Erkan Zeki Engin

Early prostate cancer diagnosis by using artificial neural networks and support vector machines

The aim of this study is to design a classifier based expert system for early diagnosis of the organ in constraint phase to reach informed decision making without biopsy by using some selected features. The other purpose is to investigate a relationship between BMI (body mass index), smoking factor, and prostate cancer. The data used in this study were collected from 300 men (100: prostate adenocarcinoma, 200: chronic prostatism or benign prostatic hyperplasia). Weight, height, BMI, PSA (prostate specific antigen), Free PSA, age, prostate volume, density, smoking, systolic, diastolic, pulse, and Gleason score features were used and independent sample t-test was applied for feature selection. In order to classify related data, we have used following classifiers; scaled conjugate gradient (SCG), Broyden-Fletcher-Goldfarb-Shanno (BFGS), and Levenberg-Marquardt (LM) training algorithms of artificial neural networks (ANN) and linear, polynomial, and radial based kernel functions of support vector machine (SVM). It was determined that smoking is a factor increases the prostate cancer risk whereas BMI is not affected the prostate cancer. Since PSA, volume, density, and smoking features were to be statistically significant, they were chosen for classification. The proposed system was designed with polynomial based kernel function, which had the best performance (accuracy: 79%). In Turkish Family Health System, family physician to whom patients are applied firstly, would contribute to extract the risk map of illness and direct patients to correct treatments by using expert system such proposed.

The classification of human tremor signals using artificial neural network

Tremor is an involuntary movement characterized by regular or irregular oscillations of one or several body segments. Physiological and pathological tremor in motor control can be defined as roughly sinusoidal movements with particular amplitude and frequency profiles. The electrophysiological analysis of human tremor has a long tradition. Tremor time series belongs to stochastic signals. This because the mechanism of generating them is so complex and exposed to so many uncontrollable influence that mathematical equations describing them contain random quantities. In this study, we concerned with tremor classification for the purpose of medical diagnosis. Accelerometer based tremor signals belong to Parkinsonian, essential, and healthy subjects were considered for this aim. Following features were extracted from tremor signals for classification by artificial neural network (ANN); linear prediction coefficients, wavelet transform detail coefficients, wavelet transform based entropy and variance, power ratio, and higher-order cumulants. Scaled-conjugate (SCG) and BFGS (Broyden-Fletcher-Goldfarb-Shanno) gradient learning algorithms were used. Despite BFGS algorithm had more sensitivity value (92.27%), SCG algorithm had more specificity value (89.01%). According to overall performance, BFGS algorithm (91.02%) was better than SCG algorithm (88.48%).

A Multivariate Analysis of Objective Voice Changes After Thyroidectomy Without Laryngeal Nerve Injury

OBJECTIVE To evaluate the impact of thyroidectomy and the possible effects of factors such as patient sex, operation type, and surgeon experience on objective voice parameters of patients undergoing thyroidectomy without laryngeal nerve injury. DESIGN Prospective study. SETTING University hospital. PATIENTS Thirty-six patients undergoing primary thyroidectomy because of thyroid disease. MAIN OUTCOME MEASURES The effect of thyroidectomy on voice was examined by recording the voices of the patients before and 1 week after thyroidectomy. The Multi-Dimensional Voice Program was used for capturing and analyzing the voice samples. RESULTS On postoperative examination of objective voice changes, thyroidectomy had no multivariate effect on the combination of voice parameters. Patient sex, type of surgery, and surgeon experience had no effect on the combination of voice parameters before and after thyroidectomy. Regardless of within-patient factors (type of surgery, patient sex, and surgeon experience), 4 acoustic parameters (highest fundamental frequency, standard deviation of average fundamental frequency, phonatory average fundamental frequency range in semitones, and degree of subharmonics) significantly decreased after thyroidectomy (P < .05). Although they tended to be worse, none of the acoustic parameters showed significant changes in male patients. However, significant changes in some of the acoustic parameters of female patients were observed. Highest fundamental frequency, standard deviation of average fundamental frequency, phonatory average fundamental frequency range in semitones, absolute jitter, relative average perturbation, pitch perturbation quotient, shimmer in decibels, percentage of shimmer, amplitude perturbation quotient, noise to harmonic ratio, and degree of subharmonics values were all lower in female patients after thyroidectomy (P < .05). CONCLUSIONS Voice changes may occur after thyroidectomy without any evident laryngeal injury, and deterioration and amelioration of acoustic parameters can be observed to occur differently among male and female patients. Preoperative and postoperative objective voice analyses may be helpful in documenting voice changes.

Wavelet Transformation Based Watermarking Technique for Human Electrocardiogram (ECG)

Nowadays, watermarking has become a technology of choice for a broad range of multimedia copyright protection applications. Watermarks have also been used to embed prespecified data in biomedical signals. Thus, the watermarked biomedical signals being transmitted through communication are resistant to some attacks. This paper investigates discrete wavelet transform based watermarking technique for signal integrity verification in an Electrocardiogram (ECG) coming from four ECG classes for monitoring application of cardiovascular diseases. The proposed technique is evaluated under different noisy conditions for different wavelet functions. Daubechies (db2) wavelet function based technique performs better than those of Biorthogonal (bior5.5) wavelet function. For the beat-to-beat applications, all performance results belonging to four ECG classes are highly moderate.

Portable heart rate monitoring system

Telemedicine is producing a great impact in the monitoring of patients located in non-clinical environments such as homes, gymnasiums, schools, remote military bases, ships, and rural area. A number of applications, ranging from data collection to chronic patient monitoring, and even to the control of therapeutic procedures, are being implemented in many parts of the world. As part of this growing trend, this paper explains the design of a portable heart rate monitoring system. A prototype system consists of analog data acquisition and a module which has a memory for recording heart rate values and corresponding time sequences. This module connected to computer via serial data communication protocol. At the computer side, we use interface software which is enable to graphically display the recorded heart rate values.

The evaluation of EEG response to photic stimulation in normal and diseased subjects

In this paper, our aim is to determine two photic stimulation frequencies, which would represent normal and diseased subjects, separately. Following features were extracted for this aim; linear prediction coefficients (LPC), subband wavelet entropy (SWE), subband wavelet variance (SWV), and relative power (RP). After extracting related features, analysis of variance (ANOVA) statistical test was used for the statistical evaluation of these features. According to the obtained results, wavelet transform-based entropy gave the best results to determine the representing stimulation frequencies. As a result, 29 Hz stimulation frequency was determined as the most representative frequency for normal subjects, whereas 8 Hz stimulation frequency was determined as the most representative frequency for diseased subjects.

Novel esophageal speech therapy method in total laryngectomized patients: biofeedback by intraesophageal impedance

The loss of the best communication port after total laryngectomy surgery makes speech rehabilitation an important goal. Our aim was to improve the quality of esophageal speech (ES) using online esophageal multichannel intra-luminal impedance (MII) as a new biofeedback method. Twenty-six total laryngectomized patients were included. Before ES therapy, an esophageal motility test was carried out. MII catheters were placed in all subjects who were then randomized into two groups. Group 1 included 13 cases, who were retrained according to the classical method. Group 2 included 13 cases, who were retrained according to the simplified animation of air movements within the esophagus and upper stomach resulting from the modifications of intra-esophageal air kinetics gained by MII. The level of speech proficiency was evaluated relative to pretraining levels using perceptual scales in the third and sixth months. Acoustic voice was analyzed. The number of syllables read per minute and the intelligibility of monosyllabic and dissyllabic words were calculated. In this study, MII was used for the first time in alaryngeal speech rehabilitation as a biofeedback method; an overall sufficient speech level was achieved by 68.4% at the end of therapy, whereas attendance was 90%. A statistically significant improvement was found in both groups in terms of ES level compared with the pretraining period although there was no significant difference between groups. Although we did not observe the expected difference between groups suggested by our hypothesis, MII may be used as an objective tool to show patients how to swallow and regurgitate air during training, and may thus expedite ES therapy both for the speech therapist and the patient in the future.

Estimation of oxygen saturation with laser optical imaging method

The aim of this study is to determine the estimation of hemoglobin concentration and oxygen saturation of tissue by non-invasively functional laser imaging for early skin cancer diagnosis. The early diagnosis of melanoma is a key factor that remarkably reduces the mortality rate. Diffuse reflectance spectroscopy is a very useful device for diagnosis and treatment purposes under in-vivo conditions. At this point, the aforementioned device, which takes into account the scattering of tissue, is to determine the concentration of chromophores (or optical absorbers) due to attenuated light strikes to the superficial layer of tissue. Laser-type light based imaging techniques in medical diagnosis substantially produce good results. So the aim of this study is to estimate HbO2 % and Hb% concentrations.

Image processing based vascular network analysis

Uncontrolled new blood vessel formation (angiogenesis) is really important step for the growth and metastasis of cancer cells. Nowadays, drugs which inhibit new vessel growth have been developed for cancer treatment. In cancer treatment management, it is vital to know which extend the vascularity reaches in order to guide the treatment. Vascular imaging devices and methods of treatment are available in clinical applications. However, many of them are invasively, non-portable and expensive systems. On the other hand, early diagnosis of skin and breast cancers gradually increased, which greatly increases the rate of survival. Hence, the need for a portable, cheap and non-invasive system plays an important role for the monitoring of early diagnostic conditions and/or the structure of a diagnosed tumor. To detect the new blood vessel formation, analysis of vessel complexity and width can be used on the recorded images by the systems. In this study; artificial vascular networks are used to determine the image processing techniques for using these analyses.

Determination of glucose concentration by pulsed laser based photoacoustic method

The aim of this study is to asist in the measurement of blood glucose concentration in an in-vitro medium by using the photoacoustic method. The tissue was mimicked with gel phantoms and different glucose concentrations were added. The sample which was in the setup was illuminated by a pulsed based laser. The acoustic vibrations that arose in the phantom medium were detected via electret microphone and processed with an electronical circuit and then recorded. When the results were evaluated, there was a correlation with glucose concentration proportions and with the photoacoustic signal amplitudes.