Ahmet Emir Kavak Year: 2011
Advisor: Prof. Yekta Ülgen
Abstract: Air quality monitoring in operating rooms is of prime importance because particles that carry microbiological contamination generate serious risks during surgical operations. Air quality monitoring is executed by two separate methods: Microbiological sampling and particle counting. The aim in this project is to investigate statistical correlation between these two methods. If so, particle-counting technique could replace frequent microbiological sampling. Microbiological sampling and particle counting techniques are applied in five operating rooms located at three different hospitals with a total number of 360 measurements taken before (at rest) and during surgery. Spearman's correlation coefficient is used for measuring the level of correlation between two methods. The bacteria counts are classified after their sizes and tested for size-by-size correlation. Then, the bacterial counts measured at the same site are cumulatively added together and correlated with the particle counts at each particle size-range. No correlation is found when size-by-size correlation is performed. When cumulative bacteria counts are considered in 'at-rest' conditions, the number of particles in 5.0-10.0 μm and 10.0-25.0 μm size ranges correlated with bacteria counts. Particles of 1.0-5.0 μm size ranges and particles larger than 25.0 μm correlated with bacteria counts during surgery.
Thesis No: 334 Autostereoscopic Displays in Computer Assisted Surgery
Caner Gümüş Year: 2011
Advisor: Prof. Mehmed Özkan
Abstract: Using images for diagnosis, therapy and surgery is a widely used option as evidenced by many articles in the literature. In conventional systems standard two-dimensional (2D) displays are used to view the images even though the images are three-dimensional (3D) reconstructed. In this study, we used a computerized surgical assistant whose core functionalities are provided by an open source software package called Slicer. Slicer uniquely integrates several facets of image quided therapy into a single environment and has capabilities for visualization, surgical planning and guidance. We performed basic tasks of computer assisted surgery such as registration, segmentation, surface model generation and 3D visualization on medical images and displayed the reconstructed images on an autostereoscopic display so that the images with depth can be viewed without the need to wear any special eyeglasses or headgear.
Thesis No: 335 Design Of A 32” Color TFT LCD PACS Monitor and Its Clinical Evaluation Through
ROC Analysis
Sadık Hakan Kayabaşı Year: 2011
Advisor: Assoc. Prof. Albert Güveniş
Abstract: One of the most important stages in digital radiology process is the transfer of the image to the observer, as the variations in light and color from a physical display. The common practice in observing these visual results is to use a medical grade LCD monitor. The main problem with medical grade monitors is their high cost. First objective of the study was to design a 32” LCD PACS monitor to be as compatible as possible to a medical grade LCD monitor with a remarkable cost advantage. The second was to test the hypothesis for a significant difference between a medical grade LCD monitor and the designed one in terms of diagnostic image quality. After the design’s validation, 60 digital radiographs with definite findings were obtained in cooperation with the authors of a previous study. Three experienced radiologists from Acıbadem Hospital examined these radiographs both on a medical grade Reference branded LCD monitor and on the 32” design. To check observers’ performance, the receiver operating characteristic (ROC) curves for all monitor-reader cases were statistically compared by using the same content and observers. The area under curve (AUC) of each ROC curve was used as a metric for detecting lung modules in the radiographs. With 95% confidence interval, the hypothesis was tested for a significant statistical difference between the related monitors. AUC for Reference monitor for observer 1, 2 and 3 were calculated as 0.634, 0.703 and 0.755 respectively. AUC for 32” design were 0.811, 0.746 and 0.811 For observer 1, the 32”design showed superior performance. For observer 2 and 3, Though AUC was far better on behalf of the new design; no significant statistical difference could be proven. As a result, it is possible to implement the new 32” design as a PACS monitor for medical diagnosis purposes without sacrificing any diagnostic value.
Thesis No: 336 Intraoperative Measurement Of Human Spastic Gracilis Muscle Isometric Forces as
a Function Of Knee Angle
Fatma Oya Aytürk Year: 2011
Advisor: Assoc. Prof. Can Yücesoy
Abstract: Spasticity is a neuromuscular disease which is associated with increased muscle tone, stiffness and impaired motor control and consequently functional limitations. Improved understanding of spasticity requires the collection of substantial directly measured length- force characteristics of spastic muscles. Studies including direct measurement of human muscle force are very rare due to limited access to the muscle. With the method developed in this study, isometric length (knee-angle)-force characteristics of human spastic Gracilis muscle are measured intraoperatively for the first time in literature. Experimental data is collected during the surgical operations performed by Prof. Dr. Yener Temelli and his group in Istanbul University School of Medicine. In 7 subjects (average ager: 8±4.6), isometric muscle forces are measured by buckle force transducers at five different knee angles (of 120°, 90°, 60°, 30° and 0°). Mean peak Gracilis muscle force and mean optimal knee angles are measured to be 41.19±41.07 N and 30±31.6° respectively. Knee-angle force characteristics of 7 subjects showed inter-subject variability and peak Gracilis muscle forces were not correlated with the anthropometric data of subjects. Gracilis muscle exerted non-sero force in each condition indicating that functional joint range of motion is at least as wide as from full knee extension to 120° of knee flexion. A finding of major importance is that knee angle-force characteristics of spastic Gracilis muscle are found to be not representative of the pathological condition occuring at the joint and are comparable with the ones obtained from healthy subjects in a previous study of our research group. Moreover, length history of muscle was shown presently to affect muscle force in most of the subjects.
Thesis No: 337 Software Development for X-Ray Fluoroscopy and MR Image Fusion
Emre Özdal Year: 2011
Advisor: Assoc. Prof. Cengizhan Öztürk
Abstract: In interventional radiology, treatment is routinely done under X-Ray Fluoroscopy (XF) using special catheters or needles. XF is a fast modality and allows one to follow the endovascular path with contrast injections and to track devices during the interventions. On the other hand, one of the most important problems in XF imaging is the lack of soft tissue contrast; besides the obvious problem of ionizing radiation. During the intervention, additional soft tissue information could decrease the risk by providing important extra guidance to the surgeon. For instance: in endovascular cardiac interventions, detailed anatomical positions of infarcted segments of the heart could highlight target or weak zones on the myocardial wall. Main goal of this study is to overcome the lack of soft tissue contrast of XF. To achieve this goal soft tissue information gathered from a priori imaging modality is used. MRI is the best candidate with excellent soft tissue contrast and ability to image any cross section in the scanned volume. On the other hand, MRI has deficient features in temporal resolution when compared to XF, instrument compatibility problems with high magnetic and RF fields and limited patient access because of the shape of the magnet bore. Therefore combination of these two modalities all deficient features could be avoided by complementing each other. Combined usage of XF and MRI requires registration of images from both modalities and a reliable software platform is needed for preclinical and clinical studies. In this work, a new implementation of XFM Suite in Extensible Imaging Platform (XIP) software provides all the tools for X-Ray fused with MRI (XFM) clinical studies performed in X-Ray-MRI suites (XMR) with real-time registration and fusion of images.
Thesis No: 338 Calculating The Magnitude Of Transfer Functions For Hearing Aids By Insertion Gain
Measurements
Yunus Karamavuş Year: 2011
Advisor: Prof. Mehmed Özkan
Abstract: While the hearing loss is one of the most common problem in our lives, it is also the most compensable disability. In general, hearing impairment can be fixed by a hearing aid which amplifies sound levels at different frequencies. A transfer function of a hearing aid determines how much the input signal will be amplified at certain frequencies. However, this function may vary depending on whether the hearing aid is worn or not. Therefore, when a hearing aid is in use, expected results are mostly not achieved if the calibration is done without wearing it. This phenomena is known as a fitting problem in literature and due to the altered transfer function artifacts a small incidence of people having hearing abnormalities prefer using a hearing aid. In this thesis, to compensate the fitting problem, we designed and developed a hearing aid analyzer which uses a Real Ear Measurement (REM) method. Determining the electroacustic performance of hearing aid in situ is essential for ideal fitting and it varies according to ear shape, ear canal, and hearing sensitivity. In REM, an insertion gain measurement is obtained by inserting a probe microphone into the ear canal and calculating the gain between input and output signals. Fine tuning is performed using the transfer function based on insertion gain measurement. REMs allow determining individual based, actual characteristics of hearing aids. The superiority of our system comes from the fact that it is capable of measuring the transfer function while the hearing aid is in situ. Moreover, our design is battery powered and small in diameter which makes it portable. Finally, our system complies with international standards.
Thesis No: 339 Endovenous Laser Ablation: Different Lasers And Delivery Techniques
Meral Filiz Somunyudan Year: 2011
Advisor: Assoc. Prof. Murat Gülsoy
Abstract: Endovenous Laser Ablation (EVLA) has become a popular minimally invasive alternative to stripping in the treatment of saphenous vein reflux. Several wavelengths have been proposed; of which 810, 940 and 980- nm are the most commonly used. Ther- mal shrinkage of collagenous tissue during EVLA plays a significant role in the early and late results of the treatment. Longer wavelengths (>1000-nm) show greater water absorption than shorter wavelengths and may have some advantages for EVLA due to water molecules in the vein lumen cells. However, the most appropriate wavelength is still the subject of debate. The laser light delivery technique is another criteria that affect the success of EVLA. Bare fiber, which is generally used for medical application, delivers the laser light locally focused to application area. Radial fiber, which is a new technique for laser light delivery, emanates the laser energy homogenously circularly to vessel lumen. The aim of this study is to compare the efficacy of 980 and 1940-nm laser wavelengths in the treatment of varicose veins. In this study, 980 and 1940-nm lasers were used to irradiate stripped human veins. Different power settings (8/10W for 980- nm, 2/3W for 1940-nm) and laser delivery techniques (radial and bare fiber) were used to compare their effects. As a conclusion, 1940-nm TM-fiber laser and radial fiber are promising methods in the treatment of varicose veins.
Thesis No: 340 Effects Of Contactor Size And Stimulation Distance On The Response Properties Of
Rapidly Adapting Tactile Fibers Innervating The Rat Glabrous Skin
İsmail Devecioğlu Year: 2011
Advisor: Assoc. Prof. Burak Güçlü
Abstract: Recent population models used space-invariant attenuation functions. But this assumption is not enough to construct realistic population models, because mechanical properties of the skin vary along the skin [1]. In this study, it is hypothesized that response profiles of mechanoreceptive Fibers are not symmetric along proximo-distal axis due to the varying mechanical properties of the skin. In this study, sinusoidal mechanical vibrations were applied perpendicular to the skin of adult rats. Single-unit responses were recorded from sciatic nerve. Five different stimulus locations (2 distal, 1 RF center, 2 proximal) and three different contactor sizes (area: 0.39 mm2, 1.63 mm2, 2.96 mm2) were used. Averages of absolute spike thresholds (a0) and entrainment thresholds (a1) of rapidly adapting fibers were plotted as a function of stimulation distance and contactor size. Also, mechanical impedance of the skin was measured at 5 different locations on distal phalanx of one rat. 2-way ANOVA showed that the effect of stimulus location on a0 and a1 was significant (p<0.001), whereas contactor size had no significant effects (p=0.642). Post- hoc Tukey test showed that thresholds for proximal stimulations were higher than those for distal stimulations. A mechanical model which explains the mechanical behavior of the skin and its effects on mechanoreceptive fibers' response is presented. According to these results rate-intensity functions of RA fibers shift asymmetrically along the proximodistal axis, which suggests that the mechanical stimulus is transmitted better towards the proximal direction. This may counterbalance the effects of innervation density and cause uniform psychophysical detection thresholds for the finger.
Thesis No: 341 Graph Theoretical Analysis Of Functional Human Brain Network
Adem Umut Günebakan Year: 2011
Advisor: Prof. Ahmet Ademoğlu
Abstract: Recent studies suggest that, human brain has a small-world behavior which is reflected by locally and globally efficient processing. To investigate this behavior a short term/working memory experiment was designed which had manipulation and retention conditions. Our goal was to be able to explore the differences between the organizations of the brain during the execution of these tasks within a graph theoretical context. The retention condition required the subject only to remember a visually applied stimulus whereas the manipulation condition to visually manipulate the stimulus before keeping it in mind. Brain activity information was recorded through the electroencephalography (EEG) device. After preprocessing, the collected EEG data was decomposed into classical frequency bands and phase locking values (PLV) between each pair of electrodes were computed with the help of the Hilbert Transform. After applying threshold to PLV matrices to build binary unweighted networks, the graph theoretical analysis was applied to determine and compare the main dynamics of functional coupling during manipulation and retention. This analysis was carried out in a time dependent manner to better monitor the variations of these dynamics. It was found that the brain exhibited a highly efficient behavior in the local and global sense both during manipulation and retention; and thus the brain had a small-world characteristic during the execution of these tasks. The statistical analysis revealed significant differences between the efficiency values of retention and manipulation. The analysis of node and edge centrality values of different frequency bands showed prominent effects in the upper alpha gamma bands. The finding of this thesis study supports the feasibility of the graph theoretical analysis for analyzing complex brain networks.
Thesis No: 342 Bayesian Modeling And Interference For Functional Magnetic Resonance Imaging Of
The Visual Cortex
Meltem Sevgi Year: 2011
Advisor: Prof. Cengizhan Öztürk
Abstract: For the most effective use of functional magnetic resonance imaging (fMRI), mapping the brain signals to a statistically valid map is crucial. The common approach to create a statistic at each voxel is applying the frequentist or the classical statistics. However, there are many challenges raised by the use of classical statistics to test the functional data such as the multiple comparison problem, and the limitation in the interpretation of the parameters. As an alternative, a Bayesian approach can be used to assess the data based on the posterior probability distributions of the parameters. In this study, the power of Bayesian inference was compared against classical inference in random effect analyses: A group data collected from visually stimulated volunteers was assessed following a simulation study. In order to assess the results of the statistical inference for the group level, the variation of the effect sizes with respect to stimulus frequency was used. A comparison was performed between the change in the effect sizes of lateral geniculate nuclei (LGN) and primary visual area (V1) during graded visual stimulation by using the posterior probability maps (PPMs) with an effect size threshold of zero. This comparison became possible with the fact that once we had the posterior probabilities the activity in LGN was able to be visualized by changing the effect size threshold and without decreasing the significance threshold, which is not possible to achieve with classical inference where the data is tested against the null hypothesis. Despite of the small magnitude of activation in LGN we could show the connectivity between V1 and LGN and the differences in response characteristics during graded visual stimulation.
Thesis No: 343 Volumetric MRI Analysis Of Thalamic Stroke Patients
Pınar Özel Year: 2012
Advisor: Prof. Ahmet Ademoğlu
Abstract: The thalamic pain syndrome is a weakly deduced phenomenon thet develops as a complication of asmall stroke in the thalamus. Because this syndrome results in many different physiologinal disturbances, it iscritical to understand its physiological basis to diagnose and to treat it. It is important to determine the whole volume of the lesion and to label it with reference to an anatomical atlas in order to identify the stroke region more accurately. In this thesis, we develop a software tool to help the neuroradiologists inperpret the stroke region more accurately by estimating its volume and identifying its anatomical label. To reduce the anatomical variability, the first step consists of spatial registration and normalization of brain images. This is achieved using the procedure implemented in the Statistical Parametric Mapping (SPM) package in Matlab. Normalized MR images are used to identify the lesions in the brain, to calculate its volume and to label it in a graphical user environment. Three types of neuropsychological tests i.e. are Mini – Mental State Examination (MMSE), Frontal Behavioral invertory (BDI) are evaluated in estimating their correlations with the total volume of selected regions. Neuroradiologists can potentially benefit from the sorfware to diagnose and to treat the thalamic stroke patients. Furthermore, such a platform may help the clinicans to interact with each other distantly by exchanging more objective information about the clinical neuroanatomical assessment of their patients. Additionally, correlation analysis of neuropsychological tests and stroke volume in the Talairach atlas plays a role for the neuropsyhologinal and neuroanatomical data fusion for better assessment.
Thesis No: 344 Tactile Mental Rotation in Blindfolded and Congenitaly Blind Subjects
Betül Polat Year: 2012
Advisor: Assoc. Prof. Burak Güçlü
Abstract: Mental rotation is the process of imagining an object rotated into a different orientation in space. This well-known visual phenomenon may be used to understand the cognitive processing by applying it to the tactile modality. Linear correlation between response times and angular orientations of the explored objects shows the mental rotation effect. Twelve sighted, 12 congenitally blind subjects participated in this study. All subjects were right handed. Gender and age were balanced. The sighted were blind- folded through the experiments. Two tactile L-shaped objects were glued on cardboards as pairs rotated at five orientations (0o, 45o, 90o, 135o, 180o). A passive touch method was developed, subjects' hands were steady on a platform and objects were placed and lifted with a lever. Subjects used their palms to passively touch the objects. The subjects were instructed to explore the objects tactually and decide if the pairs were same or different - different meaning mirror as known to the experimenter. Response times and accuracies were recorded. Correlation analysis and ANOVA were performed using Matlab. Results showed that both the blindfolded and congenitally blind subjects used mental rotation process during tactile exploration of the stimuli. The results support the idea that an analog representation is used in the cortex which totally lacked visual input. The data presented in this study, combined with the literature further supports the hypothesis that spatial properties of the objects are encoded similarly for touch and vision.
Thesis No: 345 Development Of an Acqusition and Image Enhancement Platform for Digital Dental Imaging
Abdükadir Yazıcı Year: 2012
Advisor: Prof. Cengizhan Öztürk
Abstract: With technologıcal developments in X-ray imaging, digital systems started to replace analog systems. Principle advantages of digital X-ray imaging are; it has lower cost (in the long run), is ready to be processed with a computer, could acquire 3D images and has better imaging quality even at low radiation doses. Usage of digital technology in dental imaging is also increasing rapidly. Although, dental dose levels are significantly less than other X-ray imaging techniques, its cumulative effects could be harmful in the long term. Main aim of this thesis is to set up a digital dental imaging system and to implement its image processing software. A digital dental X- ray system consists of three of the shelf components; a digital intraoral X-ray sensor, a portable X-ray tube and a computer. The new digital system and its acquisition software are tested on phantom and cadaver tests. Furthermore, effects of several image enhancement techniques on low dose dental radiographic images are examined.
Thesis No: 346 1940-nmThulium Fiber Laser Ceramic Bracket Debonding
İrem Demirkan Year: 2012
Advisor: Assoc. Prof. Murat Gülsoy
Abstract: The aim of the study was to determine the proper laser parameters for 1940-nm Thulium Fiber Laser for ceramic bracket removing. In order to assess the effectiveness of 1940-nm Thulium Fiber Laser in orthodontic ceramic bracket debonding, poly- crystalline ceramic brackets were bonded to mandibular bovine teeth with adhesive agent. The samples were divided into 9 different groups due to applied laser power and laser duration, debonding method used. There was a control group that had no laser application. The efficiency of the laser was investigated together with the required debonding forces and intrapulpal temperature changes. In this study, keeping intrapulpal temperature changes below the threshold value that is accepted 5, 5 ◦C must be accepted as a must. In most of the lasing groups, the increases in intrapulpal temperature changes were observed almost below the threshold value 5,5◦C. The findings revealed that 1940-nm Thulium Fiber Laser irradiation could reduce the needed debonding force or SBS (shear bond strength) values significantly compared to control group. Irradiation of the specimens by 1940-nm Thulium Fiber Laser caused more than 50% reduction in the needed debonding force when compared to the control group. Different application methods: non-scanning and scanning were studied to assess the effects of the distinct configurations. Scanning method was tried to reduce the intrapulpal temperature rise during laser irradiation but in this study side effects of this method were faced. It was revealed that different application methods did not create any remarkable differences. In more than 50% of samples with energies 25 J or more, adhesive remnant hasn't been observed on enamel surfaces for the laser groups.
Thesis No: 347 Implementation of Tomosynthesis in Dental X-Ray Imaging
Share with your friends: |