Abstract

Valvular regurgitation has been widely recognized as the central cause of morbidity and mortality. Even though the clinician can detect the presence of regurgitation by mere physical examination; diagnostic methods become inevitable while estimating the severity of valvular regurgitation and in the transformation of cardiac chambers as in reaction to the volume overload condition. Lately, the non-invasive recognition and assessment of the severity and etiology of valvular regurgitation has been facilitated through a promising new technology, the Echocardiography with Doppler. Accurate measurements of regurgitant volume in patients is of utmost importance since it aids in the estimation of the progression of the disease which in turn is vital for determining the optimal time for surgical repair or replacement.Color space conversion and anisotropic diffusion segmentation techniques have been utilized in this paper for the preprocessing stage of the quantification of mitral regurgitation. Flow field measurements were carried out with the aid of proximal flow convergence method. A calculated value of flow rate, regurgitant orifice area, regurgitant fraction and the regurgitant volume for a regurgitant orifice in the cardiovascular system can be obtained from the potential Color Doppler visualization of the flow convergence region. The research proposed provides a significant assessment of the echocardiographic and Doppler techniques employed in the evaluation of mitral valvular regurgitation in the patients. Additionally it also proffers the estimation of mildness, severity and eccentricity of mitral valvular regurgitation on basis of the scientific literature and a consensus of a panel of experts.

Abstract

Biometrics is the science of identifying a person using physiological or behavioral features. These features range from physical traits like fingerprints, faces, retina etc. to personal behaviors such as signatures. Compared to traditional methods, biometric features are much harder for intruders to copy or forge, and it is very rare for them to be lost. Hence, identification systems making use of biometric features that offer a much more secure and reliable performance. Many biometrics such as face, fingerprint and iris images, have been studied extensively for personal verification purposes in the past few decades. However, verification using vein patterns is less developed compared to other human traits.

This paper presents a hand vein authentication system using near infrared images of hand vein patterns. In order to evaluate the system performance, hand vein patterns from WASETdatabase were used. The database contains hand vein patterns of 100 different persons and of different gender, 5 images per person acquired at different intervals, for right hand. The images are of size 320x240 with a grey scale resolution of 8 bits per pixel. The data set is for normal persons who do not complain from any diseases such as arthritis. In verification testing analysis a single image is used to represent the template and 5 images for testing. Each of the 5 images is matched with a single template.

Unlike most biometric systems that carry out comparisons based on a pre-selected feature set, this proposed system directly recognizes the vein pattern by measuring their Hausdorff Distance. The modification suggested in the proposed matching classifier is found to improve the efficiency of the system. MATLAB tool has been used to realize the vein pattern recognition system.

Abstract

This paper is aimed at the selection of suitable wavelet family and hence suitable mother wavelet for the analysis of Fetal Phonocardiographic (fPCG) signals. Fourier based analyzing tools have some limitations concerning frequency and time resolutions. Although Wavelet Transform (WT) overcomes these limitations, it requires proper selection of mother wavelet. In this study the suitable mother wavelet is selected on the basis of properties of different wavelet families and characteristics of the fPCG signals. The Mean Squared Error (MSE) is used to evaluate the performance of the presented approach. The results show that, the fourth order Coiflets wavelet has a better performance for the analysis of fPCG  signals when using the rigorous SURE threshold algorithm with soft thresholding rule. The proposed approach is simple and proves to be effective when applied for the selection of suitable mother wavelet for analysis of the fPCG signals.These de-noised signals can be used for the accurate determination of Fetal Heart Rate (FHR) and further diagnostic applications of the fetus.

Abstract

The fetal heart sound serves as an effective diagnostic tool for a large class of fetal stress provocation. When the fetal heart sound is recorded and displayed graphically, the technique is known as fetal phonocardiography (fPCG). The fPCG signals, exhibit innate rhythms and periodicity that is more readily be appreciated in terms of frequency than time unit. This need arises to consider the fPCG signals, not only in terms of time but also in terms of frequency domain. The main advantage over the time domain representation is that, it allows a clear visualization of the signal frequency content and in many cases help to understand underplaying physiological phenomena. This work investigates methods to estimate frequency domain characteristics of fPCG signals. In this paper, various digital signal processing techniques are applied;compared and finally most appropriate technique is used for the spectrum inference of the fetal heart sound signal.

Abstract

Morphological Image Processing has became a powerful tool in Digital Image Processing, It allows processing of images to enhance image areas, segment objects, detect edge and analyze structures. It quantifies many aspects of the geometrical structure of the way that agrees with the human intuition and perception. The technique developed for binary images are a major step forward in the application of this theory to gray level images & color images. The Morphological Image Processing is based on geometrically altering image structure. In the binary image setting, an image is probed by one or more structuring elements to either extract information or to filter the image, and a similar probing also occurs in the grey scale image. Based on basic operations of Dilation, Erosion, Opening Closing, Top-hat, Morph- Gradient & Median filter one can construct a class of Morphological Image Processing Tool, which can be used in the place of a Linear Image Processing. Whereas the Linear Image Processing sometimes distort the underlying geometric form of an image, but in Morphological image Processing, the information of the image is not lost. In the Morphological Image Processing the original image can be reconstructed by using Dilation, Erosion, Opening and Closing operations for a finite no of times.The major objective of this paper is to reconstruct the class of such finite length Morphological Image Processing tool in a suitable mathematical structure using Java jdk 1.05 and also Matlab7.04.The Morphological Image Processing is implemented and successfully tested in Medical Image Analysis in detection of Tuberculosis & Segmentation of tree structure images such as retinal angiographic images.

Abstract

This paper describes a novel approach to the compression of fingerprint images using new mathematical transform, namely, the curvelet transform that has proven to show promising results over ridgelet and wavelet transforms. Wavelets are well suited to point singularities, however they have a problem with orientation selectivity, and therefore, they do not represent two-dimensional singularities (e.g. smooth curves) effectively. Ridgelets provide sparse representation to smooth objects with straight edges. But in image processing, edges are typically curved rather than straight. However at sufficiently fine scales, a curved edge is almost straight. By deploying ridgelets in a localized manner, a new transform considered in this paper has been developed and it can capture the curved edges effectively. A comparision has been carried out among Wavelets, Ridgelets and Curvelets based compression techniques for fingerprint image. A high quality compression has been achieved using curvelets compared to the existing Techniques.

Abstract

Biodiesel is one of the most promising alternative fuels for diesel engines because they are potential renewable, nontoxic, biodegradable, clean burning, high lubricity, low environmental impact, derived from vegetable oils and could be used directly in diesel engines without requiring extensive engine modifications. In the present work, the experiments have been conducted on a single cylinder, 4-stroke, Direct Injection constant speed diesel engine with different blends of Mahua Oil Methyl Ester (MOME) by varying the load to investigate the engine performance and exhaust emissions. From the investigation, it is concluded that the performance of Mahua oil methyl esters are good and results indicated that B20 has closer performance to diesel. The response surface models for rake ower (BP), carbon monoxide (CO), ydro arbons (HC), and oxides of nitrogen (NOx) emissions have been developed with the experimental data. The adequacy of the developed models is verified by using co-efficient of determination and analysis of variance (ANOVA) method. The results showed an average accuracy of 0.997 for brake power, 0.959 for NOx, 0.624 for CO, and 0.61 for HC emissions.

Abstract

This paper is focused to evaluate the effect of oxygenated diesel fuel on the engine performance and emissions, identify the oxygenates suitable for future diesel formulations, and to perform limited emissions and the performance testing of the oxygenated diesel blends and to select the suitable oxygenated compounds for future diesel formulations. From the literature survey and initial screening of 71 oxygenates, the following compounds are selected based on oxygen content, flashpoint etc. Diethyl carbonate, 2 Ethoxy ethanol and Diethyl ether - oxygenated compounds [1] These selected oxygenates are blended with diesel fuel in proportions of 2.5, 5 and 7.5 % by volume and experimental study is conducted in a single cylinder naturally aspirated Direct injection diesel engine. An eddy current dynamometer is coupled with the engine and the experimental study is carried out to evaluate the effect of the three oxygenated compounds on emissions and performance [2]. Comparison of performance of plain diesel and blended diesel with different oxygenated compounds having different mixing levels of oxygenates is carried out. The performance of oxygenates on power, brake thermal efficiency, cylinder peak pressure, HC, CO and NO_X , smoke level and particulate  matter are studied. All the three oxygenated compounds are found to be effective at reducing Smoke and Particulate levels by 30 to 40 % in 7.5 % blending, 2 Ethoxy ethanol is more effective than the other two compounds. Comparing the cost and performance Diethyl carbonate gives optimum value and Diethyl ether follows. There is a moderate increase in power output and also in thermal efficiency.