Boğaziçi University Library Digital Archive
The materials comprising the insitutional and private collections that are acquired by or donated to Boğaziçi University are reserved and opened to the access in compliance with copyright regulations. Digital Archive is a service of Boğaziçi University.
Communities in Digital Archive
Select a community to browse its collections.
- Bogazici University Theses
Recent Submissions
Predicting kidney tumor subtype from CT images using radiomics and clinical features
(Thesis (M.S.)-Bogazici University. Institute of Biomedical Engineering, 2022., 2022) Şirin, Duygu.; Güveniş, Albert.
This study aims to evaluate the performance of machine learning methods in predicting the subtype (clear-cell vs. non-clear- cell) of kidney tumors using clinical patient and radiomics data from CT images. CT images of 192 malignant kidney tumor cases (142 clear-cell, 50 other) from TCIA’s KiTS-19 Challenge were used in the study. There were several different tumor subtypes in the other group, most of them being chromophobe or papillary RCC. Patient clinical data were combined with the radiomic features extracted from CT images. Features were extracted from 3D images and all of the slices were included in the feature extraction process. Initial dataset consisted of 1157 features of which 1130 were radiomics and 27 were clinical. Features were selected using Kruskal Wallis - ANOVA test followed by Lasso Regression. After feature selection, 8 radiomic features remained. None of the clinical features were considered important for our model as a result. Training set classes were balanced using SMOTE. Training data with the selected features were used to train the Coarse Gaussian SVM and Subspace Discriminant classifiers. Coarse Gaussian SVM was faster compared to Subspace Discriminant with a training time of 0.47 sec and 11000 obs/sec prediction speed. Training duration of Subspace Discriminant was 4.1 sec with 960 obs/sec prediction speed. For Coarse Gaussian SVM was found as 0.86 while for Subspace Discriminant AUC was 0.85. Both models produced promising results on classifying malignant tumors as ccRCC or non-ccRCC. NOTE Keywords : Kidney Tumor, Clear-Cell, Machine Learning, CT Imaging.
Response analysis and damage mitigation of buried continuous pipes subjected to faulting actions
(Thesis (Ph.D.)-Bogazici University.Kandilli Observatory and Earthquake Research Institute, 2022., 2022) Perdibuka, Dardan.; Edinçliler, Ayşe.; Uçkan, Eren.
Buried continuous steel pipelines are critical lifelines failure of which under fault rupture incidents may lead to significant and deteriorating environmental and socio economic outcomes. Proper understanding and estimation of the mechanical behavior of buried steel pipes under such geohazards and investigation of means of mitigating these deleterious effects is of paramount importance. This thesis aimed at developing rigorous and simplified numerical models of the problem to realistically simulate the behavior of buried continuous pipes under strike-slip fault rupture-induced permeant ground deformations. The response of buried pipe cases under the fault load was investigated with respect to the variation of fault crossing angle (β) and pipe wall thickness (t). The second phase of this dissertation involved the investigation of the effect of four mitigation techniques to protect the buried pipe against fault rupture-induced damages. Lastly, a case study involving the evaluation of the effect of using CFRP wraps on the response of Thames Water Pipe which suffered great damage during the devastating 1999 Izmit is presented. The outcomes of this thesis indicate that the performance of the pipeline is sensitive to the variation of fault crossing angle and pipe wall thickness, increasing both parameters lead to overall improved pipe performance. Results indicate that all mitigation approaches offer certain degrees of improvement, where most effective mitigation approach is the wrapping of the pipeline surface with CFRP wraps while the use of controlled-low strength material was the least effective approach. Comparison of simplified and rigorous numerical models revealed that a good agreement exist between the approaches. Lastly, evaluation of the response of Thames Water Pipe protected using CFRP indicates that despite the considerable reduction in stresses and strains complete avoidance of failure for this particular case does not seem to be attainable.
Modular safety-critical control of legged robots
(Thesis (M.S.) - Bogazici University. Institute for Graduate Studies in Science and Engineering, 2022., 2022) Tosun, Berk.; Samur, Evren.
With recent performance improvements, legged robots will soon enter our lives to stay. Various control algorithms are already employed in deploying existing robots, and many more algorithms are still in the making. Safety concerns during the operation of legged robots must be addressed to enable their widespread use and ease their development. Especially machine learning- based control methods would benefit from model-based constraints to improve their safety. This thesis presents a modular safety filter to improve safety, i.e., reduce the chance of a fall of a legged robot. The availability of a robot capable of locomotion is assumed, i.e., a nominal controller exists. During locomotion, terrain properties around the robot are estimated through machine learning which uses a minimal set of proprioceptive signals. A novel deep-learning model utilizing an efficient transformer architecture is used for terrain estimation. A quadratic program combines the terrain estimations with inverse dynamics and a novel control barrier function constraint to filter and certify nominal control signals. The result is an optimal controller that acts as a filter and the filtered control signal allows the safe locomotion of the robot. The resulting approach is general and could be transferred with low effort to any other legged system.
A reinforcement learning based controller to minimize forces on the crutches of a lower-limb exoskeleton
(Thesis (M.S.) - Bogazici University. Institute for Graduate Studies in Science and Engineering, 2022., 2022) Utku, Aydın Emre.; Samur, Evren.; Öncü, Sinan.
The majority of the metabolic energy consumption of a lower-limb exoskeleton user comes from the upper body effort, since the lower body can be considered to be passive. However, the upper body effort of lower limb exoskeleton users is ignored during motion controller development process in the literature. In this thesis study, deep reinforcement learning is used to develop a locomotion controller that minimizes the ground reaction forces (GRF) on crutches. The rationale for minimizing the ground reaction forces is to minimize the upper body effort of the user. A model of the human-exoskeleton system with crutches is created in URDF and XML formats. Reward functions that encourage the forward displacement of the center of mass of the exoskeleton-human system without falling and extreme joint torques are shaped. The state-of-the-art methods, Twin Delayed Deep Deterministic Policy Gradient (TD3) and Proximal Policy Optimization (PPO), are employed with the RaiSim and MuJoCo physics simulators and with different algorithm specific parameters in multiple training trials. The employed networks generate the joint torques based on the joint angle and velocities along with the ground reaction forces on feet and crutch tips. These generated joint torques are directly sent to the exoskeleton model and a new state is observed after implementing the action that the deep RL framework provides. Policies trained by the TD3 and PPO methods on RaiSim are observed to fail to generate proper control commands for a stable and natural looking gait. In general, it is observed that the PPO method generated higher rewards than the TD3 method on RaiSim. After failing to develop a desired policy with RaiSim, MuJoCo is employed as the simulator. Eventually, a policy that can generate a reasonable gait with a desired crutch usage and with 35% minimization in GRFs with respect to the baseline policy is developed.
Exploring epistemic cognition in different task contexts
(Thesis (M.S.) - Bogazici University. Institute for Graduate Studies in Science and Engineering, 2022., 2022) Köksal, Alp.; Mercan, Fatih.
In this study epistemic performance in different task contexts was studied. There were, 2 physics professors, 4 teaching physics undergraduate students and 4 social sciences undergraduate students. Apt-AIR framework was adopted as the analytical framework. Data were collected using think aloud protocol and semi-structured interviews. In data analysis, (a) various examples for cognitive elements of epistemic performance were identified, (b) various behaviours related to metacognitive elements of epistemic performance were encountered and (c) comparisons were made between groups’ epistemic performances. The first task context was related with physics. Physicists have demonstrated a wide range of cognitive and metacognitive elements of epistemic performance. By doing so they have helped us to paint a picture of disciplinary characteristics of physics. All groups have identified experimentation and observations as reliable processes for producing or evaluating a knowledge claim. The theme of the second task context was mandatory covid vaccine. While evaluating knowledge claims and choosing sides, physicists have looked for the data. They have argued there were adequate amount of data during the time the video was shot. Teaching physics students, mostly trusted to their personal experiences and emotions. The frequency of demonstrated metacognitive elements of epistemic performance was lowest among teaching physics students compared to other two groups. The third group, social sciences students have kept well-being of the society at front while engaging with the second issue. The findings of the study tells us epistemic performance is contextual and Apt-AIR framework works as an analytical framework while capturing the cognitive and metacognitive elements of epistemic performance. The results suggest, there is a need to develop pre service teachers’ ability to perform apt epistemic performance.