Papers by Salvatore Gaglio
IEEE Access
This work was partially funded by the European Union-NextGenerationEU-MUR D.M. funds 737/2021-res... more This work was partially funded by the European Union-NextGenerationEU-MUR D.M. funds 737/2021-research project ''A Trustworthy Clinical Decision Support System for non-communicable diseases detection and prediction.'' ABSTRACT The SARS-CoV-2 virus pandemic had devastating effects on various aspects of life: clinical cases, ranging from mild to severe, can lead to lung failure and to death. Due to the high incidence, datadriven models can support physicians in patient management. The explainability and interpretability of machine-learning models are mandatory in clinical scenarios. In this work, clinical, laboratory and radiomic features were used to train machine-learning models for COVID-19 prognosis prediction. Using Explainable AI algorithms, a multi-level explainable method was proposed taking into account the developer and the involved stakeholder (physician, and patient) perspectives. A total of 1023 radiomic features were extracted from 1589 Chest X-Ray images (CXR), combined with 38 clinical/laboratory features. After the preprocessing and selection phases, 40 CXR radiomic features and 23 clinical/laboratory features were used to train Support Vector Machine and Random Forest classifiers exploring three feature selection strategies. The combination of both radiomic, and clinical/laboratory features enabled higher performance in the resulting models. The intelligibility of the used features allowed us to validate the models' clinical findings. According to the medical literature, LDH, PaO2 and CRP were the most predictive laboratory features. Instead, ZoneEntropy and HighGrayLevelZoneEmphasis-indicative of the heterogeneity/uniformity of lung texture-were the most discriminating radiomic features. Our best predictive model, exploiting the Random Forest classifier and a signature composed of clinical, laboratory and radiomic features, achieved AUC=0.819, accuracy=0.733, specificity=0.705, and sensitivity=0.761 in the test set. The model, including a multi-level explainability, allows us to make strong clinical assumptions, confirmed by the literature insights. INDEX TERMS Chest X-ray images, clinical and laboratory features, COVID-19 prognosis, explainable AI, machine learning classifiers, predictive models, radiomic features.
Cognitive Computation
This work aims to implement an automated data-driven model for breast cancer detection in mammogr... more This work aims to implement an automated data-driven model for breast cancer detection in mammograms to support physicians’ decision process within a breast cancer screening or detection program. The public available CBIS-DDSM and the INbreast datasets were used as sources to implement the transfer learning technique on full-field digital mammography proprietary dataset. The proprietary dataset reflects a real heterogeneous case study, consisting of 190 masses, 46 asymmetries, and 71 distortions. Several Yolo architectures were compared, including YoloV3, YoloV5, and YoloV5-Transformer. In addition, Eigen-CAM was implemented for model introspection and outputs explanation by highlighting all the suspicious regions of interest within the mammogram. The small YoloV5 model resulted in the best developed solution obtaining an mAP of 0.621 on proprietary dataset. The saliency maps computed via Eigen-CAM have proven capable solution reporting all regions of interest also on incorrect pred...
IEEE Access
Wireless Sensor Networks (WSNs) represent a key component in emerging distributed computing parad... more Wireless Sensor Networks (WSNs) represent a key component in emerging distributed computing paradigms such as IoT, Ambient Intelligence, and Smart Cities. In these contexts, the difficulty of testing, verifying, and monitoring applications in their intended scenarios ranges from challenging to impractical. Current simulators can only be used to investigate correctness at source code level and with limited accuracy. This paper proposes a system and a methodology to model and verify symbolic distributed applications running on WSNs. The approach allows to complement the distributed application code at a high level of abstraction in order to test and reprogram it, directly, on deployed network devices. The proposed intelligent architecture enables the execution of distributed applications and the verification of the supplied correctness conditions. This paper shows the feasibility of the proposed approach and its effectiveness even when networks include resource-constrained nodes with some sample applications and quantitative experiments measuring the overhead introduced by the monitoring operations.
Sensors
Unmanned Aerial Vehicles (UAVs) are often studied as tools to perform data collection from Wirele... more Unmanned Aerial Vehicles (UAVs) are often studied as tools to perform data collection from Wireless Sensor Networks (WSNs). Path planning is a fundamental aspect of this endeavor. Works in the current literature assume that data are always ready to be retrieved when the UAV passes. This operational model is quite rigid and does not allow for the integration of the UAV as a computational object playing an active role in the network. In fact, the UAV could begin the computation on a first visit and retrieve the data later. Potentially, the UAV could orchestrate the distributed computation to improve its performance, change its parameters, and even upload new applications to the sensor network. In this paper, we analyze a scenario where a UAV plays an active role in the operation of multiple sensor networks by visiting different node clusters to initiate distributed computation and collect the final outcomes. The experimental results validate the effectiveness of the proposed method in...
Sensors
We propose a methodology to verify applications developed following programming patterns inspired... more We propose a methodology to verify applications developed following programming patterns inspired by natural language that interact with physical environments and run on resource-constrained interconnected devices. Natural language patterns allow for the reduction of intermediate abstraction layers to map physical domain concepts into executable code avoiding the recourse to ontologies, which would need to be shared, kept up to date, and synchronized across a set of devices. Moreover, the computational paradigm we use for effective distributed execution of symbolic code on resource-constrained devices encourages the adoption of such patterns. The methodology is supported by a rule-based system that permits runtime verification of Software Under Test (SUT) on board the target devices through automated oracle and test case generation. Moreover, verification extends from syntactic and semantic checks to the evaluation of the effects of SUT execution on target hardware. Additionally, by...
It is introduced a semantic coding of words based the SVD tech nique for generating a semantic sp... more It is introduced a semantic coding of words based the SVD tech nique for generating a semantic space in which words are then mappe d. Th proposed technique is based on the introduction of a link energyformula related to digrams frequency. To test the effectiveness of the porposed approa ch, the english translation of the set of Grimm Fairy Tales has been analyzed and a b i-dimensional visual representation has been obtained using the Sammon pr ojection algorithm.
One of the most important functions of concepts is that of producing classifications; and since t... more One of the most important functions of concepts is that of producing classifications; and since there are at least two different types of such things, we better give a preliminary short description of them both. The first kind of classification is based on the existence of a property common to all the things that fall under a concept. The second, instead, relies on similarities between the objects belonging to a certain class A and certain elements of a subclass AS of A, the so-called ‘stereotypes.’ In what follows, we are going to call ‘proto-concepts’ all those concepts whose power of classification depends on stereotypes, leaving the term ‘concepts’ for all the others. The main aim of this article is showing that, if a proto-concept is given simply in terms of the ability to make the appropriate distinctions, then there are stimulus-response cognitive systems — whose way of manipulating information is based on Neural Networks (NN) — able to make the appropriate distinctions typic...
WIT Transactions on Information and Communication Technologies, 1970
A cognitive architecture for an artificial vision system, aimed for an autonomous intelligent sys... more A cognitive architecture for an artificial vision system, aimed for an autonomous intelligent system, is presented with particular attention to performance aspects. The architecture, based on the active vision paradigm, is cognitive in the sense that several cognitive hypotheses have been postulated as guidelines for its design in order to achieve good performance.
Robotics, 2020
This paper describes an interactive storytelling system, accessible through the SoftBank robotic ... more This paper describes an interactive storytelling system, accessible through the SoftBank robotic platforms NAO and Pepper. The main contribution consists of the interpretation of the story characters by humanoid robots, obtained through the definition of appropriate cognitive models, relying on the ACT-R cognitive architecture. The reasoning processes leading to the story evolution are based on the represented knowledge and the suggestions of the listener in critical points of the story. They are disclosed during the narration, to make clear the dynamics of the story and the feelings of the characters. We analyzed the impact of such externalization of the internal status of the characters to set the basis for future experimentation with primary school children.
2021 IEEE International Conference on Smart Computing (SMARTCOMP), 2021
Simulations are indispensable to reduce costs and risks when developing and testing algorithms fo... more Simulations are indispensable to reduce costs and risks when developing and testing algorithms for unmanned aerial vehicles (UAV) especially for applications in high risk scenarios like search and rescue (SAR) operations and post-disaster damage assessment. Many UAV applications require real-time tasks for which the timeliness of computations is fundamental. However, standard simulation tools are not guaranteed to run in sync with real-time events, leading to unreliable assessments of the ability of the target hardware to perform specific tasks. In this work we present a simulation and test system able to run UAV tasks on resource-constrained target hardware possibly adopted in these applications. The system allows for hardware-in-the-loop simulations in which a virtual UAV provided with virtual sensors is controlled by the software under test (SUT) running on the target hardware, while simulated and real time are kept in sync. We provide experimental results from the execution of several increasingly difficult tasks in the system.
2017 AEIT International Annual Conference, 2017
Nowadays, the population's average age is constantly increasing, and thus the need for specialize... more Nowadays, the population's average age is constantly increasing, and thus the need for specialized home assistance is on the rise. Smart homes especially tailored to meet elderly and disabled people's needs can help them maintaining their autonomy, whilst ensuring their safety and well-being. This paper proposes a complete context-aware system for Ambient Assisted Living (AAL), which infers user's actions and context, analyzing its past and current behavior to detect anomalies and prevent possible emergencies. The proposed system exploits Dynamic Bayesian Networks to merge raw data coming from heterogeneous sensors and infer user's behavior and health conditions. A rule-based reasoner is able to detect and predict anomalies in such data, sending appropriate alerts to caregivers and family members. The effectiveness of the proposed AAL system is demonstrated by extensive experimental results carried out in a simulated smart home.
2019 IEEE International Conference on Smart Computing (SMARTCOMP), 2019
Being part of one of the fastest growing area in Artificial Intelligence (AI), virtual assistants... more Being part of one of the fastest growing area in Artificial Intelligence (AI), virtual assistants are nowadays part of everyone's life being integrated in almost every smart device. Alexa, Siri, Google Assistant, and Cortana are just few examples of the most famous ones. Beyond these off-the-shelf solutions, different technologies which allow to create custom assistants are available. IBM Watson, for instance, is one of the most widely-adopted question-answering framework both because of its simplicity and accessibility through public APIs. In this work, we present a virtual assistant that exploits the Watson technology to support students and staff of a smart campus at the University of Palermo. Some in progress results show the effectiveness of the approach we propose.
AI*IA 2017 Advances in Artificial Intelligence, 2017
In recent years, the percentage of the population owning a smartphone has increased significantly... more In recent years, the percentage of the population owning a smartphone has increased significantly. These devices provide the user with more and more functions, so that anyone is encouraged to carry one during the day, implicitly producing that can be analysed to infer knowledge of the user's context. In this work we present a novel framework for Human Activity Recognition (HAR) using smartphone data captured by means of embedded triaxial accelerometer and gyroscope sensors. Some statistics over the captured sensor data are computed to model each activity, then real-time classification is performed by means of an efficient supervised learning technique. The system we propose also adopts a participatory sensing paradigm where user's feedbacks on recognised activities are exploited to update the inner models of the system. Experimental results show the effectiveness of our solution as compared to other state-of-the-art techniques.
Journal of Visual Language and Computing, 2020
The work describes a module that has been implemented for being included in a social humanoid rob... more The work describes a module that has been implemented for being included in a social humanoid robot architecture, in particular a storyteller robot, named NarRob. This module gives a humanoid robot the capability of mimicking and acquiring the motion of a human user in real-time. This allows the robot to increase the population of his dataset of gestures. The module relies on a Kinect based acquisition setup. The gestures are acquired by observing the typical gesture displayed by humans. The movements are then annotated by several evaluators according to their particular meaning, and they are organized considering a specific typology in the knowledge base of the robot. The properly annotated gestures are then used to enrich the narration of the stories. During the narration, the robot semantically analyses the textual content of the story in order to detect meaningful terms in the sentences and emotions that can be expressed. This analysis drives the choice of the gesture that accompanies the sentences when the story is read.
Proceedings of the 17th International Conference on Computer Systems and Technologies 2016, 2016
Several distributed applications running over the Internet use Reputation Management Systems (RMS... more Several distributed applications running over the Internet use Reputation Management Systems (RMSs) to guarantee reliable interactions among unknown agents. Because of the heterogeneity of the existing RMSs, their assessment in terms of correctness and resistance to security attacks is not a trivial task. This work addresses this issue by presenting a novel parallel simulator aimed to support researchers in evaluating the performances of a RMS since the design phase. Preliminary results obtained by simulating two different attacks confirm the suitability of the proposed framework to evaluate different RMSs.
Methods for Adaptable Usability
In this chapter the role of multimodality in intelligent, mobile guides for cultural heritage env... more In this chapter the role of multimodality in intelligent, mobile guides for cultural heritage environments is discussed. Multimodal access to information contents enables the creation of systems with a higher degree of accessibility and usability. A multimodal interaction may involve several human interaction modes, such as sight, touch and voice to navigate contents, or gestures to activate controls. We first start our discussion by presenting a timeline of cultural heritage system evolution, spanning from 2001 to 2008, which highlights design issues such as intelligence and context-awareness in providing information. Then, multimodal access to contents is discussed, along with problems and corresponding solutions; an evaluation of several reviewed systems is also presented. Lastly, a case study multimodal framework termed MAGA is described, which combines intelligent conversational agents with speech recognition/ synthesis technology in a framework employing RFID based location and Wi-Fi based data exchange.
The main aim of this paper is contributing to what in the last few years has been known as comput... more The main aim of this paper is contributing to what in the last few years has been known as computational creativity. This will be done by showing the relevance of a particular mathematical representation of Gärdenfors's conceptual spaces to the problem of modelling a phenomenon which plays a central role in producing novel and fruitful representations of perceptual patterns: analogy.
Mathematical patterns are an important subclass of the class of patterns. The main task of this p... more Mathematical patterns are an important subclass of the class of patterns. The main task of this paper is examining a particular proposal concerning the nature of mathematical patterns and some elements of the cognitive structure an agent should have to recognize them.
Euro-Par 2005 Parallel Processing, 2005
Uploads
Papers by Salvatore Gaglio