Full correction for spatially distributed speed-of-sound in echo ultrasound based on measuring aberration delays via transmit beam steering

Aberrations of the acoustic wave front, caused by spatial variations of the speed-of-sound, are a main limiting factor to the diagnostic power of medical ultrasound imaging. If not accounted for, aberrations result in low resolution and increased side lobe level, over all reducing contrast in deep tissue imaging. Various techniques have been proposed for quantifying aberrations by analysing the arrival time of coherent echoes from so-called guide stars or beacons. In situations where a guide star is missing, aperture-based techniques may give ambiguous results. Moreover, they are conceptually focused on aberrators that can be approximated as a phase screen in front of the probe. We propose a novel technique, where the effect of aberration is detected in the reconstructed image as opposed to the aperture data. The varying local echo phase when changing the transmit beam steering angle directly reflects the varying arrival time of the transmit wave front. This allows sensing the angle-dependent aberration delay in a spatially resolved way, and thus aberration correction for a spatially distributed volume aberrator. In phantoms containing a cylindrical aberrator, we achieved location-independent diffraction-limited resolution as well as accurate display of echo location based on reconstructing the speed-of-sound spatially resolved. First successful volunteer results confirm the clinical potential of the proposed technique.

Reusable Knowledge-based Components for Building Software Applications: A Knowledge Modelling Approach

In computer science, different types of reusable components for building software applications were proposed as a direct consequence of the emergence of new software programming paradigms. The success of these components for building applications depends on factors such as the flexibility in their combination or the facility for their selection in centralised or distributed environments such as internet. In this article, we propose a general type of reusable component, called primitive of representation, inspired by a knowledge-based approach that can promote reusability. The proposal can be understood as a generalisation of existing partial solutions that is applicable to both software and knowledge engineering for the development of hybrid applications that integrate conventional and knowledge based techniques. The article presents the structure and use of the component and describes our recent experience in the development of real-world applications based on this approach.

Increasing the Intelligence of Virtual Sales Assistants through Knowledge Modeling Techniques

Shopping agents are web-based applications that help consumers to find appropriate products in the context of e-commerce. In this paper we argue about the utility of advanced model-based techniques that recently have been proposed in the fields of Artificial Intelligence and Knowledge Engineering, in order to increase the level of support provided by this type of applications. We illustrate this approach with a virtual sales assistant that dynamically configures a product according to the needs and preferences of customers.

Modeling Commercial Knowledge to Develop Advanced Agent-based Marketplaces for E-commerce

This paper argues about the utility of advanced knowledge-based techniques to develop web-based applications that help consumers in finding products within marketplaces in e-commerce. In particular, we describe the idea of model-based approach to develop a shopping agent that dynamically configures a product according to the needs and preferences of customers. Finally, the paper summarizes the advantages provided by this approach.

Neural network-based wind vector retrieval from satellite scatterometer data

Obtaining wind vectors over the ocean is important for weather forecasting and ocean modelling. Several satellite systems used operationally by meteorological agencies utilise scatterometers to infer wind vectors over the oceans. In this paper we present the results of using novel neural network based techniques to estimate wind vectors from such data. The problem is partitioned into estimating wind speed and wind direction. Wind speed is modelled using a multi-layer perceptron (MLP) and a sum of squares error function. Wind direction is a periodic variable and a multi-valued function for a given set of inputs; a conventional MLP fails at this task, and so we model the full periodic probability density of direction conditioned on the satellite derived inputs using a Mixture Density Network (MDN) with periodic kernel functions. A committee of the resulting MDNs is shown to improve the results.

Neural network-based wind vector retrieval from satellite scatterometer data

Obtaining wind vectors over the ocean is important for weather forecasting and ocean modelling. Several satellite systems used operationally by meteorological agencies utilise scatterometers to infer wind vectors over the oceans. In this paper we present the results of using novel neural network based techniques to estimate wind vectors from such data. The problem is partitioned into estimating wind speed and wind direction. Wind speed is modelled using a multi-layer perceptron (MLP) and a sum of squares error function. Wind direction is a periodic variable and a multi-valued function for a given set of inputs; a conventional MLP fails at this task, and so we model the full periodic probability density of direction conditioned on the satellite derived inputs using a Mixture Density Network (MDN) with periodic kernel functions. A committee of the resulting MDNs is shown to improve the results.

Knowledge-based Virtual Reconstruction of Museum Artifacts

Within the framework of heritage preservation, 3D scanning and modeling for heritage documentation has increased significantly in recent years, mainly due to the evolution of laser and image-based techniques, modeling software, powerful computers and virtual reality. 3D laser acquisition constitutes a real development opportunity for 3D modeling based previously on theoretical data. The representation of the object information rely on the knowledge of its historic and theoretical frame to reconstitute a posteriori its previous states. This project proposes an approach dealing with data extraction based on architectural knowledge and Laser statement informing measurements, the whole leading to 3D reconstruction. The experimented Khmer objects are exposed at Guimet museum in Paris. The purpose of this digital modeling meets the need of exploitable models for simulation projects, prototyping, exhibitions, promoting cultural tourism and particularly for archiving against any likely disaster and as an aided tool for the formulation of virtual museum concept.

On the microstructure, growth pattern and original porosity of belemnite rostra: insights from calcitic Jurassic belemnites

Calcitic belemnite rostra are usually employed to perform paleoenvironmental studies based on geochemical data. However, several questions, such as their original porosity and microstructure, remain open, despite they are essential to make accurate interpretations based on geochemical analyses.This paper revisits and enlightens some of these questions. Petrographic data demonstrate that calcite crystals of the rostrum solidum of belemnites grow from spherulites that successively develop along the apical line, resulting in a “regular spherulithic prismatic” microstructure. Radially arranged calcite crystals emerge and diverge from the spherulites: towards the apex, crystals grow until a new spherulite is formed; towards the external walls of the rostrum, the crystals become progressively bigger and prismatic. Adjacent crystals slightly vary in their c-axis orientation, resulting in undulose extinction. Concentric growth layering develops at different scales and is superimposed and traversed by a radial pattern, which results in the micro-fibrous texture that is observed in the calcite crystals in the rostra.Petrographic data demonstrate that single calcite crystals in the rostra have a composite nature, which strongly suggests that the belemnite rostra were originally porous. Single crystals consistently comprise two distinct zones or sectors in optical continuity: 1) the inner zone is fluorescent, has relatively low optical relief under transmitted light (TL) microscopy, a dark-grey color under backscatter electron microscopy (BSEM), a commonly triangular shape, a “patchy” appearance and relatively high Mg and Na contents; 2) the outer sector is non-fluorescent, has relatively high optical relief under TL, a light-grey color under BSEM and low Mg and Na contents. The inner and fluorescent sectors are interpreted to have formed first as a product of biologically controlled mineralization during belemnite skeletal growth and the non-fluorescent outer sectors as overgrowths of the former, filling the intra- and inter-crystalline porosity. This question has important implications for making paleoenvironmental and/or paleoclimatic interpretations based on geochemical analyses of belemnite rostra.Finally, the petrographic features of composite calcite crystals in the rostra also suggest the non-classical crystallization of belemnite rostra, as previously suggested by other authors.

A low overhead error confinement method based on application statistical characteristics

Reliability has emerged as a critical design constraint especially in memories. Designers are going to great lengths to guarantee fault free operation of the underlying silicon by adopting redundancy-based techniques, which essentially try to detect and correct every single error. However, such techniques come at a cost of large area, power and performance overheads which making many researchers to doubt their efficiency especially for error resilient systems where 100% accuracy is not always required. In this paper, we present an alternative method focusing on the confinement of the resulting output error induced by any reliability issues. By focusing on memory faults, rather than correcting every single error the proposed method exploits the statistical characteristics of any target application and replaces any erroneous data with the best available estimate of that data. To realize the proposed method a RISC processor is augmented with custom instructions and special-purpose functional units. We apply the method on the proposed enhanced processor by studying the statistical characteristics of the various algorithms involved in a popular multimedia application. Our experimental results show that in contrast to state-of-the-art fault tolerance approaches, we are able to reduce runtime and area overhead by 71.3% and 83.3% respectively.

A path-based framework for analyzing large markov models

Reliability and dependability modeling can be employed during many stages of analysis of a computing system to gain insights into its critical behaviors. To provide useful results, realistic models of systems are often necessarily large and complex. Numerical analysis of these models presents a formidable challenge because the sizes of their state-space descriptions grow exponentially in proportion to the sizes of the models. On the other hand, simulation of the models requires analysis of many trajectories in order to compute statistically correct solutions. This dissertation presents a novel framework for performing both numerical analysis and simulation. The new numerical approach computes bounds on the solutions of transient measures in large continuous-time Markov chains (CTMCs). It extends existing path-based and uniformization-based methods by identifying sets of paths that are equivalent with respect to a reward measure and related to one another via a simple structural relationship. This relationship makes it possible for the approach to explore multiple paths at the same time,· thus significantly increasing the number of paths that can be explored in a given amount of time. Furthermore, the use of a structured representation for the state space and the direct computation of the desired reward measure (without ever storing the solution vector) allow it to analyze very large models using a very small amount of storage. Often, path-based techniques must compute many paths to obtain tight bounds. In addition to presenting the basic path-based approach, we also present algorithms for computing more paths and tighter bounds quickly. One resulting approach is based on the concept of path composition whereby precomputed subpaths are composed to compute the whole paths efficiently. Another approach is based on selecting important paths (among a set of many paths) for evaluation. Many path-based techniques suffer from having to evaluate many (unimportant) paths. Evaluating the important ones helps to compute tight bounds efficiently and quickly.

Components and circuits for tunneling diode based high frequency sources

Terahertz (THz) technology has been generating a lot of interest because of the potential applications for systems working in this frequency range. However, to fully achieve this potential, effective and efficient ways of generating controlled signals in the terahertz range are required. Devices that exhibit negative differential resistance (NDR) in a region of their current-voltage (I-V ) characteristics have been used in circuits for the generation of radio frequency signals. Of all of these NDR devices, resonant tunneling diode (RTD) oscillators, with their ability to oscillate in the THz range are considered as one of the most promising solid-state sources for terahertz signal generation at room temperature. There are however limitations and challenges with these devices, from inherent low output power usually in the range of micro-watts (uW) for RTD oscillators when milli-watts (mW) are desired. At device level, parasitic oscillations caused by the biasing line inductance when the device is biased in the NDR region prevent accurate device characterisation, which in turn prevents device modelling for computer simulations. This thesis describes work on I-V characterisation of tunnel diode (TD) and RTD (fabricated by Dr. Jue Wang) devices, and the radio frequency (RF) characterisation and small signal modelling of RTDs. The thesis also describes the design and measurement of hybrid TD oscillators for higher output power and the design and measurement of a planar Yagi antenna (fabricated by Khalid Alharbi) for THz applications. To enable oscillation free current-voltage characterisation of tunnel diodes, a commonly employed method is the use of a suitable resistor connected across the device to make the total differential resistance in the NDR region positive. However, this approach is not without problems as the value of the resistor has to satisfy certain conditions or else bias oscillations would still be present in the NDR region of the measured I-V characteristics. This method is difficult to use for RTDs which are fabricated on wafer due to the discrepancies in designed and actual resistance values of fabricated resistors using thin film technology. In this work, using pulsed DC rather than static DC measurements during device characterisation were shown to give accurate characteristics in the NDR region without the need for a stabilisation resistor. This approach allows for direct oscillation free characterisation for devices. Experimental results show that the I-V characterisation of tunnel diodes and RTD devices free of bias oscillations in the NDR region can be made. In this work, a new power-combining topology to address the limitations of low output power of TD and RTD oscillators is presented. The design employs the use of two oscillators biased separately, but with the combined output power from both collected at a single load. Compared to previous approaches, this method keeps the frequency of oscillation of the combined oscillators the same as for one of the oscillators. Experimental results with a hybrid circuit using two tunnel diode oscillators compared with a single oscillator design with similar values shows that the coupled oscillators produce double the output RF power of the single oscillator. This topology can be scaled for higher (up to terahertz) frequencies in the future by using RTD oscillators. Finally, a broadband Yagi antenna suitable for wireless communication at terahertz frequencies is presented in this thesis. The return loss of the antenna showed that the bandwidth is larger than the measured range (140-220 GHz). A new method was used to characterise the radiation pattern of the antenna in the E-plane. This was carried out on-wafer and the measured radiation pattern showed good agreement with the simulated pattern. In summary, this work makes important contributions to the accurate characterisation and modelling of TDs and RTDs, circuit-based techniques for power combining of high frequency TD or RTD oscillators, and to antennas suitable for on chip integration with high frequency oscillators.

Constraint Programming-based Job Dispatching for Modern HPC Applications

A High-Performance Computing job dispatcher is a critical software that assigns the finite computing resources to submitted jobs. This resource assignment over time is known as the on-line job dispatching problem in HPC systems. The fact the problem is on-line means that solutions must be computed in real-time, and their required time cannot exceed some threshold to do not affect the normal system functioning. In addition, a job dispatcher must deal with a lot of uncertainty: submission times, the number of requested resources, and duration of jobs. Heuristic-based techniques have been broadly used in HPC systems, at the cost of achieving (sub-)optimal solutions in a short time. However, the scheduling and resource allocation components are separated, thus generates a decoupled decision that may cause a performance loss. Optimization-based techniques are less used for this problem, although they can significantly improve the performance of HPC systems at the expense of higher computation time. Nowadays, HPC systems are being used for modern applications, such as big data analytics and predictive model building, that employ, in general, many short jobs. However, this information is unknown at dispatching time, and job dispatchers need to process large numbers of them quickly while ensuring high Quality-of-Service (QoS) levels. Constraint Programming (CP) has been shown to be an effective approach to tackle job dispatching problems. However, state-of-the-art CP-based job dispatchers are unable to satisfy the challenges of on-line dispatching, such as generate dispatching decisions in a brief period and integrate current and past information of the housing system. Given the previous reasons, we propose CP-based dispatchers that are more suitable for HPC systems running modern applications, generating on-line dispatching decisions in a proper time and are able to make effective use of job duration predictions to improve QoS levels, especially for workloads dominated by short jobs.

Self-supervised information retrieval: a novel approach based on Deep Metric Learning and Neural Language Models

Most of the existing open-source search engines, utilize keyword or tf-idf based techniques to find relevant documents and web pages relative to an input query. Although these methods, with the help of a page rank or knowledge graphs, proved to be effective in some cases, they often fail to retrieve relevant instances for more complicated queries that would require a semantic understanding to be exploited. In this Thesis, a self-supervised information retrieval system based on transformers is employed to build a semantic search engine over the library of Gruppo Maggioli company. Semantic search or search with meaning can refer to an understanding of the query, instead of simply finding words matches and, in general, it represents knowledge in a way suitable for retrieval. We chose to investigate a new self-supervised strategy to handle the training of unlabeled data based on the creation of pairs of ’artificial’ queries and the respective positive passages. We claim that by removing the reliance on labeled data, we may use the large volume of unlabeled material on the web without being limited to languages or domains where labeled data is abundant.

Advanced X-ray techniques for the structural characterisation of Prussian Blue Analogue cathode materials

The thesis is dedicated to the implementation of advanced x-ray-based techniques for the investigation of the battery systems, more predominantly, the cathode materials. The implemented characterisation methods include synchrotron based x-ray absorption spectroscopy, powder x-ray diffraction, 2-dimensional x-ray fluorescence, full field transmission soft x-ray microscopy, and laboratory x-ray photoelectron spectroscopy. The research highlights the different areas of expertise for each described method, in terms of material characterisation, exploring their complementarities and intersections. The results are focused over manganese hexacyanoferrate and partially Ni substituted manganese hexacyanoferrate, through both organic and aqueous battery systems. In aqueous system, the modification of cathode composition has been observed with various techniques, indicating to the processes occurring in bulk, surface, locally or in long-range, including with the speciation by 2-dimensional scanning, and the time-resolution, by the implementation of the operando measurements. In organic media, the inhomogenisation of the cathode material during the aging process was investigated by the development of the special image treatment procedure for the maps, obtained from the transmission soft x-ray microscopy. It worth mentioning, that apart from the combination of the outcomes from the various x-ray measurements, the exploration of the new capabilities was also conducted, namely, probing the oxidation state of the element with the synchrotron-based 2-dimensional x-ray fluorescence technique, which, generally, with conventional set up, is not possible to achieve. The results and methodology from this thesis can, of course, be generalised on the characterisation of the other battery systems, and not only, as the x-ray techniques are one of the most informative and sophisticated methods for advanced structural investigation of the materials.

Knowledge-based Chord Embeddings

This thesis develops AI methods as a contribution to computational musicology, an interdisciplinary field that studies music with computers. In systematic musicology a composition is defined as the combination of harmony, melody and rhythm. According to de La Borde, harmony alone "merits the name of composition". This thesis focuses on analysing the harmony from a computational perspective. We concentrate on symbolic music representation and address the problem of formally representing chord progressions in western music compositions. Informally, chords are sets of pitches played simultaneously, and chord progressions constitute the harmony of a composition. Our approach combines ML techniques with knowledge-based techniques. We design and implement the Modal Harmony ontology (MHO), using OWL. It formalises one of the most important theories in western music: the Modal Harmony Theory. We propose and experiment with different types of embedding methods to encode chords, inspired by NLP and adapted to the music domain, using both statistical (extensional) knowledge by relying on a huge dataset of chord annotations (ChoCo), intensional knowledge by relying on MHO and a combination of the two. The methods are evaluated on two musicologically relevant tasks: chord classification and music structure segmentation. The former is verified by comparing the results of the Odd One Out algorithm to the classification obtained with MHO. Good performances (accuracy: 0.86) are achieved. We feed a RNN for the latter, using our embeddings. Results show that the best performance (F1: 0.6) is achieved with embeddings that combine both approaches. Our method outpeforms the state of the art (F1 = 0.42) for symbolic music structure segmentation. It is worth noticing that embeddings based only on MHO almost equal the best performance (F1 = 0.58). We remark that those embeddings only require the ontology as an input as opposed to other approaches that rely on large datasets.