Department of Medical and Surgical Sciences (DIMEC)
Gastone Castellani
Full Professor of Applied Physics and Biophysics | University of Bologna
Bologna, Italy | Email: Gastone.Castellani@unibo.it | ORCID: 0000-0003-4892-925X
Web: unibo.it/sitoweb/gastone.castellani/en | Google Scholar: scholar profile
Department of Medical and Surgical Sciences (DIMEC)
Profile
Gastone Castellani is Full Professor of Applied Physics and Biophysics at the University of Bologna
and former Director of the Galvani Center for Bioinformatics, Biophysics and Biocomplexity. His
research focuses on artificial intelligence, machine learning, systems biology, and advanced statistical
modeling for biomedical data, with particular emphasis on multimodal data integration, predictive
modeling, synthetic data generation, and personalized medicine.
A central component of his work is the development of AI-driven methodologies for clinical decision
support, including risk stratification, prognosis, and treatment optimization in complex diseases,
particularly in oncology and hematology. His research also includes Bayesian and probabilistic
modeling approaches, enabling robust inference, uncertainty quantification, and interpretable machine
learning in high-dimensional biomedical settings.
He has made significant contributions to synthetic data generation, developing generative models
and simulation-based frameworks for multi-omics and medical imaging data, supporting data sharing,
privacy-preserving analytics, and the validation of AI systems in healthcare. His work integrates
heterogeneous data sourcesgenomics, transcriptomics, imaging, and clinical datathrough networkbased
and statistical physics-inspired approaches.
He has also made foundational contributions to the BCM theory of bidirectional synaptic plasticity,
extending Hebbian learning principles through biophysically grounded and stochastic formulations.
He has developed interdisciplinary research at the interface of physics, biology, medicine, and data
science, and has been actively involved in numerous national, European, and international AI for
Medicine projects, serving as Principal Investigator, Co-Principal Investigator, and participant.
He has also served as Associate Research Professor at Brown University for over 20 years, conducting
research on neural networks, learning systems, and large-scale biological data analysis, including
collaborative work with Prof. Leon N. Cooper (Nobel Laureate in Physics), with whom he has been
associated as a researcher.
Publications: over 200 peer-reviewed papers Citations: approx. 16,000 H-index: 57
Research Interests
• Artificial intelligence and machine learning for biomedical and clinical data
• AI-driven clinical decision support systems and predictive modeling in precision medicine
• Bayesian and probabilistic modeling for high-dimensional biomedical data
• Synthetic data generation, generative models, and privacy-preserving data analytics
• Federated learning and distributed AI for healthcare applications
• Systems biology, multi-omics integration, and network medicine
• Integration of heterogeneous data (genomics, transcriptomics, imaging, clinical data)
• Medical imaging, radiomics, and computational imaging analysis
• Statistical physics, stochastic processes, and complex systems modeling
• Theoretical foundations of learning systems and neural models (including BCM theory)
• Computational biomedicine and translational AI applications in oncology, hematology, and aging
Academic Positions
Full Professor of Applied Physics and Biophysics 2015–present
University of Bologna, Department of Medical and Surgical Sciences
SSD FIS/07. Teaching and research in medical physics, biophysics, AI, and biomedical data science.
Director, Galvani Center for Bioinformatics, Biophysics and Biocomplexity 2010–2020
University of Bologna
Scientific leadership of interdisciplinary research on bioinformatics, biophysics, and AI for medicine.
Associate Research Professor 2000–2020
Brown University, Providence, RI, USA
Research collaboration on machine learning, neural networks, and large-scale gene expression analysis.
Associate Professor of Physics 2005–2010
University of Bologna, Department of Physics
Research and teaching in applied physics and biophysics.
Researcher in Physics 1996–2005
University of Bologna
Faculty of Veterinary Medicine.
Postdoctoral Fellow 1995
CNR, Italy
Postdoctoral research activity in biophysics and modeling.
Visiting Scientist 1988–1990
Società Italiana Sincrotrone, Trieste, Italy
Research activity in synchrotron-related biophysical studies.
Education
PhD in Physics 1995
University of Bologna
Field: Biophysics. Dissertation on network models of the immune system.
Laurea / MSc in Physics 1992
University of Bologna
Field: Biophysics. Mathematical modeling of the immune system.
Laurea / MSc in Biology 1988
University of Bologna
Field: Chemistry and Biophysics. X-ray diffraction studies of collagen.
Major Scientific Contributions
• Development of AI- and machine learning-based approaches for stratification, prognosis, and
decision support in hematological malignancies and personalized medicine.
• Contributions to synthetic data generation frameworks and federated learning pipelines for
biomedical applications.
• Integration of multi-omics, imaging, and clinical data through statistical and network-based
models.
• Foundational work on stochastic and biophysical modeling, including neural learning models,
entropy-based descriptions, and complex systems approaches.
• Translational applications in radiomics, microbiome modeling, aging, oncology, and computational medicine.
Competitive Funding and European Projects
SYNTHIA – Innovative Health Initiative 2024–2029
University of Bologna, Use Case Leder
Synthetic data generation framework for integrated validation of use cases and AI healthcare
applications. Project no. 101172872.
METASTRA – Horizon Europe 2023–2028
University of Bologna, Task Leader
Computational models for fracture risk stratification in patients with vertebral metastases. Project
no. 101080135.
AIRC Individual Grant 2022-2029
Principal Investigator / key investigator
Artificial intelligence for genomics and personalized medicine in myelodysplastic syndromes.
GENOMED4ALL – Horizon 2020 2021-2026
University of Bologna, WP Leader of AI methods
Genomics and personalized medicine through AI in hematological diseases.
In Silico World – Horizon 2020 2021–2024
University of Bologna, Task Leader
Lowering barriers to the adoption of in silico trials. Project no. 101016503.
VEO – Horizon 2020 2019-2025
University of Bologna, Participant
Data analytics and modeling for emerging infectious disease observatory.
HARMONY PLUS 2019-2023
University of Bologna, WP Co-Leader
Big biomedical data integration for hematological malignancies.
HARMONY 2017-2022
University of Bologna, WP Leader for the AI and Big Data algorithms
Large-scale data analytics and computing infrastructure for hematology.
IMforFUTURE – ITN 2017-2023
University of Bologna, WP Leader and scientific co-coordinator
Training network in methods for future data.
Leadership, Service, and Academic Roles
• Head of the DIMEC (Department of Medical and Surgical Sciences) Artificial Intelligence, Data
sharing and privacy working group
• Director, Specialty School in Medical Physics, University of Bologna
• Coordinator of the PhD Program in Physics, University of Bologna
• Vice Director, Specialty School in Medical Physics, University of Bologna
• Member of research committees for European grants and academic evaluation
• Co-founder of the Galvani Center for Bioinformatics, Biophysics and Biocomplexity
• Co-founder of SysBioHealth / systems biology scientific initiatives
• Referee for international journals including PNAS, Bioinformatics, Physical Review, and others
Selected Publications
Artificial Intelligence and Precision Medicine
• Castellani G., et al. Clinical and Genomic-Based Decision Support System to Define the Optimal
Timing of Allogeneic Hematopoietic Stem-Cell Transplantation in Patients With Myelodysplastic
Syndromes. Journal of Clinical Oncology, 2024.
• Castellani G., et al. Characterization and Clinical Implications of p53 Dysfunction in Patients
With Myelodysplastic Syndromes. Journal of Clinical Oncology, 2025.
• Castellani G., et al. Data-driven, harmonised classification system for myelodysplastic syndromes.
The Lancet Haematology, 2024.
• Castellani G., et al. Machine learning risk stratification strategy for multiple myeloma.
HemaSphere, 2025.
• Castellani G., et al. Artificial Intelligence model to predict resistances in Gram-negative
bloodstream infections. npj Digital Medicine, 2025.
Systems Biology, Multi-Omics, and Network Medicine
• Castellani G., et al. Covering Hierarchical Dirichlet Mixture Models on binary data to enhance
genomic stratifications in onco-hematology. PLOS Computational Biology, 2024.
• Castellani G., et al. Systems medicine of inflammaging. Briefings in Bioinformatics, 2016.
• Castellani G., et al. Network integration of multi-tumour omics data suggests novel targeting
strategies. Nature Communications, 2018.
Biophysical and Theoretical Modeling
• Castellani G., et al. Toward a microscopic model of bidirectional synaptic plasticity. PNAS,
2009.
• Castellani G., et al. A biophysical model of bidirectional synaptic plasticity. PNAS, 2001.
• Castellani G., et al. Energy Consumption and Entropy Production in a Stochastic Formulation
of BCM Learning. Journal of Computational Biology, 2021.
Awards and Distinctions
• Chaos and Complexity Prize, Blois, France
• Vice President Research Award, Brown University
• Research fellowships and awards from CNR, SIS, INFN, and other institutions
International Collaborations
Active collaborations include Brown University, European consortia in hematology and AI for
medicine, and interdisciplinary networks in systems biology, radiomics, synthetic data generation,
and computational biomedicine.
Teaching and Mentoring
Teaching activity includes medical physics, applied physics, biophysics, artificial intelligence in
medicine, and quantitative methods for biomedical data. Supervision includes PhD students, medical
physics residents, postdoctoral fellows, and interdisciplinary research teams.
Languages and Technical Skills
Languages: Italian (native), English (professional proficiency)
Programming / technical skills: Python, C, C++, Linux, machine learning workflows, biomedical data analysis, statistical modeling, AI methods for multimodal biomedical data.
Bologna,
