NAVIGATOR: an Italian regional imaging biobank to promote precision medicine for oncologic patients

Rita Borgheresi, Andrea Barucci, Sara Colantonio, Gayanè Aghakhanyan, Massimiliano Assante, Elena Bertelli, Emanuele Carlini, Roberto Carpi, Claudia Caudai, Diletta Cavallero, Dania Cioni, Roberto Cirillo, Valentina Colcelli, Andrea Dell’Amico, Domnico Di Gangi, Paola Anna Erba, Lorenzo Faggioni, Zeno Falaschi, Michela Gabelloni, Rosa Gini, Lucio Lelii, Píetro Lió, Antonio Lorito, Silvia Lucarini, Paolo Manghi, Francesco Mangiacrapa, Chiara Marzi, Maria Antonietta Mazzei, Laura Mercatelli, Antonella Mirabile, Francesco Mungai, Vittorio Miele, M Olmastroni, Pasquale Pagano, Fabiola Paiar, Giancarlo Panichi, Maria Antonietta Pascali, F Pasquinelli, Jorge Eduardo Shortrede, Lorenzo Tumminello, Luca Volterrani, Emanuele NeriEuropean 2022 · DOI

Abstract

NAVIGATOR is an Italian regional project boosting precision medicine in oncology with the aim of making it more predictive, preventive, and personalised by advancing translational research based on quantitative imaging and integrative omics analyses. The project's goal is to develop an open imaging biobank for the collection and preservation of a large amount of standardised imaging multimodal datasets, including computed tomography, magnetic resonance imaging, and positron emission tomography data, together with the corresponding patient-related and omics-related relevant information extracted from regional healthcare services using an adapted privacy-preserving model. The project is based on an open-source imaging biobank and an open-science oriented virtual research environment (VRE). Available integrative omics and multi-imaging data of three use cases (prostate cancer, rectal cancer, and gastric cancer) will be collected. All data confined in NAVIGATOR (i.e., standard and novel imaging biomarkers, non-imaging data, health agency data) will be used to create a digital patient model, to support the reliable prediction of the disease phenotype and risk stratification. The VRE that relies on a well-established infrastructure, called D4Science.org, will further provide a multiset infrastructure for processing the integrative omics data, extracting specific radiomic signatures, and for identification and testing of novel imaging biomarkers through big data analytics and artificial intelligence.

Introduction

Precision medicine relies heavily on the ability to extract actionable insights from vast quantities of quantitative data. However, a significant "translational gap" exists in modern oncology: while hospitals generate massive amounts of medical imaging, these data typically remain siloed within Picture Archiving and Communication Systems (PACS). These systems are optimized for clinical viewing and diagnosis rather than large-scale data mining or cross-institutional research.

The NAVIGATOR project addresses this hurdle by establishing a first-of-its-kind regional imaging biobank in Tuscany, Italy. By integrating multimodal imaging (CT, MRI, PET) with clinical records and "omics" data, NAVIGATOR aims to bridge the gap between raw medical images and personalized cancer treatment through the creation of comprehensive "digital patient models."

System Architecture and Methodology

The NAVIGATOR framework is designed to handle the technical, ethical, and legal complexities of sharing sensitive medical data across a regional health network. The architecture is built upon two primary pillars:

  1. The Biobank Infrastructure: Unlike a standard PACS, this biobank utilizes a flexible data model compliant with international standards such as DICOM, MIABIS, and OMOP-CDM. It employs a dual-storage strategy: a Postgres database manages clinical and molecular variables, while a structured file-based repository stores the high-resolution DICOM images.
  2. The Virtual Research Environment (VRE): Powered by the D4Science.org infrastructure and the gCube software toolkit, the VRE serves as a centralized "digital laboratory." It provides researchers with web-based access to high-performance computing resources. Within this environment, clinicians and scientists can execute radiomics extraction pipelines and develop AI models using familiar tools like Jupyter Notebooks and RStudio, without requiring extensive backend IT expertise.
The multi-layered architecture of the NAVIGATOR project, illustrating the flow from data sources to the Virtual Research Environment and specific clinical use cases.

A core innovation of the project is the Digital Patient Model. This model integrates imaging biomarkers with "holomics" (genomics, immunomics) and administrative health records from the Agenzia Regionale di Sanità (Regional Health Agency) to provide a holistic view of the disease phenotype and improve risk stratification.

Clinical Focus and Datasets

The project initially focuses on three primary oncological use cases, with a target of collecting data from over 1,500 patient cases across the Tuscany region:

  • Prostate Cancer: Utilizing multiparametric MRI to improve risk stratification and diagnostic accuracy.
  • Rectal Cancer: Identifying "complete responders" to neoadjuvant therapy, which could allow some patients to avoid invasive surgery.
  • Gastric Cancer: Developing models to predict patient response to multimodal treatment regimens.

Data is aggregated from multiple clinical centers, including the University of Pisa and various regional hospitals, ensuring a diverse and representative dataset for the Tuscan population.

Strategic Innovations

NAVIGATOR introduces several key concepts to the field of medical data management:

  • Biobank vs. PACS: The project emphasizes that the value of an imaging biobank lies in the explicit linkage between images and the quantitative biomarkers (radiomics) extracted from them, a feature missing in standard clinical systems.
  • The "Broad Consent" Model: To comply with GDPR while facilitating research, NAVIGATOR uses a "broad consent" framework. This allows data to be used for future ethical research projects without the need to re-contact patients for every individual study.
  • Democratizing AI: By hosting analytical tools within the VRE, the project lowers the barrier to entry for medical professionals to engage in big data analytics and AI development.

Current Status and Future Outlook

As a position paper, NAVIGATOR reports significant progress in infrastructure establishment. The underlying D4Science platform is already highly scalable, serving over 7,000 users. The project has successfully established standardized acquisition and segmentation protocols across regional centers to ensure data reproducibility—a common pitfall in radiomics research.

However, challenges remain. The Italian legal system currently lacks specific legislation for research biobanks, requiring the project to navigate complex "soft law" interpretations of GDPR. Furthermore, while the current focus is on three abdominal cancers, future work will involve scaling the infrastructure to include other malignancies and connecting NAVIGATOR to international networks like the European Biobanking and BioMolecular Resources Research Infrastructure (BBMRI-ERIC).

Summary updated

Abstract

NAVIGATOR is an Italian regional project boosting precision medicine in oncology with the aim of making it more predictive, preventive, and personalised by advancing translational research based on quantitative imaging and integrative omics analyses. The project's goal is to develop an open imaging biobank for the collection and preservation of a large amount of standardised imaging multimodal datasets, including computed tomography, magnetic resonance imaging, and positron emission tomography data, together with the corresponding patient-related and omics-related relevant information extracted from regional healthcare services using an adapted privacy-preserving model. The project is based on an open-source imaging biobank and an open-science oriented virtual research environment (VRE). Available integrative omics and multi-imaging data of three use cases (prostate cancer, rectal cancer, and gastric cancer) will be collected. All data confined in NAVIGATOR (i.e., standard and novel imaging biomarkers, non-imaging data, health agency data) will be used to create a digital patient model, to support the reliable prediction of the disease phenotype and risk stratification. The VRE that relies on a well-established infrastructure, called D4Science.org, will further provide a multiset infrastructure for processing the integrative omics data, extracting specific radiomic signatures, and for identification and testing of novel imaging biomarkers through big data analytics and artificial intelligence.
BibTeX
@inproceedings{borgheresi2022navigator,
  author = {Rita Borgheresi and Andrea Barucci and Sara Colantonio and Gayanè Aghakhanyan and Massimiliano Assante and Elena Bertelli and Emanuele Carlini and Roberto Carpi and Claudia Caudai and Diletta Cavallero and Dania Cioni and Roberto Cirillo and Valentina Colcelli and Andrea Dell’Amico and Domnico Di Gangi and Paola Anna Erba and Lorenzo Faggioni and Zeno Falaschi and Michela Gabelloni and Rosa Gini and Lucio Lelii and Píetro Lió and Antonio Lorito and Silvia Lucarini and Paolo Manghi and Francesco Mangiacrapa and Chiara Marzi and Maria Antonietta Mazzei and Laura Mercatelli and Antonella Mirabile and Francesco Mungai and Vittorio Miele and M Olmastroni and Pasquale Pagano and Fabiola Paiar and Giancarlo Panichi and Maria Antonietta Pascali and F Pasquinelli and Jorge Eduardo Shortrede and Lorenzo Tumminello and Luca Volterrani and Emanuele Neri},
  title = {{NAVIGATOR: an Italian regional imaging biobank to promote precision medicine for oncologic patients}},
  booktitle = {European},
  year = {2022},
  doi = {10.1186/s41747-022-00306-9},
}
Figure