Data mining plays an important role in a variety of fields including bioinformatics, multimedia, business intelligence, marketing, and medical diagnosis. Analysis of today’s huge and complex data involves several data mining algorithms including clustering and classification. The computational complexity of machine learning and data mining algorithms, that are frequently used in today’s applications such as embedded systems, makes the design of efficient hardware architectures for these algorithms a challenging issue for the development of such systems. The aim of this work is to optimize the performance of hardware acceleration for data mining applications in terms of speed and area. Most of the previous accelerator architectures proposed in the literature have been obtained using ad hoc techniques that do not allow for design space exploration, some did not consider the size (number of samples) and dimensionality (number of features in each sample) of the datasets. To obtain practical architectures that are amenable for hardware implementation, size and dimensionality of input datasets are taken into consideration in this work. For one-dimensional data, algorithm-level optimizations are investigated to design a fast and area-efficient hardware accelerator for clustering one-dimensional datasets using the well-known K-Means clustering algorithm. Experimental results show that the optimizations adopted in the proposed architecture result in faster convergence of the algorithm using less hardware resources while maintaining the quality of clustering results. The computation of similarity distance matrices is one of the computational kernels that are generally required by several machine learning and data mining algorithms to measure the degree of similarity between data samples. For these algorithms, distance calculation is considered a computationally intensive task that accounts for a significant portion of the processing time. A systematic methodology is presented to explore the design space of 2-D and 1-D processor array architectures for similarity distance computation involved in processing datasets of different sizes and dimensions. Six 2-D and six 1-D processor array architectures are developed systematically using linear scheduling and projection operations. The obtained architectures are classified based on the size and dimensionality of input datasets, analyzed in terms of speed and area, and compared with previous architectures in the literature. Motivated by the necessity to accommodate large-scale and high-dimensional data, nonlinear scheduling and projection operations are finally introduced to design a scalable processor array architecture for the computation of similarity distance matrices. Implementation results of the proposed architecture show improved compromise between area and speed. Moreover, it scales better for large and high-dimensional datasets since the architecture is fully parameterized and only has to deal with one data dimension in each time step. / Graduate / 2019-12-31

Cardiac pacing has been the only effective treatment in the management of patients with symptomatic bradycardia caused by sinus node dysfunction or atrioventricular block for decades. Conventional dual-chamber pacing is performed by implanting two leads in right atrial (RA) appendage and right ventricular (RV) apex separately. RV apex is the most commonly applied pacing site because it can be easily reached and allows a chronically stable position and stimulation thresholds. However, large randomized clinical trials have suggested that right ventricular apical (RVA) pacing may cause abnormal ventricular contraction and reduce pump function and lead to myocardial hypertrophy, in particular in patients with impaired left ventricular (LV) function. Recent studies have also reported a reduced LV systolic function in patients with pacing indications and preserved ejection fraction. The deleterious effects of RVA pacing on LV function may be related to the abnormal electrical and mechanical activation pattern or ventricular dyssynchrony. During RVA pacing, conduction of the electrical wave front propagates slowly through ventricular myocardium rather than through the His-Purkinje conduction system, comparable to left bundle branch block (LBBB). In addition , RVA pacing alters ventricular synchrony and loading conditions which may result in diastolic heart failure with abnormal LV relaxation, high filling pressure and low cardiac output state. Furthermore, it is possible that left atrial (LA) remodeling and reduction of atrial function may occur during RVA pacing . However, it is not been carefully studied. / Echocardiography is a convenient, non-invasive and established tool to assess cardiac function in clinical practice. Conventional two-dimensional echocardiography is useful to assess cardiac chamber size, volume and function. With the development of real time three-dimensional echocardiography (RT3DE) and color tissue Doppler imaging (TDI), echocardiography provides further valuable information and more accurate measurements which include myocardial velocity and parameters of dyssynchrony. In the present study, the main echocardiographic parameters including the maximal left atrial volume (LAVmax), pre-atrial contraction volume (LAVpre) and the minimal left atria l volume (LAVmin) were assessed by two-dimensional echocardiography. Peak systolic (Sm-la), peak early diastolic (Em-la), peak late diastolic (Am-la) velocities of left atrium (LA) and atrial conduction time (from onset of P wave on electrocardiogram to onset of atrial velocity) were measured by TDI. / In a cross-sectional study, ninety-eight patients who had been implanted with RVA-based dual-chamber pacemakers were enrolled. Four patients with pacing dependent were excluded. Eventually 94 patients were included in the final analysis. Echocardiography was performed (iE33, Philips) during intrinsic ventricular conduction (V-sense) and RVA pacing (V-pace) modes with 15 minutes between switching modes. We aimed to investigate if RVA pacing has any acute effects on atrial remodeling and function in patients with preserved ejection fraction (LV ejection fraction> 45%). The result showed that during V-pace, LA volumes increased significantly when compared with V-sense (LAVmax: 52.0 +/- 18.8 vs. 55.2 +/- 21.1 ml, p = 0.005; LAVpre: 39.8 +/- 16.4 vs. 41.3 +/- 16.6 ml, p = 0.014; LAVmin: 27.4 +/- 14.0 vs. 29.1 +/- 15.1 ml, p = 0.001) . TDI parameters showed significant reduction in Sm-la (3.0 +/- 1.1 vs. 2.7 +/- 0.9 cm/s, p < 0.01), Em-la (2.7 +/- 1.1 vs. 2.4 +/- 1.0 cm/s, p = 0.001). However, there was no change in Am-la. / In a prospective study, patients with symptomatic bradycardia, preserved ejection fraction, and received RVA pacing were recruited. Echocardiography was performed at both baseline and one year follow up through a standard protocol by experienced echocardiographers. LA volumes and velocities as well as intra- and interatrial dyssynchrony were measured offline with the use of dedicated software. The objectives of this study were to investigate: (1) if RVA pacing has any deleterious effects on LA remodeling and function during long-term follow up; (2) if RA appendage pacing has separate effects on atrial pump function, intra- and interatrial dyssynchrony; (3) if atrial dysfunction and dyssynchrony can predict atrial high rate episodes (AHREs) burden in the first year of RVA pacing. The main findings of this study were: (a) at one year follow up, LA volumes and indexes were increased with reduction in passive emptying fraction and total emptying fraction. Atrial velocities showed significant reduction when compared with baseline; (b) in multivariate regression analysis, the ratio of transmitral early diastolic filling velocity to mitral annular early diastolic velocity (E/e') > 15 at one year and reduction of LV ejection fraction ≥ 5% were independent predictors of reduction of Am-la > 30%; (c) high percent of RA appendage pacing prolonged atrial conduction and induced intra- and interatrial dyssynchrony. (d) Am-la < 5.3 cm/s can predict AHREs burden which had a sensitivity of 71% and specificity of 75%. / In conclusion, our studies suggest even short-term RVA pacing induces LA dilatation and impaired passive atrial function, though it did not have direct effect on active atrial contractility. However, chronic RVA pacing results in LA remodeling and reduces atrial function with decreased contractility. This was more likely to occur in those with impaired LV ejection fraction and evidence of diastolic dysfunction. Atrial dysfunction and interatrial dyssynchrony can predict AHREs burden after chronic RVA pacing. Therefore, measures that may minimize such adverse effect of pacing on atrial function need to considered for patients receiving RVA pacing, such as the use of new pacing modalities. / Xie, Junmin. / "December 2010"--Abstract. / Adviser: Yu Cheuk Man. / Source: Dissertation Abstracts International, Volume: 73-04, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2011. / Includes bibliographical references (leaves 142-161). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [201-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.

A new structure for the implementation of bit/serial adaptive IIR filter is presented. The bit level system consists of gated full adders for the arithmetic unit and data latches for the data path. This approach allows recursive operation of the IIR filter to be implemented without any global interconnections, minimal delay time, chip area and I/O pins. The coefficients of the filter can be updated serially in real time for time invariant and adaptive filtering. A fourth order bit/serial IIR filter is implemented on a 2 micron CMOS technology clocked at 55 MHz. / Graduation date: 1992

Thesis (Ph.D.), Memorial University of Newfoundland, 2000. / Includes index. Restricted until June 2001. Bibliography: leaves 82-84.

In the first half of this work, titin's role in cardiac function was studied using intact cardiac myocytes. The development of a carbon fiber based cell-attachment system allowed diastolic and systolic function of the isolated intact myocyte to be investigated. Addition of actomyosin inhibitor to the intact myocyte revealed that the majority of the cell's diastolic stiffness is due to titin but that actomyosin interaction exists as well and contributes ~ 30% of total diastolic stiffness. The details of this study are provided in chapter 1. Heart failure with preserved ejection fraction (HFpEF) accounts for up to 50% of total heart failure cases and is characterized by increased diastolic stiffness. An effective treatment for HFpEF does not exist. Reducing titin stiffness as a therapeutic strategy for lowering left ventricular (LV) chamber stiffness in HFpEF is currently under consideration. To understand the functional consequence of reduced titin stiffness on global cardiac function a Rbm20 Δᴿᴿᴹ mouse model was created. The Rbm20 Δᴿᴿᴹ model has deficiency in titin splicing that results in expression of very large and compliant titin isoforms in the sarcomeres. Study of Rbm20 Δᴿᴿᴹ cells revealed that cellular diastolic stiffness was inversely related to the size of titin and was reduced in a graded manner in Rbm20 Δᴿᴿᴹ heterozygous (+/-) and homozygous (-/-) cells. Importantly, reduced titin-based stiffness manifested in vivo as reduced LV chamber stiffness, which could be observed by echocardiography and pressure volume (PV) analysis. The systolic function of Rbm20 Δᴿᴿᴹ was studied by measuring the Frank-Starling mechanism (FSM), first at the intact myocyte level. The FSM was reduced in Rbm20 Δᴿᴿᴹ +/- and -/- with the largest reduction in -/- cells. PV analysis demonstrated a reduced FSM at the LV chamber level, consistent with the result at the cellular level. Surprisingly, exercise testing showed an enhanced exercise performance in cardiac specific Rbm20 Δᴿᴿᴹ +/- mice (relative to wild-type mice). Thus, this work indicates that increasing titin compliance improves diastolic function but negatively impacts systolic function. Importantly, findings suggest that the beneficial effect of improving diastolic function is a dominant effect. This work is described in Chapter 2.

This thesis presents the ASIC implementation of the RSA algorithm, which is one of the most widely used Public Key Cryptosystems (PKC) in the world. In RSA Cryptosystem, modular exponentiation of large integers is used for both encryption and decryption processes. The security of the RSA increases as the number of the bits increase. However, as the numbers become larger (1024-bit or higher) the challenge is to provide architectures, which can be implemented in hardware, operate at high clock speeds, use a minimum of resources and can be used in real-time applications. In this thesis, a semi-custom VLSI implementation of the RSA Cryptosystem is performed for both 512-bit and 1024-bit processes using 0.35&micro / m AMI Semiconductor Standard Cell Libraries. By suiting the design into a systolic and regular architecture, the broadcasting signals and routing delays are minimized in the implementation. With this regular architecture, the results of 3ns clock period (627Kbps) using 87K gates (8.7mm2 with I/O pads) for the 512-bit implementation, and 4ns clock period (237Kps) using 132K gates (10.4mm2 with I/O pads) for the 1024-bit implementation have been achieved. These results are obtained for the worst-case conditions and they include the post-layout routing delays. The design is also verified in real time using the Xilinx V2000E FPGA on the Celoxica RC1000 Hardware. The 1024-bit VLSI implementation has been sent to IMEC for fabrication as a prototype chip through Europractice Multi-Project Wafer (MPW) runs.

Abstract The heart responds to increased hemodynamic stress by increased cardiac myocyte size, enhanced protein synthesis and altered gene expression. Regulation of hypertrophic adaptation involves a number of neural and hormonal factors, which act on the cardiovascular system. The aim of the present study was to elucidate the regulation of gene expression of natriuretic peptides and adrenomedullin (AM) in cardiac overload in vivo. Furthermore, the interactions of AM and angiotensin II (Ang II) in cardiac function and development of left ventricular hypertrophy were studied both in vivo and in vitro. The effects of cardiac hypertrophy on the regulation of natriuretic peptides (atrial natriuretic peptide, ANP and B-type natriuretic peptide, BNP) and AM gene expression were studied during pressure overload in the hearts of two hypertensive rat strains, angiotensinogen-renin transgenic rats and spontaneously hypertensive rats as well as their normotensive control strains. Increased workload resulted in rapid upregulation of both BNP and AM gene expression in all rat strains; the response of AM was, however, augmented in hypertensive rats. Direct left ventricular wall stretch induced AM gene expression in isolated, perfused rat hearts, whereas stretching of cultured cardiac myocytes downregulated AM mRNA levels. In cultured cardiac cells exposed to Ang II, endothelin-1 or the α-agonist phenylephrine, Ang II-induced myocyte hypertrophy was selectively inhibited by AM. In vivo, AM interacted with Ang II in circulation by attenuating the hypertensive effects of Ang II, and in the heart by augmenting the Ang II-induced improvement in cardiac systolic function. However, AM had no direct modulatory effects on Ang II-induced left ventricular hypertrophy. These results show that cardiac wall stretch is a major stimulus for the early induction of AM gene expression in both normal and hypertrophied ventricle, and the response in hypertrophied myocardium is augmented. Furthermore, cardiac non-muscle cells may be involved in mediating effects of direct stretch. In vitro, AM acts as a selective inhibitor of Ang II-induced myocyte hypertrophy, suggesting a cardioprotective role for AM to counteract the local renin-angiotensin system and Ang II in cardiac hypertrophy and heart failure. Circulating AM appears to act mainly as a regulator of vascular tone and cardiac function.

Background: During incremental exercise, systolic blood pressure (SBP) increases due to increasing cardiac output. However, the impact of workload on SBP has often been overlooked. Indexing the increase in SBP to the increase in workload could provide a way of accounting for this. Athletes often reach higher maximal SBP (SBPmax) than untrained subjects, which has been attributed to their superior cardiac capacity. How this affects the relation between SBP and workload is not established. Aim: We sought to characterise the novel metrics SBP/VO2-slope and SBP/Watt-slope in endurance athletes and to analyse possible correlations between these metrics and maximal oxygen uptake (VO2max) in a population of endurance athletes and healthy, non-athletic subjects. We also sought to compare the SBP response of athletes to values predicted by newly published reference equations accounting for workload. Methods: In 24 endurance athletes and 5 healthy non-athletes we assessed the workload-indexed blood pressure response during a graded bicycle ergometer test. SBPmax was defined as the last SBP during exercise, VO2max as the mean of the two highest consecutive VO2 measurements at end of exercise. Results: The mean SBP/VO2-slope was 31.1 ± 9.7 mmHg/l/min and the mean SBP/Watt-slope was 0.28 ± 0.08 mmHg/Watt. We found no significant correlation between VO2max and the SBP/VO2-slope or the SBP/Watt-slope, nor with SBP at 50 W or at 200 W. In males there was a significant correlation between VO2max and SBPmax. The endurance athletes had less steep SBP/Watt-slopes and higher SBPmax than predicted by reference equations.  Conclusion: The SBP/VO2-slope offers a precise way of indexing blood pressure to workload and could provide a valuable tool in future studies investigating the SBP response to exercise. Our results suggest that different reference equations than in the general population might be needed when evaluating the SBP response in athletes.

Cardiovascular disease (CVD) is the leading cause of death worldwide. Pre-menopausal women have a lower incidence and severity of CVD compared to age-matched men. However, at the onset of menopause, CVD increases. A central feature in patients with CVD is excessive chronic sympathetic stimulation (CSS) of β-adrenergic receptors (β-AR’s). Clinical and animal studies show estrogen deficiency and age exacerbate cardiac β-AR signaling and contractile function. This led to the hypothesis that prolonged estrogen deficiency followed by CSS worsens left ventricular cardiac function and hypertrophy in the aged female heart. Female mice underwent bilateral ovariectomy or SHAM surgery at 2.5 months of age. At 12 months post-ovariectomy, mice were infused with Isoproterenol (400μg/kg/h) via mini-osmotic pumps for three days to induce CSS. This observation demonstrates prolonged estrogen deficiency worsens cardiac function and structure in aged female hearts. Thereby emphasizing the importance of clinical intervention and prevention for CVD in menopausal women.

Isolated systolic hypertension (ISH) is a growing health concern in the United States (US) black population. The stratified prevalence of untreated ISH has not been fully investigated in non-Hispanic blacks. Cross-sectional data on 4625 non-Hispanic blacks aged ≥18 years were collected from the National Health and Nutrition Examination Survey 1999-2010, representing a probability sample of the US civilian noninstitutionalized black population. The 6-year prevalence of ISH and 95% confidence intervals (CIs) were estimated by conducting weighted frequency and logistic procedures. The prevalence of untreated ISH was 11.2% among non-Hispanic black adults in 1999-2010. Individuals who received lower education (high school or below) had higher prevalence of untreated ISH than those with higher education (12.8% (95% CI: 11.3-14.2%) vs. 9.0% (95% CI: 7.5-10.6%)). The prevalence of untreated ISH was higher in young men than in young women (4.3% (95% CI: 3.3-5.4%) vs. 1.8% (95% CI: 0.9-2.7%)), and higher in middle-aged adults with lower education than in middle-aged adults with higher education (14.1% (95% CI: 11.4-16.7%) vs. 7.7% (95% CI: 5.5-9.8%)). Compared with 1999-2004, the prevalence of untreated ISH in 2005-2010 decreased for old individuals (27.7% vs. 40.8%), old men (24.4% vs. 40.0%) and old individuals who received higher education (21.4% vs. 40.7%). Untreated ISH is more prevalent in old blacks, and significant reduction of the prevalence in this group suggests that public health interventions, lifestyle modifications or health awareness are in the right direction.