Murat Hamit

Feature Extraction and Classification on Esophageal X-Ray Images of Xinjiang Kazak Nationality

Esophageal cancer is one of the fastest rising types of cancers in China. The Kazak nationality is the highest-risk group in Xinjiang. In this work, an effective computer-aided diagnostic system is developed to assist physicians in interpreting digital X-ray image features and improving the quality of diagnosis. The modules of the proposed system include image preprocessing, feature extraction, feature selection, image classification, and performance evaluation. 300 original esophageal X-ray images were resized to a region of interest and then enhanced by the median filter and histogram equalization method. 37 features from textural, frequency, and complexity domains were extracted. Both sequential forward selection and principal component analysis methods were employed to select the discriminative features for classification. Then, support vector machine and K-nearest neighbors were applied to classify the esophageal cancer images with respect to their specific types. The classification performance was evaluated in terms of the area under the receiver operating characteristic curve, accuracy, precision, and recall, respectively. Experimental results show that the classification performance of the proposed system outperforms the conventional visual inspection approaches in terms of diagnostic quality and processing time. Therefore, the proposed computer-aided diagnostic system is promising for the diagnostics of esophageal cancer.

Liver hydatid CT image segmentation based on localizing region active contours and modified parametric active contours

Liver hydatid disease is a common parasitic disease in farm and pastoral areas and seriously damages peoples health. Based on CT imaging features of this disease, we perform experiments on the segmentation of the Liver hydatid CT image. The experimental results show that: according to the intensity distribution characteristics expressed by different tissues in liver hydatid CT image, localizing region active contours and modified parametric active contours is practical to liver and hydatid lesions segmentation simultaneously from liver hydatid CT images. we obtain the optimal segmentations and improve the internal and external statistical smoothness of liver hydatid CT image by adjusting local selected area size and a reasonable location dimension should be commonly employed because of the sensitive of the initialization.

Feature extraction and analysis on Xinjiang uygur medicine image by using color histogram

With the rapid development of multimedia technology and network technology and wide application of digital image, more and more attention has been paid for Content-based image retrieval technology. For a long time Xinjiang uygur hospitals and medical institutions accumulated a large amount of underutilized data of uygur medicine. In this paper, the image color histogram feature of botanical and animal drugs of Xinjiang uygur medicine has been extracted. First, the image size has been normalized, and extract the color histogram and analyse color histogram characteristics with statistics method, at last, the classification ability of features is evaluated by Bayes discriminant analysis. Experimental results show that high accuracy for botanical image classification is existed by using color histogram feature. This study would have a certain extent for the content-based medical image retrieval for Xinjiang uygur medicine.

Feature extraction and analysis on X-ray image of Xinjiang Kazak Esophageal cancer by using gray-level histograms

Image feature extraction technology has been widely used in medical field, but there is no related research on image feature extraction of high morbidity of esophageal cancer in Xinjiang. In this paper, feature extraction based on gray-scale histograms was applied in X-ray barium angiogram in high morbidity of esophageal cancer in Xinjiang. For the X-ray barium angiogram, converting RGB image to the grayscale intensity image, through median filter to remove the image noise and using histogram equalization to enhance the contrast; then the gray-level histograms was used to get the features of the images, and the classification ability of features was evaluated by Bayes discriminant analysis. The result show that feature classification ability was different when classifying different images, which provides a new direction for the research of computer aided diagnosis system for the high incidence of Kazak esophageal cancer in Xinjiang Uygur Autonomous Region.

Assessment of pain with mechanical nociceptive stimuli by the change of arterial wall impedance

The understanding and treatment of pain is one of the oldest challenges in the field of clinical medicine. In this study, as a first step toward adequate pain assessment, we propose a method to evaluate the reactions of the automatic nervous system in response to painful stimuli by observing arterial wall impedance. Under the proposed method, the mechanical impedance (stiffness) of the arterial wall is calculated from blood pressure and photoplethysmogram measurements on a beat-to-beat basis. In the experiments, we tested eight male subjects (aged 22-23) by applying external forces (1-3 [N]) to the central parts of their palms as painful stimuli, and evaluated changes in levels of arterial wall stiffness during stimulation. The results indicated that stiffness during stimulation showed a significant increase (p = 0.007, p = 0.014 and p = 0.018 for the stimulus changes from 0 to 1 [N], 1 to 2 [N] and 2 to 3 [N] for all subjects). We also compared the coefficients of variation in the measured stiffness and visual analog scale (VAS) values during stimulation, and found that the mean coefficients of variation for stiffness (0.37, 0.27 and 0.26 for the stimuli of 1, 2 and 3 [N] for all subjects, respectively) were smaller than the ones of the VAS values (0. 67, 0.51 and 0.50, respectively). From these results, it was confirmed that changes in the level of measured stiffness can be used to quantify the level of pain felt by a patient.

All-optical switching characteristics analysis in an optimized nonlinear Bragg grating with a π phase shift

We numerically study the nonlinear switching characteristics of optical transmission through optimized fiber Bragg grating with a π phase shift. The nonlinear coupled-mode equations were solved numerically based on the time-dependent transfer-matrix method. The result shows that the π phase shift grating is superior to the uniform grating in the enhancement, corresponding to the local intensity of the light inside the grating. It shows that the use of π phase shift gratings reduces effectively the switching threshold, but the on-off contrast is generally declined which can be generally improved through the introduction of tapered parameters. In addition, the narrowed transmitted pulse for positive-tapered nonlinear Bragg grating is a Bragg soliton owing to the balance of anomalous group velocity dispersion and self-phase modulation (SPM). It can be found that the tapered nonlinear Bragg grating with a π phase shift is more preferable for achieving the larger on-off contrast.