Journal of Medical Signals and Sensors
Volume:3 Issue: 3, Jul-Sep 2013

Due to the daily mass production and the widespread variation of medical X-ray images, it is necessary to classify these for searching and retrieving proposes, especially for content-based medical image retrieval systems. In this paper, a medical X-ray image hierarchical classification structure based on a novel merging and splitting scheme and using shape and texture features is proposed. In the first level of the proposed structure, to improve the classification performance, similar classes with regard to shape contents are grouped based on merging measures and shape features into the general overlapped classes. In the next levels of this structure, the overlapped classes split in smaller classes based on the classification performance of combination of shape and texture features or texture features only. Ultimately, in the last levels, this procedure is also continued forming all the classes, separately. Moreover, to optimize the feature vector in the proposed structure, we use orthogonal forward selection algorithm according to Mahalanobis class separability measure as a feature selection and reduction algorithm. In other words, according to the complexity and inter-class distance of each class, a sub-space of the feature space is selected in each level and then a supervised merging and splitting scheme is applied to form the hierarchical classification. The proposed structure is evaluated on a database consisting of 2158 medical X-ray images of 18 classes (IMAGECLEF 2005 database) and accuracy rate of 93.6% in the last level of the hierarchical structure for an 18-class classification problem is obtained.

In this study, a method for determining the location and extent of myocardial infarction using Body Surface Potential Map data that is obtained from PhysioNet challenge 2007 database has been suggested. This data is related to four patients with myocardial infarction that we used from two patients as training set to determine rules, and from two other patients for testing set and the conclusion of the proposed model. First T-wave amplitude, T-wave integral, Q-wave amplitude and R-wave amplitude as four features of ECG signals were extracted. Then with definition and application of several rules and threshold levels for those features, areas that are with myocardial infarction and their extent were diagnosed. To determine the precise location and extent of myocardial infarction, 17-segments standard model of left ventricle was used. Finally, overall accuracy of this method that expressed with SO parameter, CED parameter and EPD parameter was obtained to 1.16, 1 and 5.3952 respectively for two patients in test set. Two main advantages of this method are simplicity and high accuracy.

The main purpose of this paper is to introduce afast method for gene prediction in DNA sequences based on the period-3 property in exons. First, the symbolic DNA sequences are converted to digital signal using the EIIP method. Then, to reduce the effect of background noise in the period-3 spectrum, we use the Discrete Wavelet Transform (DWT) at three levels and apply it on the input digital signal. Finally, the Goertzel algorithm is used to extract period-3 components in the filtered DNA sequence. The proposed algorithm leads to increase the speed of process and therefor reduce the computational complexity. Detection of small size exons in DNA sequences, exactly, is another advantage of our algorithm. The proposed algorithm ability in exon prediction is compared with several existing methods at the nucleotide level using: (i) specificity - sensitivity values; (ii) Receiver Operating Curves (ROC); and (iii) area under ROC curve. Simulation results show that our algorithm increases the accuracy of exon detection relative to other methods for exon prediction.

Most of the erythrocyte related diseases are detectable by hematology images analysis. At the first step of this analysis, segmentation and detection of blood cells are inevitable. In this study, a novel method using a line operator and watershed algorithm is rendered for erythrocyte detection and segmentation in blood smear images, as well as reducing over-segmentation in watershed algorithm that is useful for segmentation of different types of blood cells having partial overlap. This method uses gray scale structure of blood cell, which is obtained by exertion of Euclidian distance transform on binary images. Applying this transform, the gray intensity of cell images gradually reduces from the center of cells to their margins. For detecting this intensity variation structure, a line operator measuring gray level variations along several directional line segments is applied. Line segments have maximum and minimum gray level variations has a special pattern that is applicable for detections of the central regions of cells. Intersection of these regions with the signs which are obtained by calculating of local maxima in the watershed algorithm was applied for cells’ centers detection, as well as a reduction in over-segmentation of watershed algorithm. This method creates 1300 sign in segmentation of 1274 erythrocytes available in 25 blood smear images. Accuracy and sensitivity of the proposed method are equal to 95.9% and 97.99%, respectively. The results show the proposed method’s capability in detection of erythrocytes in blood smear images.

The Monte Carlo method is the most accurate method for simulation of radiation therapy equipment. The linear accelerators are currently the most widely used machines in radiation therapy centers. In this work, a Monte Carlo modelling of the Siemens ONCOR linear accelerator in 6 MV and 18 MV beams was performed. The results of simulation were validated by measurements in water by ionization chamber and EDR 2 film in solid water. The linac’s x-ray particular are so sensitive to the properties of primary electron beam. Square field size of 10 × 10 cm2 produced by the jaws was compared with ionization chamber and film measurements. Head simulation was performed with BEAMnrc and dose calculation with DOSXYZnrc For film measurements and 3ddose file produced by DOSXYZnrc analyzed used homemade MATLAB program. At 6 MV, the agreement between dose calculated by Monte Carlo modelling and direct measurement was obtained to the least restrictive of 1%, even in the build-up region. At 18 MV, the agreement was obtained 1%, except for in the build-up region. In the build-up region, the difference was 1% at 6 MV and 2% at 18 MV. The mean difference between measurements and Monte Carlo simulation is very small in both of ONCOR x-ray energy. The results are highly accurate and can be used for many applications such as patient dose calculation in treatment planning and in studies that model this linac with small field size such as Intensity Modulated Radiation Therapy technique.

Substitution of arterial with venous blood samples to estimate blood gas status is highly preferable due to practical and safety concerns. Numerous studies support the substitution of arterial by venous blood samples, reporting strong correlations between arterial and venous values. This study further investigated the predictive ability of venous blood samples for arterial Acid-Base Balance (pH) and pressure of carbon dioxide (pCO2). Participants were 51 post-brain surgery patients receiving mechanical ventilation, who had blood samples taken simultaneously from radial artery of the wrist and elbow vein. showed significant associations between arterial and venous pH and pCO2. However, the variation of regression residuals was not homogenous, and the regression line did not fit properly to the data, indicating that simple linear regression is sub-optimal for prediction of arterial pH and pCO2 by venous blood sample. Although highly significant correlations were found between arterial and venous blood pH and pCO2, the results did not support the reliability of prediction of arterial blood pH and pCO2 by venous blood samples across a range of concentrations.

The speech is an easily accessible signal which clearly represents the characteristics of larynx and vocal folds. Therefore, application of some proper machine learning algorithms on a small part of a recorded speech signal may help in non-invasive diagnosing of vocal fold diseases. Since there are some experimental evidences that suggest the existence of chaotic behavior in speech production system, in this paper a new method is proposed to predict vocal fold pathologies using its chaotic characteristics. The proposed method is based on modeling of pathological voice as a speech trajectory in the reconstructed phase space.

Studies show that any complications including hemorrhage, lack of blood supply, lack of oxygen supply, and death of cells in a tissue, will have a clear effect on electrical properties of that tissue. Thus, by measuring impedance of a set of tissues, potential problems of the damaged tissue may be found. Since electrical impedance is closely related to the measuring frequency, obviously, every tissue exhibits its own specific impedance according to its electrical properties at each frequency. This research project investigates design and manufacture method of a device for measuring tissue impedance at different frequencies. To this end, design of a multi frequency sinusoidal current source is required. This current source is built using a single harmonic Generator sample (DDS AD9835) with working frequency (design-point frequency) between 1 Hz and 10 MHz and accuracy of 1 Hz, and microcontroller (PIC16F628) capability. For measurement and display of tissue impedance, ARM AT91SAMs256 microcontroller was used. Thus, with this hardware created, it shows that there are significant impedance changes between mouse tissues.