics. The framework offers a robust tool for real-time simulation of laser-induced thermal processes in semiconductor systems.
