Yıldız Technical University Research Information System
Not a member yet
91324 research outputs found
Sort by
Unveiling knee morphology with SHAP: shaping personalized medicine through explainable AI
No full textPurpose: This study aims to enhance personalized medical assessments and the early detection of knee-related pathologies by examining the relationship between knee morphology and demographic factors such as age, gender, and body mass index. Additionally, gender-specific reference values for knee morphological features will be determined using explainable artificial intelligence (XAI). Methods: A retrospective analysis was conducted on the MRI data of 500 healthy knees aged 20–40 years. The study included various knee morphological features such as Distal Femoral Width (DFW), Lateral Femoral Condyler Width (LFCW), Intercondylar Femoral Width (IFW), Anterior Cruciate Ligament Width (ACLW), and Anterior Cruciate Ligament Length (ACLL). Machine learning models, including Decision Trees, Random Forests, Light Gradient Boosting, Multilayer Perceptron, and Support Vector Machines, were employed to predict gender based on these features. The SHapley Additive exPlanation was used to analyze feature importance. Results: The learning models demonstrated high classification performance, with 83.2% (±5.15) for classification of clusters based on morphological feature and 88.06% (±4.8) for gender classification. These results validated that the strong correlation between knee morphology and gender. Conclusion: The study found that DFW is the most significant feature for gender prediction, with values below 78–79 mm range indicating females and values above this range indicating males. LFCW, IFW, ACLW, and ACLL also showed significant gender-based differences. The findings establish gender-specific reference values for knee morphological features, highlighting the impact of gender on knee morphology. These reference values can improve the accuracy of diagnoses and treatment plans tailored to each gender, enhancing personalized medical care
Experimental thermal and environmental impact performance evaluations of hydrogen-enriched fuels for power generation
No full textThe transition to a low-carbon energy future requires a multi-faceted approach, including the enhancement of existing power generation technologies. This study provides a comprehensive experimental evaluation of hydrogen enrichment as a strategy to improve the performance and reduce the emissions of a power generator. A 3.65 kW power generator that is equipped with spark-ignition engine is systematically tested with five distinct base fuels: gasoline, propane, methane, ethanol, and methanol. Each fuel is volumetrically blended with pure hydrogen in ratios of 5 %, 10 %, 15 %, and 20 % using a custom-developed dual-fuel carburetor. The key parameters, including exhaust emissions (CO2, CO, HC, NOx), cylinder exit temperature, electrical power output, and thermodynamic efficiencies (energy and exergy), are meticulously measured and analyzed. The results reveal that hydrogen enrichment is a powerful tool for decarbonization, consistently reducing carbon-based emissions across all fuels. At a 20 % hydrogen blend, CO2 emissions are reduced by 22–31 %, CO emissions by 39–60 %, and HC emissions by 21–60 %. This environmental benefit, however, is accompanied by a critical trade-off: a severe increase in NOx emissions, which rose by 200–420 % due to significantly elevated combustion temperatures. The power outputs are increased by 2–16 %, with hydrogen addition enabling lower-energy–density fuels like methane and propane to achieve performance parity with gasoline. Thermodynamic analysis confirms these gains, with energy efficiency showing marked improvement, particularly for methane, which has increased from 42.0 % to 49.9 %. While hydrogen enrichment presents a viable pathway for enhancing engine performance and reducing the carbon emissions of power generators, the profound increase in NOx necessitates the integration of advanced control and after-treatment systems for its practical and environmentally responsible deployment
İNŞAAT SEKTÖRÜNDE DÖNGÜSEL YAKLAŞIM VE PAYDAŞ HARİTALAMASI: KİLİT PAYDAŞLARIN BELİRLENMESİ
No full textDigital ethnography of asynchronous online professional development: Voices from K-12 secondary school English teachers in Istanbul
No full textVolume Control in a Hybrid Mock Circuit of the Cardiovascular System using Command Filtered Backstepping Approach
No full textFarklılıkların Yönetiminin Örgütsel Yabancılaşma Üzerindeki Etkisinin Öğretmen Görüşlerine Göre İncelenmesi
No full textExtracting 3-Dimensional Brain Model From 2-Dimensional Brain CT Images With Deep Learning Derin grenme ile 2-Boyutlu Beyin BT Goruntulerinden 3-Boyutlu Beyin Modelinin Elde Edilmesi
No full textClinical imaging techniques such as computerized tomography (CT) are widely used by today's doctors to obtain information about the patient and decide on the appropriate treatment. In addition to these 2D CT images of the patient's relevant organ, a high resolution 3D model can be used to facilitate the preparation process of specialist doctors before risky surgeries and to minimize the problems they may encounter during the operation. Thus, the surgical preparation process can be carried out more efficiently with the model to be calculated. Segmentation has an important place in obtaining a 3D brain model from 2D brain CT images, and in this study, U-Net architecture was used in medical image segmentation. After segmentation, a high- accuracy and easy-to-analyze 3D brain model was calculated with Marching Cubes, a 3D modeling algorithm
MammosighTR: Nationwide Breast Cancer Screening Mammogram Dataset with BI-BADS Annotations for Artificial Intelligence Applications
No full textPromoting Sustainable Life Through Global Citizenship-Oriented Educational Approaches: Comparison of Learn–Think–Act Approach-Based and Lecture-Based SDG Instructions on the Development of Students’ Sustainability Consciousness
No full textPromoting individuals’ sustainability consciousness (SC) is one of the important way of ensuring a sustainable world and finding ways toward a better life. Therefore, the purpose of the present study was to compare the effects of learn–think–act approach-based instruction and lecture-based instruction on the development of sustainability consciousness in students, with the Sustainable Development Goals (SDGs) acting as the subject of the instructions. The research was conducted with 80 seventh-grade students from a state school in Istanbul, Türkiye. While 40 of them were in a class where learn–think–act approach-based SDG instruction was implemented, the other 40 participants were trained with lecture-based SDG instruction for eight weeks. A quasi-experimental research design was followed in the research. The data was collected with the Sustainability Consciousness Questionnaire and obtained before and after SDG instruction. In the data analysis, paired and independent samples t-tests were used. The findings revealed that learn–think–act approach-based SDG instruction has a significantly larger effect (d = 1.62, 95% CI) on the development of sustainability consciousness in middle school students compared to lecture-based SDG instruction