BPBarbara Puccio

Barbara Puccio

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Barbara Puccio has a total of 37 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Deep Learning-Based Malaria Parasite Image Classification on Real Microscopy Data

    by Francesco Branda, Lilia Andriani, Riccardo Lucis, Annamaria Defilippo, Ugo Lomoio, Barbara Puccio, Giancarlo Ceccarelli, Massimo Ciccozzi, Fabio Scarpa, Pierangelo Veltri, et al.

    Background: Accurate and timely malaria diagnosis with species-level identification of Plasmodium parasites is critical for guiding effective treatment and disease management. Although light microscopy remains the diagnostic gold standard, its dependence on highly trained personnel limits accessibility, particularly in resource-constrained settings. Recent advances in deep learning have enabled automated image-based diagnosis with promising performance; however, reliable differentiation among Plasmodium species continues to pose a major challenge. This study aims to evaluate and compare the effectiveness of different deep learning architectures for automated species identification from microscopy images. Methods: Three deep learning architectures were systematically assessed: a convolutional backbone (ResNet50), a Vision Transformer (ViT), and a hybrid ResNetViT model. All models were trained from scratch, without using pretrained weights, on a dataset comprising real-world thick blood smear images augmented with publicly available microscopy data from Kaggle. In the hybrid architecture, the ResNet module was used to extract robust local morphological features, while the ViT component captured long-range dependencies and contextual relationships within images. Results: In cross-validation experiments, all three architectures consistently achieved high diagnostic performance. ResNet50 attained the highest accuracy (96.9%), an F1-score of 96.3%, and an ROC-AUC of 0.997. The Vision Transformer achieved 93.1% accuracy, 91.7% F1-score, and an ROC-AUC of 0.989. The hybrid ResNetViT reached 95.2% accuracy, 94.2% F1-score, and an ROC-AUC of 0.995. These results confirm that all architectures can reliably distinguish among Plasmodium species. Although ResNet50 achieved the highest raw accuracy, the hybrid model showed the most stable calibration and the closest qualitative agreement between its attention maps and expert-annotated parasite locations. Conclusions: The findings demonstrate that convolutional, transformer-based, and hybrid deep learning architectures can be successfully trained on real microscopy data for species-level malaria diagnosis. These results support the feasibility of deploying scalable, automated diagnostic systems to improve both accuracy and accessibility of malaria detection, particularly in resource-limited healthcare settings.


  • Open Access

    Cross-domain integration of transcriptomic modalities through graph neural networks

    by Barbara Puccio, Pietro Hiram Guzzi, Stefania Scalise, Raffaele Covello, Elvira Immacolata Parrotta, Pierangelo Veltri

  • Open Access

    Extracting Clinical Insights: LLM-Powered EHR Knowledge Graphs

    by Caterina Francesca Perri, Annamaria Defilippo, Valentina Carbonari, Ugo Lomoio, Barbara Puccio

    Electronic Health Records (EHRs) represent a rich source of structured clinical information that is difficult to query at scale using traditional methods. This work proposes an end-to-end pipeline that parses synthetic EHR data, organises clinical entities into a Neo4j Knowledge Graph (KG), and enables LLM-powered question answering (QA) over patient data. The pipeline leverages open-source Qwen models for both embedding generation and natural language to Cypher query translation, combined with a LlamaIndex-based vector retrieval layer. A comparative evaluation between a vector-only RAG baseline and a KG-augmented RAG pipeline is performed using an LLM-as-Judge framework, with qualitative results suggesting improved reasoning consistency when structured graph knowledge is incorporated. An interactive Streamlit demo allows step-by-step exploration of the full pipeline.

  • Open Access

    The Goofy Game: An approach to medical AI misalignment

    by Barbara Puccio, Federico Castagna, Allan Tucker, Pierangelo Veltri

  • Open Access

    Cytoskeletal Imbalance and Axonal Vulnerability in Sporadic PSP-RS: Early Changes in a Human iPSC-Derived Neuronal Model with Altered mTOR Signaling

    by Raffaele Covello, Giorgia Lucia Benedetto, Stefania Scalise, Caterina Gabriele, Desirèe Valente, Clara Zannino, Barbara Puccio, Andrea Quattrone, Pietro Hiram Guzzi, Marco Gaspari, et al.

    Progressive supranuclear palsy-Richardson’s syndrome (PSP-RS) is a primary 4R tauopathy in which early axonal dysfunction may precede overt neurodegeneration; however, the mechanisms linking Tau dysregulation to cytoskeletal vulnerability remain poorly defined. Here, we generated induced pluripotent stem cell (iPSC)-derived midbrain dopaminergic neurons from individuals with sporadic PSP-RS and matched healthy controls and performed integrated transcriptomic and proteomic analyses. PSP-RS neurons exhibited coordinated suppression of dopaminergic and synaptic programs alongside activation of cytoskeletal remodeling and stress-related pathways. These changes were accompanied by increased Tau phosphorylation, neurofilament accumulation, and structural alterations of the axonal compartment, consistent with an early axonopathic phenotype. Notably, mechanistic target of rapamycin (mTOR) signaling significantly increased. Pharmacological inhibition of mTOR reduced Tau phosphorylation and neurofilament levels, indicating that mTOR activity contributes to the maintenance of cytoskeletal imbalance. In conclusion, our findings support a model in which early cytoskeletal dysfunction in PSP-RS arises from the convergence of Tau dysregulation, impaired structural homeostasis, and altered signaling pathways. Rather than acting as a primary driver, mTOR appears to function as a pathogenic amplifier that sustains axonal stress. This study provides a human cellular framework to investigate early axonopathic mechanisms in sporadic PSP-RS.

  • Open Access

    From data to decisions: a modular platform for modelling and simulation of infectious disease diffusion in networks

    by Francesco Branda, Annamaria Defilippo, Ugo Lomoio, Barbara Puccio, Massimo Ciccozzi, Fabio Scarpa, Pierangelo Veltri, Pietro Hiram Guzzi

    Accurately modelling diffusion dynamics in complex networks is essential for improving medical outcomes, guiding pandemic preparedness, and optimizing reso

  • Open Access

    Artificial Intelligence–Driven Structural Mining Enables Functional Inference in the Human Dark Proteome

    by Valentina Carbonari, Annamaria Defilippo, Ugo Lomoio, Caterina Francesca Perri, Barbara Puccio, Pierangelo Veltri, Pietro Hiram Guzzi

    The rapid diffusion of high-throughput sequencing technologies has generated a vast repertoire of protein-coding se- quences whose biological roles remain unknown. This discrepancy between sequence availability and functional under- standing has led to the definition of the dark proteome, comprising proteins or protein regions that lack experimentally resolved structures and reliable functional annotations. Classical sequence-based approaches often fail to characterize these targets due to extreme sequence divergence, intrinsic disorder, or membrane localization. Here, we present an inte- grated, structure-centric computational framework that leverages recent advances in artificial intelligence to enable func- tional inference in the human dark proteome. By combining deep learning–based protein structure prediction, large-scale structural alignment, and machine learning–driven surface pocket analysis, we uncover remote evolutionary relationships and conserved functional features that remain invisible to traditional bioinformatics pipelines. Our results demonstrate that artificial intelligence provides a powerful strategy to bridge the gap between genomic information and biological function, opening new avenues for systematic exploration of uncharacterized regions of the human proteome.

  • Open Access

    From Data to Decisions: A Modular Platform for modelling and Simulation of Infectious Disease Diffusion in Networks

    by Francesco Branda, Annamaria Defilippo, Ugo Lomoio, Barbara Puccio, Massimo Ciccozzi, Fabio Scarpa, Pierangelo Veltri, Pietro Hiram Guzzi

    Accurately modelling diffusion dynamics in complex networks is essential for improving medical outcomes, guiding pandemic preparedness, and optimizing resource allocation in public health. However, existing approaches often face a trade-off between predictive performance and model interpretability, limiting their utility for clinical decision-making and strategic planning. This study presents a modular computational methodology that integrates classical compartmental models with graph neural networks (GNNs) and explainable artificial intelligence (XAI) to simulate, analyse, and interpret the spread of contagion across heterogeneous network topologies. The approach captures both structural and temporal dimensions of diffusion processes, enabling granular insights into transmission pathways. Simulations are applied to critical public health scenarios, including the identification of super-spreaders and the assessment of targeted containment strategies. By combining mechanistic models with data-driven learning and explainability techniques, the methodology supports outcome forecasting, scenario comparison, and the interpretation of network-based risk factors. Results demonstrate the ability to predict diffusion trajectories with high accuracy while preserving transparency in decision-relevant variables. The approach is intended as a generalizable tool to support medical modelling and simulation with applications ranging from epidemic control to personalized risk assessment and cost-effective intervention planning.

  • Open Access

    ExDiff: A Modular and Explainable Framework Combining Network Simulation and Graph Neural Networks for Diffusion Modelling

    by Annamaria Defilippo, Ugo Lomoio, Barbara Puccio, Pierangelo Veltri, Pietro Hiram Guzzi

    Understanding and modelling diffusion processes in complex networks is critical across disciplines, including epidemiology, sociology, and information science. Despite considerable progress, existing approaches often struggle to balance predictive accuracy with interpretability, constraining their applicability in real-world decision-making. ExDiff is introduced as an interactive and modular computational framework that integrates network simulation, Graph Neural Networks (GNNs), and eXplainable Artificial Intelligence (XAI) to both model and elucidate diffusion dynamics. By combining classical compartmental models with deep learning architectures, ExDiff captures the structural and temporal features of diffusion across heterogeneous network topologies. The framework includes modules designed for network analysis, neural modelling, simulation, and interpretability. Its effectiveness is demonstrated through applications to epidemic modelling, including simulation of disease spread, evaluation of intervention strategies, and identification of structural drivers of contagion via XAI techniques. The framework is available in the following GitHub repository: https://github.com/hguzzi/ExDiff.git.