Osman Erogul

Epileptic EEG detection using the linear prediction error energy

In this study, a method is proposed to detect epileptic seizures over EEG signal. For this purpose, a linear prediction filter is used to observe the presence of spikes and sharp waves on seizure EEG recordings. Linear prediction analysis calculates a coefficient set for each window, which can best model the applied time series signal. Modeling success is observed on the prediction error signal. The presence of spikes and other seizure-specific sharp waves on the signal reduces the modeling success and increases the prediction error of the filter. It is clearly observed that, the energy of prediction error signal during seizures is much higher than that of the seizure free intervals, which indicates the energy value and can be used to locate the seizure interval. The method is applied to 250 distinct EEG records, each of which has 23.6s duration. The results of the proposed algorithm are evaluated with the ROC analysis which indicates 93.6% success in detecting the presence of seizures. As a conclusion, the linear prediction error energy method can be considered as an efficient way to detect epileptic seizures on EEG records.

Efficient sleep spindle detection algorithm with decision tree

In this study, an efficient sleep spindle detection algorithm based on decision tree is proposed. After analyzing the EEG waveform, the decision algorithm determines the exact location of sleep spindle by evaluating the outputs of three different methods namely: Short Time Fourier Transform (STFT), Multiple Signal Classification (MUSIC) algorithm and Teager Energy Operator (TEO).The EEG records collected from patients used in this study have been recorded at the Sleep Research Center in Department of Psychiatry of Gulhane Military Medicine Academy. The obtained results are in agreement with the visual analysis of EEG evaluated by expert physicians. The method is applied to 16 distinct patients, 420,570 minutes long EEG records and the performance of the algorithm was assessed for the sleep spindles detection with 96.17% sensitivity and 95.54% specificity. As a result, it is found that the proposed sleep spindle detection algorithm is an efficient method to detect sleep spindles on EEG records.

Effects of Tonsillectomy on Speech Spectrum

Changes in the speech spectrum of vowels and consonants before and after tonsillectomy were investigated to find out the impact of the operation on speech quality. Speech recordings obtained from patients were analyzed using the Kay Elemetrics, Multi-Dimensional Voice Processing (MDVP Advanced) software. Examination of the time-course changes after the operation revealed that certain speech parameters changed. These changes were mainly F3 (formant center frequency) and B3 (formant bandwidth) for the vowel /o/ and a slight decrease in B1 and B2 for the vowel /a/. The noise-to-harmonic ratio (NHR) also decreased slightly, suggesting less nasalized vowels. It was also observed that the fricative, glottal consonant /h/ has been affected. The larger the tonsil had been, the more changes were seen in the speech spectrum. The changes in the speech characteristics (except F3 and B3 for the vowel /o/) tended to recover, suggesting an involvement of auditory feedback and/or replacement of a new soft tissue with the tonsils. Although the changes were minimal and, therefore, have little effect on the extracted acoustic parameters, they cannot be disregarded for those relying on their voice for professional reasons, that is, singers, professional speakers, and so forth.

Down Syndrome Diagnosis Based on Gabor Wavelet Transform

Down syndrome is a chromosomal condition caused by the presence of all or part of an extra 21st chromosome. It has different facial symptoms. These symptoms contain distinctive information for face recognition. In this study, a novel method is developed to distinguish Down Syndrome in a custom face database. Gabor Wavelet Transform (GWT) is used as a feature extraction method. Dimension reduction is performed with Principal Component Analysis (PCA). New dimension which has most valuable information is derived with Linear Discriminant Analysis (LDA). Classification process is implemented with k-nearest neighbor (kNN) and Support Vector Machine (SVM) methods. The classification accuracy is carried out 96% and 97,34% with kNN and SVM methods, respectively. Different from the studies related with the Down Sydrome, feature selection process is applied before PCA according to the correlation between components of feature vectors. Best results are achieved with euclidean distance metric for kNN and linear kernel type for SVM. In this way, we developed an efficient system to recognize Down syndrome.

Recognition of Down syndromes using image analysis

In the present study, image processing algorithms have been applied to face photos of the patients diagnosed by Down syndrome for development of a pre-diagnostic tool. The data sets evaluated in this study are collected from children whose ages range from 5 to 6. In each of normal syndrome groups; 18 photos of the children are analyzed. The critical points on faces are obtained by using elastic face bunch graph method for all photos. 10 feature vectors are applied to artificial neural network for both training and classification. In results Down syndrome can be pre-diagnosed with the accuracy of 68,7 percent by using neural network.

Automated Detection and Classification of Sleep Apnea Types Using Electrocardiogram (ECG) and Electroencephalogram (EEG) Features

1.1 Sleep and sleep disorders Sleep, which is defined as a passive period in organic physiology until the mid-20th century, is accepted to be an indispensable period of life cycle with today’s technological advances. While wakefulness is associated with the active excitation of Central Nervous System (CNS), sleep has been recognized as a passive period by the elimination of excitation. However, recent studies have shown that sleep is independent of wakefulness, generated by a sequence of changes in CNS, and a combination of five periods with clear boundaries. Sleep is not the disruption of daily life for a period of time or a waste of time. It is an active period which is important to renew our mental and physical health everyday and is covering one– third of our lives. Sleep activity is important for resting during the working period of basal metabolism of human body. The advances in technology enabling the measurement and quantification of brain activity make possible the micro and macro analysis of brain during both sleep and wakefulness states. With the studies investigating the CNS, it is observed the existence of some centrals causing the sleep by inhibiting the other regions of brain. As a result, sleep, which is an active and other state of consciousness, is a brain state of high coordination (Erdamar, 2007). Since breathing is established autonomously during sleep, it is affected by many anatomical and physiological parameters. Depending on this situation, various sleep disorders occur. There are more than eighty known sleeping diseases. Most of them cause person’s health to deteriorate and a decrease in life quality. As a result of the research carried out for many years, a list of sleep disorders, which are generally occurring, can be seen as in Table 1. Sleep disorders can be examined in two classes, parasomnia and dissomnia.

A new dynamic renal phantom and its application to scintigraphic studies for pixel basis functional radionuclide imaging

Various phantoms have been proposed in order to simulate the physical structures of human organs, such as those used in computerized brain tomography. Studying the functional behaviour of kidney by means of functional imaging techniques suffers from a lack of dynamic renal phantom for simulation. In this study a new dynamic renal phantom (DRP) is proposed and the first test results are reported which demonstrate the significance of deconvolution analysis in scintigraphy. The main idea in the construction of our DRP involves the filtration of chemical substances from the blood by flowing it through coiled tubes surrounded by semipermeable membranes. The DRP tests are performed with Technetium-99m (Tc-99m). The semipermeable membrane in the DRP passes Tc-99m, salts and small molecules but not blood cells and large protein molecules. The proposed DRP is tested using pixel basis renal functional radionuclide imaging techniques and promising results are obtained.

A clinical engineering approach for design and management of Central Sterilization Units

Central Sterilization Unit (CSU) are the units that performs sterilization of medical devices, instruments and consumables which used in hospitals and these units are planned to provide services within a quality management system and traceability. The numbers of sterilization procedures are carried out in medium and large scale hospitals, this situation can lead to reduced efficiency of the sterilization process have become critical. In this study, using a medium scale hospital as base, planning to work in coordination with the clinical engineering unit the structure of a central sterilization unit that coordinated to work with clinical engineering unit is recommended. The following issues are discussed in detail: architecture of the CSU, departments, staff, process of monitoring, validation and quality cycles. In addition, contributions to the technical efficiency of the sterilization process from biomedical engineers and technicians which are appointed by the clinical engineering units were examined.

Conventional and multi-state cellular neural networks in segmenting breast region from MR images: Performance comparison

Automated evaluation of MR images for breast density assessment or lesion localization requires accurate segmentation of breast region from regions of the body, such as the chest muscle, lungs, heart and ribs. Breast region segmentation is very complicated in the presence of background noise, intensity inhomogeneity and partial volume artifacts on MR images. Cellular neural networks (CNNs) are massively parallel cellular structures with locally interconnected cells and learning abilities and offer efficient ways to perform many complex medical image segmentation tasks. In this study, the performance of two breast region segmentation methods based on conventional CNNs and multi-state CNNs have been compared using non fat-suppressed T2-weighted bilateral axial images selected from 23 healthy women examined using a 3 Tesla MR scanner. The images provide a range of breast fat content representing 48 fatty, 61 fibroglandular or heterogeneously dense and 28 dense breast slices. Statistical analyses show that multi-state based method performs significantly better with high average precision, high true positive volume fraction, and low false positive volume fraction with an overall performance of 99.3±1.8%, 99.5±1.3%, and 0.1±0.2%, respectively.

Classification of sleep apnea types using EEG synchronization criteria

In this study, to obtain high quality signal features in discriminating the Central Sleep Apnea (CSA) and Obstructive Sleep Apnea (OSA) from controls, both linear and nonlinear EEG synchronization methods so called Coherence Function (CF) and Mutual Information (MI) are performed. For this purpose, sleep EEG series data collected from patients and healthy volunteers are classified by using a well known and widely used Feed-Forward Neural Network (FFNN) with respect to synchronic activities between C3 and C4 recordings. The results show that the degree of central EEG synchronization during night sleep is closely linked to sleep disorders like CSA and OSA. The MI and CF provide information in meaningful collaboration to support the clinical findings. These three groups were defined with a medical expert and can be very successfully classified by using the FFNN having two hidden layers with the average area of CF curves ranged form 0 Hz to 10 Hz and the average MI values are assigned as two features. This study is a preliminary study for classifying types of sleep apnea.