A5087549459- Privacy-Preserving Technologies in Data
- Uterine Myomas and Treatments
- Ectopic Pregnancy Diagnosis and Management
- Gynecological conditions and treatments
- Gestational Trophoblastic Disease Studies
- Network Security and Intrusion Detection
- Traffic Prediction and Management Techniques
- Salivary Gland Tumors Diagnosis and Treatment
- Radiomics and Machine Learning in Medical Imaging
- Human Mobility and Location-Based Analysis
- Internet Traffic Analysis and Secure E-voting
- Model Reduction and Neural Networks
- Adversarial Robustness in Machine Learning
- Endometriosis Research and Treatment
- Probabilistic and Robust Engineering Design
- Advanced Graph Neural Networks
- Minimally Invasive Surgical Techniques
- Privacy, Security, and Data Protection
- Neural Networks and Applications
- Cryptography and Data Security
- Remote-Sensing Image Classification
- Emotion and Mood Recognition
- Geochemistry and Geologic Mapping
- Access Control and Trust
- Currency Recognition and Detection
University of Naples Federico II
2021-2024
FHNW University of Applied Sciences and Arts
2018
Ospedale San Paolo
2010-2012
University of Milan
2002-2010
Biotherapy of Genetic Diseases, Inflammatory Disorders and Cancers
2009
Ospedale Maggiore
2006
Meteorological conditions have a strong influence on air quality and can play an important role in prediction. However, due to the "black-box" nature of deep learning, it is difficult obtain trustworthy learning models when considering meteorological To address above problem, this paper, we reveal prediction by utilizing explainable learning. In (1) source data from pollutant datasets, including PM <sub xmlns:mml="http://www.w3.org/1998/Math/MathML"...
Abstract Nowadays, in the Scientific Machine Learning (SML) research field, traditional machine learning (ML) tools and scientific computing approaches are fruitfully intersected for solving problems modelled by Partial Differential Equations (PDEs) science engineering applications. Challenging SML methodologies new computational paradigms named Physics-Informed Neural Networks (PINNs). PINN has revolutionized classical adoption of ML computing, representing a novel class promising...
Network Intrusion Detection Systems (NIDS) are crucial tools for protecting networked devices from cyberattacks. Recent development in the field of Artificial Intelligence (AI) has provided tremendous advantages implementing NIDSs able to monitor network traffic and block cyberattacks real-time. In literature, it is widely recognized that effective training a NIDS requires large quantity labeled traffic, representative attacks. Nonetheless, availability public abundant datasets remains...
Nowadays, predictive medicine begins to become a reality thanks Artificial Intelligence (AI) which allows, through the processing of huge amounts data, identify correlations not perceptible human brain. The application AI in diagnostics is increasingly pervasive; use and interpretation first signs some diseases (i.e. tumours) can be detected help physicians make more accurate diagnoses reduce errors develop methods for individualized medical treatment. In this perspective, salivary gland...
New technologies are transforming medicine, and this revolution starts with data. Usually, health services within public healthcare systems accessed through a booking centre managed by local authorities controlled the regional government. In perspective, structuring e-health data Knowledge Graph (KG) approach can provide feasible method to quickly simply organize and/or retrieve new information. Starting from raw bookings system in Italy, KG is presented support extraction of medical...
Artificial intelligence-driven automation has gradually become the technical trend of new era. At present, many artificial intelligence technologies have been applied to improve level in field automation. Among them, convolutional neural network (CNN) technology is one most representative, which used detection defective products industrial automation, robot human tracking widely machine vision driven However, high dependence current application leads potential failure product system. In this...
The Forward-forward (FF) algorithm is a new method for training neural networks, proposed as an alternative to the traditional Backpropagation (BP) by Hinton. FF replaces backward computations in learning process with another forward pass. Each layer has objective function, which aims be high positive data and low negative ones. This paper presents preliminary investigation into variations of algorithm, such incorporating local create hybrid network that robustly converges while preserving...
Traffic forecasting is a crucial aspect of modern Intelligent Transportation Systems (ITS) and the Internet Vehicles (IoV), playing vital role in improving safety efficiency daily transportation activities. Despite valuable contributions traditional machine learning (ML) models advanced deep (DL) techniques, there persist challenges capturing intricate spatial temporal dependencies inherent traffic flow. In response to these challenges, we present GRAPHITE, an innovative framework that...
This paper investigates the symbiosis of Federated Learning (FL) and High-Performance Computing (HPC) architectures, unraveling challenges introduced by intricate interplay heterogeneity non-Independently Identically Distributed (non-lID) data. By leveraging Flower framework, our research delves into nuanced implications FL in diverse HPC environments. We provide a comprehensive exploration within contemporary spanning node organizations, memory hierarchies, special-ized accelerators,...
ABSTRACT Municipal waste management (MWM) poses significant challenges in the context of rapid urbanisation and population growth. Accurate forecasting production is crucial for designing sustainable strategies. However, traditional methods often struggle to capture complexities generation dynamics. This paper proposes a novel methodology leveraging deep learning techniques forecast municipal production. By harnessing power neural networks, our approach transcends limitations conventional...
In the context of rapidly advancing smart cities, efficient crowd analysis plays a crucial role in ensuring public safety, urban planning, and resource management. This paper presents novel framework that combines popular You Only Look Once (YOLO) object detection algorithm with advanced techniques, aiming to improve understanding management dynamics. The proposed leverages YOLO's real-time capabilities detect various objects within video frames, particular focus on identifying individuals....