RESEARCH RESPONSIBLE: Prof. Andrea Giachetti (e-mail: andrea.giachetti@univr.it)

RESEARCH TITLE: Shape analysis on 3D surfaces acquired with 3D scanners: algorithms and applications

RESEARCH ABSTRACT: The program starts from the results obtained in the previous year and from other results recently obtained by the research group (new descriptors for 3D shape matching). Starting from the characterization of shapes with graphs based on multiscale salient points analysis, tests on different benchmarks will be performed and optimized graph kernels will be tested. The approach will be extended also to local description and matching, using context graphs characterizing salient points through geodesic connection.

The research in this field is both related to state of the art computer graphics application (invariant shape descriptors) and hot Pattern Recognition topics (graph matching). The relevant experience available in the research group in both the fields will be particularly useful for the candidate. Resulting algorithms could be applied also in practical contexts related to medicine and anthropometry involving the group.

The expected workplan is the following:

Months 1-6 Implementation of graph kernel methods for point description and global description (Radial Symmetry Graphs) with expected outcome consisting in 1-2 conference paper submission. In parallel the work on Tubular Section Tracking will be completed (submission of journal paper)

Months 6-12 Application of the developed methods to real world contexts (e.g. Computer aided diagnosis based on shape and texture, development of body composition predictors based on shape features)

RESEARCH LENGTH: 1 year

RESEARCH RESPONSIBLE: Prof. Manuele Bicego (e-mail: manuele.bicego@univr.it)

RESEARCH TITLE: Count models and representations for bioinformatics problem

RESEARCH ABSTRACT: In recent years, several Pattern Recognition and Computer Vision problems have been successfully faced by techniques based on the count representation - a representation which characterizes a pattern with a vector of counts. However, in the bioinformatics field, the potentialities of these tools and representation have not completely been exploited, due to the methodological and applicative challenges derived from the peculiar context.

The main goal of this research project is to study the problems related to the creation of count models and representations for some specific bioinformatics problems, such as the structural bioinformatics (characterization of protein structures starting from sequences and 3D data) and the analysis of proteomics data of self-immune diseases.

RESEARCH LENGTH: 1 year

RESEARCH RESPONSIBLE: Prof. Franco Fummi (e-mail: franco.fummi@univr.it)

RESEARCH TITLE: Heterogeneous Components Integration for the Design of Smart Systems

RESEARCH ABSTRACT: This research is in the field of effectively integrating heterogeneous components for the design of smart systems. A methodology and a set of tools must be developed to allow smart systems designers to safely and correctly aggregate heterogeneous components, such as, analog sensors, MEMS, power sources, digital hardware, embedded software, etc.

Competences currently available at the department, and in particular in the Electronic Systems Design (ESD) research group, are mainly related to the formalization and verification of components behaviors, but they lack of the comprehension of the actual physical behaviors. However, the definition of abstract and high-level models of heterogeneous components is the only strategy to create effective executable models of large and complex systems and this concept is repeated many times in the cited Horizon 2020 roadmap.

It is thus necessary to identify a Post Doc with competences in electronic, modeling of analog behaviors and elaboration of power to cover the gap between low-level behaviors and their abstract representation.

RESEARCH LENGTH: 2 years

RESEARCH RESPONSIBLE: Prof. Paolo Fiorini (e-mail: paolo.fiorini@univr.it)

RESEARCH TITLE: Integration of artificial reasoning and learning methods with robot control for applications to robot-assisted surgery

RESEARCH ABSTRACT: The candidate will analize the leading methods of artificial reasoning and learning applied to robotics and will integraate the most promising approaches into the control of a robot used for minimally invasive surgery, with particular attention to robotic surgery.

RESEARCH LENGTH: 1 year

RESEARCH RESPONSIBLE: Prof. Graziano Pravadelli (e-mail: graziano.pravadelli@univr.it)

RESEARCH TITLE: Definition of a platform to save energy in Data Centers

RESEARCH ABSTRACT: Nowadays, virtualization systems and cloud computing can play a key role in Data Centesr, as they can consolidate legacy systems, rapidly create new computer clusters and provide virtualized resources over Internet. Data Centers’ energy consumption has attracted global attention because of the fast growth of the information technology (IT) industry. Energy consumption is a critical issue for IT organizations today. Industry analysts estimate that over the next 5 years up to 60% of the energy consumed in a Data Center will be used for cooling in wasteful ways as a result of lack of environmental information and automatic intelligent redistribution of virtual systems to optimize resources consumption. The goal of this project is to design a real-time Data Center infrastructure management solution capable to: (i)gather data environment from a wireless sensor networks (WSN) (ii)recover solar energy by using advanced roof panel (iii)control actuators (iv)collect data from virtualization/cloud infrastructure (v)communicate with Uninterruptible Power Supply (UPS). The final aim is to define a set of automatic policies to reduce costs /saving energy, keep your Data Center running ensuring business continuity and reducing human interaction, particularly in case of critical situations.

RESEARCH LENGTH: 2 years

RESEARCH RESPONSIBLE: Prof. Carlo Combi (e-mail: carlo.combi@univr.it)

RESEARCH TITLE: Burundi Project at the Department of Computer Science

RESEARCH ABSTRACT: The project provides the design, implementation, and management of a system, based on free and open-source software, allowing for tele-consultations acting between the hospital of Ngozi (Burundi) and the hospitals in Verona (Italy). A consultation, or second opinion, is a deliberation between physicians on a case or its treatment. In particular, the designed system should allow to physicians in Ngozi to share several kinds of medical reports with and to request special consultations to physicians in Verona.

RESEARCH LENGTH: 1 year

RESEARCH RESPONSIBLE: Prof. Tiziano Villa (e-mail: tiziano.villa@univr.it)

RESEARCH TITLE: Methodologies and algorithms for the analysis of cyberphysical systems

RESEARCH ABSTRACT: The objective is to extend the capability of verifying and synthesizing systems modeled by hybrid automata through the further development of the library ARIADNE, of which a prototype for the analysis of hybrid systems has been

already impkemented. ARIADNE can handle non-linear dynamics with a rigorous mathematical semantics.

RESEARCH LENGTH:2 years

RESEARCH RESPONSIBLE: Prof. Pasquina Marzola (e-mail: pasquina.marzola@univr.it)

RESEARCH TITLE: Physico-chemical characterization of magnetosomes extracted from the magnetotactic bacterium magnetospirillum gryphiswaldense MSR -1-

RESEARCH ABSTRACT: Investigation of physico-chemical characteristics of magnetosomes extracted from magnetotactic bacteria using: x-ray diffraction, atomic force microscopy, electron microscopy, dynamic light-scattering and Z potential.

RESEARCH LENGTH: 1 year

RESEARCH RESPONSIBLE: Prof. Alessandro Romeo (e-mail: alessandro.romeo@univr.it)

RESEARCH TITLE: Preparation and characterization of nano structured materials for biomedicals and energy applications

RESEARCH ABSTRACT: The research regards the preparation and characterization of second generation thin film solar cells, such as CdTe/CdS devices and at the same time also to analyse nanostructured materials (typically nanoparticles) for biomedical applications. From the photovoltaic side the layers of the device will be prepared by means of vacuum and non-vacuum techniques. The study will concentrate on the reduction of absorber thickness in order to reduce the material impact to cost and to environment. The materials will be analysed by means of atomic force microscopy (AFM) and x-ray diffraction spectra. Electrical characterization of the devices will be made by capacitance-voltage, drive level capacitance profiling and current-voltage techniques. Moreover the AFM technology will be applied on the bio-materials, nanoparticles, in order to study their size, morphology and more important their behaviour to different solution concentration and properties. Also physical properties like luminescence and magnetic properties will be taken into consideration. This second part will be made in collaboration with groups that will provide the nanoparticles from one side and groups that will test them from the medical effect.

RESEARCH LENGTH: 1 year

RESEARCH RESPONSIBLE: Prof. Nicola Daldosso (e-mail: nicola.daldosso@univr.it)

RESEARCH TITLE: Spectroscopic characterization and validation of new biomarkers for alcoholic abuse

RESEARCH ABSTRACT: The present research program aims at identifying and validating new biomarkers for alcoholic abuse, thus helping the prevention of work and road accidents thanks to the definition and assessment of new analytical parameters.

The main goal of the research is the characterization and validation of a group of CDT glicoforms – asialoCDT is one of the more specific markers for the alcoholic abuse, but it is not easily detected by the current analytical methods – through the use of high selective and sensitive fluorescence methods and procedures. The present AdR will be fundamental for the chemical and physical characterization of the rare earth inclusion in the asialo-CDT (to enhance sensibility). Moreover, new fluorescence equipment based on the validated techniques will be developed.

RESEARCH LENGTH: 2 years

RESEARCH RESPONSIBLE: Prof. Marco Squassina (e-mail: marco.squassina@univr.it)

RESEARCH TITLE: Existence and qualitative properties of solutions of nonlocal problems

RESEARCH ABSTRACT: Existence and multiplicity results for the fractional Laplacian with exponential nonlinearities, spectral theory for the fractional p-Laplacian, degree theory with application to quasilinear elliptic equations involving non-potential operators.

RESEARCH LENGTH: 1 year

RESEARCH RESPONSIBLE: Prof. Luca Di Persio (e-mail: luca.dipersio@univr.it)

RESEARCH TITLE: Stochastic differential equations with jumps in Mathematical Finance: applications to pricing, hedging and dynamic risk measures's problems

RESEARCH ABSTRACT: Empirical evidence shows that assuming the Geometric Brownian motion for stock prices' behaviors, as in the classical Black-Scholes (BS) approach, we miss an accurate description of financial data. Discrepancies between the BS prices and real data depend, e.g., on continuity for Brownian motion trajectories, implied volatility in option market, kurtosis and skewness of asset returns. Improvements has been achieved taking into account jump perturbations acting on asset prices motion, i.e. using Lévy processes which allow stock prices to jump while preserving the independence and stationarity of returns. The latter leads to consider general Itô-Lévy processes jointly with the theory of Backward Stochastic Differential Equations (BSDEs) with jumps to treat pricing and hedging problems also in connection with questions of stochastic optimal transportation and dynamic risk measures definition. The latter are the key points of our project which aims to a better understanding of features characterizing associated (incomplete) financial markets.

RESEARCH LENGTH: 1 year