Biomedical Engineering

Aim: The objective of this study is to compare the image quality of flat panel systems that have cesium iodide (CsI) and gadolinium oxysulfate (GOS) scintillators in their detectors. Method: CDRAD 2.0 phantom and CDRAD analyzer software is used for objective evaluation and comparison of image quality of flat panel systems with CsI and GOS scintillators. The image quality is investigated in terms of contrast and detail. Nine different flat panel systems from 6 different manufacturers are evaluated in this study. Four of these flat panels have CsI scintillators and the remaining 5 have GoS scintillators. For evaluations, 20 layers of plexiglass were placed on the top (10 layers) and bottom (10 layers) of the CDRAD 2.0 phantom to simulate a patient. 3 images were taken from each system at each dose level, which were analyzed by the CDRAD 2.0 analyzer software. Image qualities were investigated at four different dose levels (50, 100, 150, and 200 µGys). IQFinv is used as the quality metric. Results: IQFinv values of GOS systems have little variance both within the same system and between all systems. On the other hand, CsI systems have higher variance in IQFinv values within the same system and between the systems. In addition, same CsI detectors (same model from the same manufacturer) used in different systems resulted in considerably different IQFinv values. CsI systems demonstrate 4-5 times more improvement in IQFinv value with increasing dose levels compared to GOS systems. Finally, IQFinv values of CsI systems are higher than GOS systems with statistically significance (p<0.029).

Choosing the correct ventilator settings for the treatment of patients with respiratory tract disease is quite an important issue. Since the task of specifying the parameters of ventilation equipment is entirely carried out by a physician, physician's knowledge and experience in the selection of these settings has a direct effect on the accuracy of his/her decisions. Nowadays, decision support systems have been used for these kinds of operations to eliminate errors. Our goal is to minimize errors in ventilation therapy and prevent deaths caused by incorrect configuration of ventilation devices. The proposed system is designed to assist less experienced physicians working in the facilities without having lung mechanics like cottage hospitals.

Choosing the correct ventilator settings for the treatment of patients with respiratory tract disease is quite an important issue. Since the task of specifying the parameters of ventilation equipment is entirely carried out by a physician, physician's knowledge and experience in the selection of these settings has a direct effect on the accuracy of his/her decisions. Nowadays, decision support systems have been used for these kinds of operations to eliminate errors. Our goal is to minimize errors in ventilation therapy and prevent deaths caused by incorrect configuration of ventilation devices. The proposed system is designed to assist less experienced physicians working in the facilities without having lung mechanics like cottage hospitals.

Background: Digital breast tomosynthesis (DBT) is an emerging imaging modality which produces three-dimensional radiographic images of breast. DBT reconstructs tomographic images from a limited view angle, thus data acquired from DBT is not sufficient enough to reconstruct an exact image. It was proven that a sparse image from a highly undersampled data can be reconstructed via compressed sensing (CS) techniques. This can be done by minimizing the l 1 norm of the gradient of the image which can also be defined as total variation (TV) minimization. In tomosynthesis imaging problem, this idea was utilized by minimizing total variation of image reconstructed by algebraic reconstruction technique (ART). Previous studies have largely addressed 2-dimensional (2D) TV minimization and only few of them have mentioned 3-dimensional (3D) TV minimization. However, quantitative analysis of 2D and 3D TV minimization with ART in DBT imaging has not been studied. Methods: In this paper two different DBT image reconstruction algorithms with total variation minimization have been developed and a comprehensive quantitative analysis of these two methods and ART has been carried out: The first method is ART + TV 2D where TV is applied to each slice independently. The other method is ART + TV 3D in which TV is applied by formulating the minimization problem 3D considering all slices.

Background: After the release of compressed sensing (CS) theory, reconstruction algorithms from sparse and incomplete data have shown great improvements in diminishing artifacts of missing data. Following this progress, both local and non-local regularization induced iterative reconstructions have been actively used in limited view angle imaging problems. Methods: In this study, a 3D iterative image reconstruction method (ART + TV) NLM was introduced by combining local total variation (TV) with non-local means (NLM) filter. In the first step, TV minimization was applied to the image obtained by algebraic reconstruction technique (ART) for background noise removal with preserving edges. In the second step, NLM is used in order to suppress the out of focus slice blur which is the most existent image artifact in tomosynthesis imaging. NLM exploits the similar structures to increase the smoothness in the image reconstructed by ART + TV.

Chronic kidney disease can be detected with several automatic diagnosis systems. In this study, chronic kidney diseases are diagnosed with Adaboost ensemble learning algorithm. Decision tree based classifiers are used in the diagnosis. The classification performance are evaluated with kappa, mean absolute error (MAE), root mean squared error (RMSE) and area under curve (AUC) criterias. Considering the performance analyses, it is observed that adaboost ensemble learning algorithm provides better classification performance than individual classification.

Chronic kidney disease is a rising health problem and involves conditions that decrease the efficiency of renal functions and that damage the kidneys. Chronic kidney disease may be detected with several classification techniques, and these have been classified using various features and classifier combinations. In this study, we applied seven different classifiers (Naïve Bayes, HoeffdingTree, RandomTree, REPTree, Random Subspaces, Adaboost, and IBk) for the diagnosis of chronic kidney disease. The classification performances are evaluated with five different performance metrics, i.e., accuracy, kappa, mean absolute error (MAE), root mean square error (RMSE), and F measures. Considering the classification performance analyses of these methods, six reduced features provide a better and more rapid classification performance. Seven individual classifiers are applied to the six features and the best results are obtained using individual random tree and IBk classifiers.

The application of EEG-based emotional states is one of the most vital phases in the context of neural response decoding. Emotional response mostly appears in the presence of visual, auditory, tactile, and gustatory arousals. In our work, we use visual stimuli to evaluate the emotional feedback. One of the best performing methods in emotion estimation applications is the common spatial patterns (CSP). We implement CSP method in addition to the conventional Welch power spectral density-based analysis. Experimental results and topographies on the collected EEG data show that the CSP spatial filtering method implies the relationship between EEG bands, EEG channels, neural efficiency and emotional stimuli types.

With the technological development in the world especially in computers, a computer simulation was
observed with almost all things around as in a lot of fields such as engineering, medical, chemical,...etc., so can a
computer understand the human's emotions? Emotions are intrinsically connected to the way that people interact
with each other. Emotion constitutes a major influence in determining human behaviors. A human being can
read the emotional state of another human, and behave in the best way to improve their communication at that
moment. This is because emotions can be recognized through words, voice intonation, facial expression, and
body language. Can a computer understand the human's brain? Making the computer more empathic to the
human is one of the aspects of affective computing. The computer can actually take a look inside the human’s
head to observe their mental state. In this study, we will understand how can a computer understand human
brain by understanding the electric signals that come from it.