Statistical methods for analysis and processing of medical ultrasound: applications to segmentation and restoration

In this thesis two major topics inherent with medical ultrasound images are addressed: deconvolution and segmentation. In the first case a deconvolution algorithm is described allowing statistically consistent maximum a posteriori estimates of the tissue reflectivity to be restored. These estimates are proven to provide a reliable source of information for achieving an accurate characterization of biological tissues through the ultrasound echo. The second topic involves the definition of a semi automatic algorithm for myocardium segmentation in 2D echocardiographic images. The results show that the proposed method can reduce inter- and intra observer variability in myocardial contours delineation and is feasible and accurate even on clinical data.

Automated Analysis Methods for the Assessment of Bicycle Infrastructure and Demand

A first phase of the research activity has been related to the study of the state of art of the infrastructures for cycling, bicycle use and methods for evaluation. In this part, the candidate has studied the "bicycle system" in countries with high bicycle use and in particular in the Netherlands. Has been carried out an evaluation of the questionnaires of the survey conducted within the European project BICY on mobility in general in 13 cities of the participating countries. The questionnaire was designed, tested and implemented, and was later validated by a test in Bologna. The results were corrected with information on demographic situation and compared with official data. The cycling infrastructure analysis was conducted on the basis of information from the OpenStreetMap database. The activity consisted in programming algorithms in Python that allow to extract data from the database infrastructure for a region, to sort and filter cycling infrastructure calculating some attributes, such as the length of the arcs paths. The results obtained were compared with official data where available. The structure of the thesis is as follows: 1. Introduction: description of the state of cycling in several advanced countries, description of methods of analysis and their importance to implement appropriate policies for cycling. Supply and demand of bicycle infrastructures. 2. Survey on mobility: it gives details of the investigation developed and the method of evaluation. The results obtained are presented and compared with official data. 3. Analysis cycling infrastructure based on information from the database of OpenStreetMap: describes the methods and algorithms developed during the PhD. The results obtained by the algorithms are compared with official data. 4. Discussion: The above results are discussed and compared. In particular the cycle demand is compared with the length of cycle networks within a city. 5. Conclusions

The Law and Economics of Hedge Fund Regulation: A Comparison between the U.S. and the EU

This doctoral dissertation seeks to assess and address the potential contribution of the hedge fund industry to financial instability. In so doing, the dissertation investigates three main questions. What are the contributions of hedge funds to financial instability? What is the optimal regulatory strategy to address the potential contribution of hedge funds to financial instability? And do new regulations in the U.S. and the EU address the contribution of hedge funds to financial instability? With respect to financial stability concerns, it is argued that despite their benefits, hedge funds can contribute to financial instability. Hedge funds’ size and leverage, their interconnectedness with Large Complex Financial Institutions (LCFIs), and the likelihood of herding behavior in the industry can potentially undermine financial stability. Nonetheless, the data on hedge funds’ size and leverage suggest that these features are far from being systemically important. In contrast, the empirical evidence on the interconnectedness of hedge funds with LCFIs and their herding behavior is mixed. Based on these findings, the thesis focuses on one particular aspect of hedge fund regulation: direct vs. indirect regulation. In this respect, a major contribution of the thesis to the literature consists in the explicit discussion of the relationships between hedge funds and other market participants. Specifically, the thesis locates the domain of the indirect regulation in the inter-linkages between hedge funds and prime brokers. Accordingly, the thesis argues that the indirect regulation is likely to address the contribution of hedge funds to systemic risk without compromising their benefits to financial markets. The thesis further conducts a comparative study of the regulatory responses to the potential contribution of hedge funds to financial instability through studying the EU Directive on Alternative Investment Fund Managers (AIFMD) and the hedge fund-related provisions of the Dodd-Frank Wall Street Reform and Consumer Protection Act of 2010.

Deep learning for computer vision: a comparison between convolutional neural networks and hierarchical temporal memories on object recognition tasks

In recent years, Deep Learning techniques have shown to perform well on a large variety of problems both in Computer Vision and Natural Language Processing, reaching and often surpassing the state of the art on many tasks. The rise of deep learning is also revolutionizing the entire field of Machine Learning and Pattern Recognition pushing forward the concepts of automatic feature extraction and unsupervised learning in general. However, despite the strong success both in science and business, deep learning has its own limitations. It is often questioned if such techniques are only some kind of brute-force statistical approaches and if they can only work in the context of High Performance Computing with tons of data. Another important question is whether they are really biologically inspired, as claimed in certain cases, and if they can scale well in terms of "intelligence". The dissertation is focused on trying to answer these key questions in the context of Computer Vision and, in particular, Object Recognition, a task that has been heavily revolutionized by recent advances in the field. Practically speaking, these answers are based on an exhaustive comparison between two, very different, deep learning techniques on the aforementioned task: Convolutional Neural Network (CNN) and Hierarchical Temporal memory (HTM). They stand for two different approaches and points of view within the big hat of deep learning and are the best choices to understand and point out strengths and weaknesses of each of them. CNN is considered one of the most classic and powerful supervised methods used today in machine learning and pattern recognition, especially in object recognition. CNNs are well received and accepted by the scientific community and are already deployed in large corporation like Google and Facebook for solving face recognition and image auto-tagging problems. HTM, on the other hand, is known as a new emerging paradigm and a new meanly-unsupervised method, that is more biologically inspired. It tries to gain more insights from the computational neuroscience community in order to incorporate concepts like time, context and attention during the learning process which are typical of the human brain. In the end, the thesis is supposed to prove that in certain cases, with a lower quantity of data, HTM can outperform CNN.

Assessment of right ventricular systolic function: Comparison between cardiac magnetic resonance derived ejection fraction and pulsed-wave tissue Doppler imaging of the tricuspid annulus

Systolic right ventricular (RV) function is an important predictor in the course of various congenital and acquired heart diseases. Its practical determination by echocardiography remains challenging. We compared routine assessment of lateral tricuspid annular systolic motion velocity (TV(lat), cm/s) using pulsed-wave tissue Doppler imaging from the apical 4-chamber view with cardiac magnetic resonance (CMR) as reference method.

A Semiempirical Transition State Structure for the First Step in the Alkaline Hydrolysis of Cocaine. Comparison between the Transition State Structure, the Phosphonate Monoester Transition State Analog, and a Newly Designed Thiophosphonate Transition State Analog

Semiempirical molecular orbital calculations have been performed for the first step in the alkaline hydrolysis of the neutral benzoylester of cocaine. Successes, failures, and limitations of these calculations are reviewed. A PM3 calculated transition state structure is compared with the PM3 calculated structure for the hapten used to induce catalytic antibodies for the hydrolysis of cocaine. Implications of these calculations for the computer–aided design of transition state analogs for the induction of catalytic antibodies are discussed.

Association between variants of the leptin receptor gene (LEPR) and overweight: a systematic review and an analysis of the CoLaus study

Background Three non-synonymous single nucleotide polymorphisms (Q223R, K109R and K656N) of the leptin receptor gene (LEPR) have been tested for association with obesity-related outcomes in multiple studies, showing inconclusive results. We performed a systematic review and meta-analysis on the association of the three LEPR variants with BMI. In addition, we analysed 15 SNPs within the LEPR gene in the CoLaus study, assessing the interaction of the variants with sex. Methodology/Principal Findings We searched electronic databases, including population-based studies that investigated the association between LEPR variants Q223R, K109R and K656N and obesity- related phenotypes in healthy, unrelated subjects. We furthermore performed meta-analyses of the genotype and allele frequencies in case-control studies. Results were stratified by SNP and by potential effect modifiers. CoLaus data were analysed by logistic and linear regressions and tested for interaction with sex. The meta-analysis of published data did not show an overall association between any of the tested LEPR variants and overweight. However, the choice of a BMI cut-off value to distinguish cases from controls was crucial to explain heterogeneity in Q223R. Differences in allele frequencies across ethnic groups are compatible with natural selection of derived alleles in Q223R and K109R and of the ancient allele in K656N in Asians. In CoLaus, the rs10128072, rs3790438 and rs3790437 variants showed interaction with sex for their association with overweight, waist circumference and fat mass in linear regressions. Conclusions Our systematic review and analysis of primary data from the CoLaus study did not show an overall association between LEPR SNPs and overweight. Most studies were underpowered to detect small effect sizes. A potential effect modification by sex, population stratification, as well as the role of natural selection should be addressed in future genetic association studies.

Transmission of Chlamydia trachomatis through sexual partnerships: a comparison between three individual-based models and empirical data

Chlamydia trachomatis is the most common bacterial sexually transmitted infection (STI) in many developed countries. The highest prevalence rates are found among young adults who have frequent partner change rates. Three published individual-based models have incorporated a detailed description of age-specific sexual behaviour in order to quantify the transmission of C. trachomatis in the population and to assess the impact of screening interventions. Owing to varying assumptions about sexual partnership formation and dissolution and the great uncertainty about critical parameters, such models show conflicting results about the impact of preventive interventions. Here, we perform a detailed evaluation of these models by comparing the partnership formation and dissolution dynamics with data from Natsal 2000, a population-based probability sample survey of sexual attitudes and lifestyles in Britain. The data also allow us to describe the dispersion of C. trachomatis infections as a function of sexual behaviour, using the Gini coefficient. We suggest that the Gini coefficient is a useful measure for calibrating infectious disease models that include risk structure and highlight the need to estimate this measure for other STIs.

Extra-articular fractures of the proximal phalanges of the fingers: a comparison of 2 methods of functional, conservative treatment

For nonsurgical treatment of fractures of the proximal phalanges of the triphalangeal fingers, different dynamic casts have been described. The main principle behind these casts is advancement and tightening of the extensor hood, caused by a combination of blocking the metacarpophalangeal joints in flexion and actively flexing the proximal interphalangeal joints. In contrast to established treatment protocols using functional forearm casts, the Lucerne cast allows for free mobilization of the wrist joint. The purpose of the current multicenter study was to compare the results of conservative, functional treatment using 2 different methods, either a forearm cast or a Lucerne cast.

Exogenous gonadotropins have little impact on follicular but considerable effect on serum cytokine concentrations – a comparison between Natural Cycle and stimulated IVF using a multiplexed assay platform

Introduction: Throughout follicular growth and subsequent corpus luteum formation the leukocyte number increases and follicular vascularisation changes. These processes are enhanced under exogenous stimulation with gonadotropins. Cytokines released by leukocytes contribute to further recruitment and vascularisation of the follicle, and they play an important role in regulating ovarian steroidogenesis by influencing theca and granulosa–lutein cell function. Changes in cytokine and vascular endothelial growth factor (VEGF) concentrations in the ovary as a consequence of gonadotropin stimulation may negatively influence oocyte quality. In this project we have compared the intrafollicular production of inflammatory cytokines and growth factors between natural IVF cycles (NC) and classical, gonadotropin-stimulated IVF cycles (gsIVF). Material and Methods: Serum on the day of oocyte retrieval and follicular fluid (FF) were collected in 37 NC and 39 gsIVF cycles. Thirteen women within this population underwent one NC and one gsIVF cycle each. A total of 14 cytokines from Bio-Plex panels I and II were determined in matched serum and FF samples using Luminex xMAP technology on the Bio-Plex(R) platform, using the serum protocol. Results: Tumour necrosis factor-alpha, RANTES, eotaxin and interferon-gamma-induced protein-10 levels were lower in FF than in serum, and thus not further investigated. Interleukin (IL)-6, -8, -10, -15, -18, monocyte chemotactic protein-1 (MCP-1), VEGF and leukaemia inhibitory factor (LIF) showed higher median concentrations in FF than in serum, indicating possible ovarian production. Moreover, most of these showed higher evels in the gsIVF than in the NC groups in the serum, but not in the follicular fluid. IL-8 was reduced in gsIVF cycles. Conclusion: The fact that serum but not FF levels of the studied cytokines were higher in the stimulated than in the natural cycles can be attributed to the increased number of active follicles present after controlled ovarian stimulation.