Giovanni De Gasperis

Automated electrorotation: dielectric characterization of living cells by real-time motion estimation

Electrorotation (ROT) has been applied widely for determining the dielectric properties of cells (and bio-particles) with single-cell resolution. However a serious limitation of ROT has been the tedious manual measurements required. A new real-time PC-based machine vision algorithm and hardware implementation are presented that achieve measurements of cell rotational motion and analysis of ROT spectra. The system is equipped with a computer-controlled quadrature digital synthesizer and is capable of measuring a ROT spectrum of a single cell with the frequency range 1 kHz-200 MHz in less than 5 min, taking four measurement points per frequency decade. Laser tweezers are used to facilitate cell selection and positioning in order to maximize the flexibility and accuracy of the system. The performance of this system is characterized in terms of robustness, accuracy and linearity with respect to manual measurements of real spinning cells under the influence of a rotating electric field. The system is quite generally applicable to a wide variety of mammalian cell morphologies and optical appearances. Membrane capacitance values derived from automated ROT measurements averaged within 10% of those obtained from manual measurements.

Microfluidic Cell Separation by 2-dimensional Dielectrophoresis

We describe a microfluidic device for separating cells according to their dielectric properties by combining 2-dimensional dielectrophoretic forces with field-flow-fractionation. The device comprises a thin chamber in which a travelling-wave electrical field is generated by a planar, multilayer microelectrode array at the bottom. Under the balance of gravitational and dielectrophoretic levitation forces, cells introduced into the device are positioned at different equilibrium heights in a velocity profile established inside the chamber, and thereby transported at different velocities by the fluid. Simultaneously, cells are subjected to a horizontal travelling-wave dielectrophoretic force that deflects them across the flow stream. The 2-dimensional dielectrophoretic forces acting on cells and the associated velocities in the fluid-flow and travelling-field directions depend sensitively on cell dielectric properties. The responses of cultured MDA-435 human breast cancer, HL-60 human leukemia and DS19 murine erythroleukemia cells, and of peripheral blood mononuclear (PBMN) cells were studied as functions of the frequency and voltage of the applied electric signals, and of the fluid flow rate. Significant differences were observed between the responses of different cell types. Cell separation was demonstrated by the differential redistribution of MDA-435 and PBMN cells as they flowed through the device. The device can be readily integrated with other microfluidic components for microscale sample preparation and analysis.

Complex reactivity with preferences in rule-based agents

In this paper, we extend our previous work on complex reaction in rule-based logical agents. In particular, we introduce the possibility of defining and exploiting complex preferences for choosing the course of action to undertake in response to external events, also based upon a (simplified) form of modal reasoning and on sequences of past events.

FRIENDLY & KIND with your Health: Human-Friendly Knowledge-INtensive Dynamic Systems for the e-Health Domain

This paper presents our approach for addressing “Human-friendly Knowledge-INtensive Dynamic Systems” (FRIENDLY & KIND systems) from a methodological point of view, also providing tools and languages for their design, implementation and testing. FRIENDLY & KIND systems are an evolution of multiagent systems and represent a good option for engineering complex and dynamic applications like those in the e-Health domain. We will demonstrate the suitability of our approach by designing and implementing a Remote Monitoring System for oncological patients.

A Heuristic Approach to Proposal-Based Negotiation: with Applications in Fashion Supply Chain Management

In this paper, we extend and improve the formal, executable framework for automated multi-issue negotiation between two autonomous competitive software agents proposed by Cadoli. This model is based on the view of negotiation spaces (or “areas”), representing the admissible values of the goods involved in the process as convex regions. However, in order to speed up the negotiation process and guarantee convergence, there was the restriction of potential agreements to vertices included in the intersection of the two areas. We present and assess experimentally an extension to Cadolis approach where, for both participating agents, interaction is no longer vertex based, or at least not necessarily so. This eliminates the asymmetry among parties and the limitation to polyhedral negotiation areas. The extension can be usefully integrated to Cadolis framework, thus obtaining an enhanced algorithm that can be effective in many practical cases. We present and discuss a number of experiments, aimed at assessing how parameters influence the performance of the algorithm and how they relate to each other. We discuss the usefulness of the approach in relevant application fields, such as, for instance, supply chain management in the fashion industry, which is a field of growing importance in economy and e-commerce.

A Multi-Agent System for Industrial Fault Detection and Repair

A Multi Agent System is described, capable of monitoring a telecommunication industrial test & measurement setup, designed as an application of the DALI agent language. The autonomy of the MAS is necessary to supervise the measurement apparatus during off-work time without human intervention, increasing the quality and efficacy of the overall test procedure. The MAS can decide whether to recover or repair the set of software process needed to achieve a correct test sequence without user intervention.

Building an AIML Chatter Bot Knowledge-Base Starting from a FAQ and a Glossary

Chatter bots are software programs that emulate human conversation and can show human-like conversational behavior in limited knowledge domain if properly crafted. AIML, Artifcial Intelligence Markup Language, is a well known XML derived language to build chatter bot knowledge bases, in a context of case-based reasoning and textual pattern matching algorithms. A design methodology will be explained based on a novel algorithm to automatically generate AIML knowledge bases starting from a frequently asked question free text fle and a glossary of terms.A generated demonstrator chatter bot using the Italian language will be shown.

DALI for Cognitive Robotics: Principles and Prototype Implementation

DALI is a logic Prolog-based Multi Agent System Language and Framework (publicly available on GitHub) developed at University of L’Aquila since 1999, and includes features aimed at user monitoring and training in Ambient Intelligent applications. In this paper, we show how such features can be integrated and extended in view of cognitive robotic applications; we then illustrate the extensions to the DALI implementation that allow DALI agents to interact with robotic platforms even through the cloud.

Memory, Experience and Adaptation in Logical Agents

The DALI language and framework [1] has been exploited in a number of practical applications (cf. [2] for a full list of references on DALI). In order to improve the DALI language and environment, we have been developing a comprehensive approach to memory management.

Demonstrator of a Multi-Agent System for Industrial Fault Detection and Repair

A demonstrator of a Multi Agent System is described capable of monitoring a telecommunication industrial test & measurement setup, designed as an application of the DALI agent language. The autonomy of the MAS is necessary to supervise the measurement apparatus during off-work time without human intervention, increasing the quality and efficacy of the overall test procedure. The MAS can decide whether to recover or repair the set of software process needed to achieve a correct test sequence without user intervention.