A5004599844- Advanced Malware Detection Techniques
- Network Security and Intrusion Detection
- Formal Methods in Verification
- Software Testing and Debugging Techniques
- Radiomics and Machine Learning in Medical Imaging
- AI in cancer detection
- Logic, programming, and type systems
- Anomaly Detection Techniques and Applications
- Software Reliability and Analysis Research
- Software Engineering Research
- Model-Driven Software Engineering Techniques
- Digital and Cyber Forensics
- Advanced Software Engineering Methodologies
- Security and Verification in Computing
- Autonomous Vehicle Technology and Safety
- COVID-19 diagnosis using AI
- Advanced Neural Network Applications
- Business Process Modeling and Analysis
- Petri Nets in System Modeling
- Brain Tumor Detection and Classification
- Retinal Imaging and Analysis
- Advanced X-ray and CT Imaging
- Adversarial Robustness in Machine Learning
- Smart Grid Security and Resilience
- Lung Cancer Diagnosis and Treatment
University of Molise
2017-2025
Institute of Informatics and Telematics
2019-2024
National Research Council
2019-2024
University of Pisa
1996-2022
University of Sannio
2006-2017
University of Chieti-Pescara
2016
Grading laryngeal squamous cell carcinoma (LSCC) based on histopathological images is a clinically significant yet challenging task. However, more low-effect background semantic information appeared in the feature maps, channels, and class activation which caused serious impact accuracy interpretability of LSCC grading. While traditional transformer block makes extensive use parameter attention, model overlearns information, resulting ineffectively reducing proportion semantics. Therefore,...
Security and non-security requirements are two critical issues in software development. Classifying is crucial as it aids recalling security needs during the early stages of development, ultimately leading to enhanced final solution. However, remains a challenging task classify into categories automatically. In this work, we propose novel method for automatically classifying using transformer models address these challenges. fine-tuned four pre-trained transformers datasets (the original one...
Medical studies demonstrated that diabetes pathology is increasing in last decades and the trend do not tends to stop. In order help accelerate diagnosis of this paper we propose a method able classify patients affected by using set characteristic selected according World Health Organization criteria. Evaluating real-world data state art machine learning algorithms, obtain precision value equal 0.770 recall 0.775 HoeffdingTree algorithm.
Modern vehicles have lots of connectivity, this is the reason why protect in-vehicle network from cyber-attacks becomes an important issue. The Controller Area Network a de facto standard for network. However, lack security features CAN protocol makes vulnerable to attacks. message injection attack representative type which injects fabricated messages deceive original Electronic Control Units or cause malfunctions. In paper we propose method able detect four different attacks targeting...
At the end of 2019, a new form Coronavirus, called COVID-19, has widely spread in world. To quickly screen patients with aim to detect this pulmonary disease, paper we propose method aimed automatically COVID-19 disease by analysing medical images. We exploit supervised machine learning techniques building model considering data-set freely available for research purposes 85 chest X-rays. The experiment shows effectiveness proposed discrimination between and other diseases.
The tumor grading of laryngeal cancer pathological images needs to be accurate and interpretable. deep learning model based on the attention mechanism-integrated convolution (AMC) block has good inductive bias capability but poor interpretability, whereas vision transformer (ViT) interpretability weak ability. Therefore, we propose an end-to-end ViT-AMC network (ViT-AMCNet) with adaptive fusion multiobjective optimization that integrates fuses ViT AMC blocks. However, existing methods often...
The coronavirus is caused by the infection of SARS-CoV-2 virus: it represents a complex and new condition, considering that until end December 2019 this virus was totally unknown to international scientific community. clinical management patients with disease has undergone an evolution over months, thanks increasing knowledge virus, symptoms efficacy various therapies. Currently, however, there no specific therapy for know also as Coronavirus 19, treatment based on patient taking into...
Most of death causes are related to cardiovascular disease. In fact, there several anomalies afflicting the heart beat, for instance murmur or artefact. We propose a method disease detection. By gathering set feature obtainable directly from cardiac sounds, we consider this vector as input deep neural network discriminate whether sound is belonging an healthy patient with The experiment performed demonstrated effectiveness proposed approach in real-world environment.
Modern cars include a huge number of sensors and actuators, which continuously exchange data control commands. The most used protocol for communication different components in automotive system is the Controller Area Network (CAN). According to CAN, communicate by broadcasting messages on bus. In addition, standard definition does not provide information authentication, so exposing it attacks. This paper proposes method based deep learning aiming at discovering attacks towards CAN-bus....
Abstract Objective Due to the COVID-19 pandemic, our daily habits have suddenly changed. Gatherings are forbidden and, even when it is possible leave home for health or work reasons, necessary wear a face mask reduce possibility of contagion. In this context, crucial detect violations by people who do not mask. Materials and Methods For these in article, we introduce method aimed automatically whether wearing We design transfer learning approach exploiting MobileNetV2 model identify...
Laryngeal cancer tumor (LCT) grading is a challenging task in P63 Immunohistochemical (IHC) histopathology images due to small differences between LCT levels pathology images, the lack of precision lesion regions interest (LROIs) and paucity image samples. The key solving problem transfer knowledge from other identify more accurate LROIs, but following problems occur: 1) transferring without priori experience often causes negative creates heavy workload abundance types, 2) convolutional...
Background: Liver metastases are a leading cause of cancer-associated deaths in patients affected by colorectal cancer (CRC). The multidisciplinary strategy to treat CRC is more effective when the radiological diagnosis accurate and early. Despite evolving technologies accuracy, Colorectal Cancer Metastases (CRCLM) still key point. aim our study was define new patient representation different Artificial Intelligence models, using Formal Methods (FMs), help clinicians predict presence liver...