Investigation of the Effects of Statistically Significant Features on the Classification of EEG-Based Motor Imagery Tasks

Murside Degirmenci

Motor imagery (MI) task classification is highly prevalent in Electroencephalography (EEG)-based Brain-Computer Interface (BCI) research area. Extremity movement task classification and finger movement classification studies are presented in this thesis. In extremity movement classification, binary-class (right hand and left hand) and multi-class (right hand, left hand, right hand, and left hand) classifications are performed using 4 different feature extraction approaches and statistically significance-based feature selection (the independent t-test, one-way ANOVA test). Firstly, time- ...Daha fazlası

Performance Study of Darrieus Turbine Using Numerical Simulation

Abdelrahman ABUALEENEIN

In this study, CFD abilities to handle Darrieus turbine simulation were assessed and validated in light of related experimental and numerical studies. Simulation methodology is based on two-dimensional modeling of a double NACA 0018 bladed, straight blade Darrieus turbine. Modeling approach, meshing procedure, and solution parameters are mentioned and discussed in details. ANSYS fluent 18.2 is used to carry out all simulations. Experimental test’s data from literature [11] are implemented in context of results verification, also, Courant number and y plus analyses were provided. Due to the sim ...Daha fazlası

Finite Element Analysis-Based Pre-operative Planning Applications of Tooth, Tibia, and Femur Bones

Samet Çıklaçandır

Ameliyat öncesi planlama veya eğitimde, zorlu cerrahi operasyonları kolaylaştırmak için 3D modelleme ve kişiye özgü kemik modeli kullanılabilir. Bu prosedür, 3D baskı ve simülasyonların gelişmesi nedeniyle daha etkili ve uygulanabilir hale geldi. Bunun yanında, sonlu elemanlar analizi (SEA) mekanik analiz sonuçlarına yakınsayarak kemik modellerinin ve çeşitli tıbbi ekipmanların oluşturulmasını daha hızlı ve daha güvenilir hale getirebilir. Fakat, simülasyonların ve mekanik testlerin modeller üzerindeki etkisi hala tartışmalıdır. Bu sebeple bu çalışmada femur başı ve medial malleol olmak üzere ...Daha fazlası

Optimization of Drying Process of Plastic Granules Based on Stochastic Method with Neuro Regression Approach

Mustafa DİNÇ

Polymers contain moisture due to their chemical structure. Moisture in polymers is important for the quality of the final product and ease of processing. For this reason, polymers are subjected to drying before processes such as injection and extrusion. Drying is carried out at temperatures and times specified by the polymer suppliers. Over-drying leads to increased viscosity (processing difficulty) while under-drying leads to flow marks, burrs, and reduced tensile strength and impact strength. Over-drying also causes excessive energy consumption. In this study, the drying process optimization ...Daha fazlası

Software Engineering Department Master Thesis

Mamadou Lamarana Diallo

Son zamanlarda yapay zeka (AI), sunduğu çözümler nedeniyle bilimsel araştırmaların tüm alanlarını işgal etti. Sağlık da bir istisna değil. Diyabet dünyadaki en yaygın hastalıklardan biridir. Komplikasyonlarından biri, hastanın görüşünü bulanıklaştırabilen veya bozabilen ve körlüğün ana nedenlerinden biri olan diyabetik retinopatidir. Diyabetik retinopatinin erken teşhisi tedaviye büyük ölçüde yardımcı olabilir. Yapay Zeka ve özellikle derin öğrenme alanındaki son gelişmeler, birçok hastalığı erken evrelerinde tahmin etmek, öngörmek ve teşhis etmek için kullanılabilecek iddialı çözümler sunmakt ...Daha fazlası

Application of Artificial Intelligence to Paper Based Colorimetric Sensors

Elif Yüzer

Kolorimetrik analiz, biyomoleküllerin saptanmasında kullanılan ve çıplak gözle gözlemlenen maddenin varlığını ve konsantrasyonunu belirlemek için renk değişimlerini kullanılarak ölçülmesini sağlayan bir tekniktir. Sağlık, gıda, ilaç, kozmetik ve boya endüstrileri gibi birçok farklı alanda biyolojik, tıbbi ve çevresel numunelerin analizinde kolorimetrik analiz kullanılmaktadır. Bakım-noktası (POC) testleri kullanım kolaylığı ve ulaşılabilirliği nedeniyle bir çok alanda kullanılmaktadır. POC testlerinin kolorimetrik analiz ile entegrasyonu kullanım kolaylığını arttırmaktadır. Kolorimetrik ...Daha fazlası

Automatic Detection of Alzheimer’s Disease with Transfer Learning and Image Processing Techniques

Atılay Yeşilada

Alzheimer's disease continues to be a serious public health issue. Alzheimer's disease affects generally old people, and it is a neurological disorder. The most significant sign of Alzheimer's disease is memory loss, which becomes worse over time. Therefore, early identification of Alzheimer's illness can aid in diseased getting the proper care. One of the aims of this thesis is to automate the early detection of Alzheimer's disease and create a self-improving system by utilising previous experiences each time it is detected automatically. In this way, the bad effects of the disease due to the ...Daha fazlası

Nonlinear analysis of steel frames considering lateral Torsional buckling effect

The use of nonlinear analysis methods is widely preferred to determine the realistic behavior of the structures through the computer technologies developed in recent years. However, the stability problems that significantly affect the structure behavior are often neglected in nonlinear analysis and in computer programs that are frequently used to the realization of these analyses. Lateral torsional buckling behavior, which is one of the stability problems, is considered to be neglected in many studies using nonlinear analysis methods in the literature. As a result, the realistic behavior ...Daha fazlası

Improvement of engineering properties and liquefaction reduction of sandy soils using electric arc furnace slag and roof tile powder

Agolli, Drinela

The scarcity of suitable land for construction of engineering facilities and shortage of natural earth aggregates has highlighted the need for finding innovative way of construction. Nowadays problematic soils such as: soft clay, organic soils and liquefiable soils can be improved to the required civil engineering requirements by application of soil stabilization. Soil stabilization is a method intended to increase or preserve the stability of soil mass and chemical alteration of soil to improve engineering properties. Generally, ground treatment techniques used are: densification, reinforcem ...Daha fazlası

Development and Characterization of Peptide Modified Bioactive Membrane for Regeneration of Corneal Endothelial Tissue

Gülşah Sunal

Because of the drawbacks related with the current corneal transplantation techniques such as donor shortage, rejection or disease transmission risk, there is still an urgent requirement in order to produce new treatments for regenerating corneal tissue. Specially, it is also necessary to overcome the problems related with the blindness and vision losses originating from damage to corneal endothelium. By considering these, corneal tissue engineering might possess an encouraging potential to ensure alternative and innovative approaches toward the development of effective corneal tissue equivalen ...Daha fazlası

Artificial Intelligent Based Segmentation on Medical Imaging

Mahmut AĞRALI

Teknolojinin ve teorik bilginin gelişmesiyle son 20 yıldır günlük hayatımızın her alanında olduğu gibi sağlık alanında da yapay zekâ (YZ)'nın uygulanması giderek artmaktadır. Özellikle koronavirüs (COVID-19) ve iskemik inme gibi toplum sağlığını tehdit eden hastalıkların YZ ile tespit edilmesi sağlık çalışanların iş yüklerini hafifleterek daha doğru tanı sağlamaktadır. Bu hastalıkların tespit edilmesinin yanında medikal görüntü üzerinden hastalık bölgelerinin bölütlenmesi, sağlık çalışanlarına tedavide yardımcı olabilecek bilgi sağlamaktadır. Bununla birlikte medikal görüntü üzerinden b ...Daha fazlası

Classification of Hand Gestures Using Time-Frequency Analysis and Different Artificial Intelligence Methods

Deniz Hande Kısa

Hand gesture-based systems are one of the most effective technological advances. Surface electromyography (sEMG) is utilized as a popular input data for gesture classification with elevated accuracy and advanced control capability. In this thesis, which is based on the classification performance of Hilbert-Huang spectrum (HHS) images obtained from Hilbert Huang Transform (HHT) of the sEMG of the gestures, an evaluation of the results of artificial intelligence (AI) methods on hand gesture classification using HHS image is presented.