Institutional and academic context

Research and institutional partners

An IRCCS research hospital — IFOEmbedded university teaching, direct translational research and the HEAL Italia consortiumConsortium / network

The framework must satisfy data governance compatible with hospital operations, clinical research and public accountability; mapping tables, quality rules and audit logs are what make logistics data usable inside a research-intensive institution.

An academic spin-off — Clepio BiotechUniversity of Florence, LENS and CNR research, extended to the University of Bern through 3D PATHOfficial source

The framework performs technology transfer at the data level: raw-data preservation, explicit schemas, versioning and quality gates give a university-born scientific method the controls required for industrial operation and collaborative research.

A university-origin engineering firm — Beta 80Founded by Politecnico di Milano graduates; five university relationships; a decade-long partnership with Università CattolicaOfficial source

Emergency-platform work contributes source fidelity, task-level lineage and the controlled reuse of sensitive public-safety data, together with a formation pipeline of engineers who read those structures.

Lineage as the bridge to academic trustRead across the three ecosystemsProject document

Mapping tables and audit logs at IFO, raw JSON, batch identities, checksums and schema versions in OME-Zarr, TASK and TASK_COLUMN in RMS-EMMA: the same control makes analysis reproducible and research conclusions defensible, and turns a Data Object architecture into a shared language among clinicians, laboratories, engineers and public-sector specialists.

Standards and specifications

Medallion layering (Bronze · Silver · Gold)Lakehouse architecture literature

Source-preserving Bronze, reconstructed Silver entities and Gold metrics form the framework's layered computation.

Directed Acyclic GraphsGraph theory

Task-based lineage is expressed as a DAG in which each task is the atomic unit of computation.

Data contractsContract-first data engineering

Data Objects carry identity, grain, temporal semantics, quality obligations and a publication contract.

Scientific and clinical disciplines

Semantics and knowledge representationOntology and conceptual modelling

Physical representation, semantic identity and computational interpretation are separated by design.

Privacy engineeringData-protection engineering practice

Selective quarantine and controlled publication are encoded into the computation itself.

Formal and mathematical foundations

Deterministic surrogate keysHashing and identity theory

Identity is computed, not assigned, so that replay reproduces the same entity resolution.

Schema drift and semantic driftConcept-drift research

Drift is treated as a governed, observable condition rather than an operational accident.

Written treatment

• What Is a Data Object?
• Contract-First Data Architecture: Meaning Before Movement

Cited literature

1. Searle, J. R. The Construction of Social Reality (1995).
2. Evans, E. Domain-Driven Design (2003).
3. Kahn, R. E., and Wilensky, R. “A Framework for Distributed Digital Object Services” (1995).
4. Guarino, N., writings on ontology-driven information systems.

Research and institutional partners

Clepio BiotechAcademic spin-off of the University of FlorenceOfficial source

Industrial owner of the scientific data model; the OME-Zarr architecture operationalises a decade of research in light-sheet microscopy, tissue clearing and image analysis conducted between the University of Florence, LENS and CNR.

University of FlorenceAcademic spin-off recognitionOfficial source

The University officially lists Clepio among its academic spin-offs — the formal route through which university-generated knowledge, researchers and intellectual assets were translated into the company whose workflow this architecture governs.

European Laboratory for Non-linear Spectroscopy (LENS)Interdisciplinary research infrastructure, FlorenceOfficial source

Foundational research in advanced optical methods and imaging; the source of the volumetric imaging practice whose stores, scales, chunks and segmentations become governed Data Objects.

Italian National Research Council (CNR)National public research bodyOfficial source

Foundational public-research contribution in optics and imaging within the same decade-long research base.

University of Bern — 3D PATHEIC Transition project with ACMIT GmbH and the research group led by Inti ZlobecOfficial source

International translational partnership in next-generation three-dimensional analysis of biological samples; it connects a university pathology research environment to the spin-off and to the data architecture that preserves and describes the resulting image and segmentation assets.

Standards and specifications

OME-ZarrOpen Microscopy Environment

Multiresolution volumetric image stores are ingested and governed in their native scientific format.

OME-NGFFOpen Microscopy Environment — Next Generation File Format

Node-level metadata, chunk grids and coordinate transformations are interpreted from NGFF metadata.

SHA-256 checksums and manifestsCryptographic integrity standards

Immutable CSV publication packages carry manifests and checksums before Parquet materialisation.

Apache ParquetColumnar storage specification

Final scientific Data Objects are materialised as Parquet datasets.

Scientific and clinical disciplines

Computational biology and biomedical imagingScientific imaging research

Segmentation labels, stitched products, object-level measurements and scientific aggregates are modelled as governed objects.

Cellpose segmentation outputsDeep-learning cell segmentation

Segmented objects, their relationships and aggregates are carried through the platform as first-class Data Objects.

Reproducible researchScientific method

Deterministic reprocessing across multiresolution levels makes scientific claims replayable.

Formal and mathematical foundations

Expected chunk inventoryDiscrete mathematics

N_chunks = ∏ᵢ ceil(shapeᵢ / chunk_shapeᵢ) validates completeness before the binary payload fully arrives.

Reference workloadMeasured platform specification

40 metadata nodes, 18 multidimensional arrays, six scales per scientific product and 41,116 expected chunks.

Written treatment

• What Is a Data Object?
• Determinism Is Not Repetition: The Conditions of Reproducible Data

Cited literature

1. Searle, J. R. The Construction of Social Reality (1995).
2. Evans, E. Domain-Driven Design (2003).
3. Kahn, R. E., and Wilensky, R. “A Framework for Distributed Digital Object Services” (1995).
4. Guarino, N., writings on ontology-driven information systems.

Research and institutional partners

Politecnico di MilanoAcademic origin of Beta 80's founders; current university relationshipOfficial source

Beta 80 was created by students and recent engineering graduates of the Politecnico; the relationship continues through internships and teaching. Emergency-call platforms are exactly the sociotechnical systems in which university engineering formation must become robust operational products.

University of Milano-Bicocca · University of Bergamo · University of Sannio · University of SalernoBeta 80's published university networkOfficial source

Curricular and extracurricular internships, lectures and professional testimony in STEM programmes. The geography mirrors the company's engineering base in Lombardy and its development operations in southern Italy — the talent pipeline behind emergency-data engineering.

Università Cattolica del Sacro CuoreMulti-year applied digital-transformation partnershipOfficial source

An Agile Software Factory supporting the rapid development of university applications, including distance-learning and online administrative services, with advanced observability in a partnership spanning more than a decade. Not a biomedical research collaboration: the university is a complex institutional environment in which application governance, resilience and service continuity are tested at scale — requirements directly relevant to emergency platforms.

NUE 112 legacy corpus as a research assetAcademic reading of the governed historical datasetProject document

Under rigorous privacy, security and public-safety restrictions, the preserved corpus supports algorithm evaluation, semantic research, simulation, operator-training methods and comparative benchmarking.

Beta 80 — emergency-management domainPortfolio context; the technical document does not name the vendorPortfolio context

Contextual attribution supported by Beta 80's documented 112/118 business: Advanced Mobile Location, eCall, Where Are U integration and Next Generation 112 services.

Standards and specifications

NUE 112 emergency-number data modelEuropean single emergency number

Historical emergency-response records are preserved verbatim as immutable Bronze evidence.

Immutable evidence captureArchival and audit theory

No foreign keys, no central event hub and no silent normalisation of source identifiers in Bronze.

Scientific and clinical disciplines

Emergency-response operations researchPublic-safety systems

Calls, locations, organisations and operational narratives are prepared as a corpus for analytics and benchmarking.

Machine-learning corpus preparationApplied ML methodology

Bronze objects are shaped for training, benchmarking and validation without destroying original payloads.

Formal and mathematical foundations

Source-preserving invariantsProvenance research

Evidence precedes interpretation: reconstruction is deferred to a later, explicitly governed layer.

Written treatment

• Separating the Fact from the Local Format
• Append-Only Architectures and the Ethics of Preserving History

Cited literature

1. Kant, I. Critique of Pure Reason (1781/1787).
2. Latour, B. Pandora’s Hope (1999).
3. Hacking, I. Representing and Intervening (1983).
4. Floridi, L. The Philosophy of Information (2011).

Research and institutional partners

Politecnico di Milano and Beta 80's university networkFounding academic origin; internships and STEM teaching across five universitiesOfficial source

Milano-Bicocca, Bergamo, Sannio and Salerno join the Politecnico in a formation network that supplies the software, geospatial and reliability engineering competence behind semantic reconstruction of emergency events.

Università Cattolica del Sacro CuoreLong-term software-factory and observability partnershipOfficial source

Applied experience of application governance and service continuity in a large institutional setting, informing the governance of reconstructed operational entities.

Standards and specifications

Entity resolutionRecord-linkage literature

Source-shaped records are reconstructed into governed operational entities under explicit rules.

Temporal semanticsBitemporal data modelling

Event time, observation time and record time are distinguished rather than collapsed.

Scientific and clinical disciplines

Emergency-response analyticsPublic-safety systems

Reconstructed entities support operational and analytical interpretation of emergency events.

Formal and mathematical foundations

Deterministic reconstructionReproducibility theory

Semantic reconstruction is replayable from immutable Bronze evidence.

Written treatment

• Dismantling the Source, Reconstructing Information
• From Conceptual Lineage to Executable Lineage

Research and institutional partners

Istituti Fisioterapici Ospitalieri (IFO), RomeDual-IRCCS research hospital — Istituto Nazionale Tumori Regina Elena and Istituto Dermatologico San GallicanoPortfolio context

Institutional owner of the programme; the only Italian hospital organisation including two IRCCS institutes, with a technology estate of approximately EUR 85 million and 57 agreements signed or active with institutions and universities. The ETL specification itself does not name IFO — the attribution comes from the project context.

Sapienza University of RomeNursing degree agreement traced to 16 July 1998; DOMINO and NANO-COVID-TEST joint programmesOfficial source

An enduring educational structure that places university teaching inside the hospital, alongside joint translational work on the tumour microenvironment, nanophotonic diagnostics, experimental medicine and machine-learning-supported drug discovery. A governed logistics model therefore also supports case-based education and data literacy, under privacy constraints.

University of Rome Tor VergataProspective multicentre study; laboratory collaborations; HEAL Italia partnerOfficial source

Haematology and stem-cell research, including transcriptional and epigenetic mechanisms of treatment resistance in multiple myeloma — a setting in which drug, device and cost data must be modelled reliably.

Campus Bio-Medico University of RomeMulticentre multiple-myeloma study with IFO and Tor VergataOfficial source

A direct bridge among university haematology units and IFO clinical research; also convention for the tutoring of medical trainees.

University of VeronaORIENTATE study, IFO as promoterOfficial source

Clinical partner for the repositioning of decitabine in advanced pancreatic cancer, with the Verona oncology unit coordinating the clinical work — an example of the trial operations that consumption and budget data must support.

University of MilanIFO-led PNRR projectOfficial source

HPV and host body-fluid biomarkers for the detection of head-and-neck cancer relapse.

University of Bari Aldo MoroJoint venture with the Regina Elena Institute and the CASPUR university-computing consortiumOfficial source

Formed part of the computational-biology framework underpinning IFO's analytical capability.

University of Urbino Carlo BoJoint research with IFO and SapienzaOfficial source

Functional assessment of ATM variants in atypical or mild forms of Ataxia-Telangiectasia.

University “G. d’Annunzio” of Chieti-PescaraPreclinical therapeutic-target researchOfficial source

Allosteric kinesin inhibitors with antitumour efficacy in gastric adenocarcinoma models.

HEAL Italia — national precision-medicine partnershipCoordinated by the University of Palermo; IFO participates in several spokes, including data-intensive and predictive-model componentsConsortium / network

Places IFO inside a national network for precision medicine, advanced laboratory research, integrated clinical-data networks and technology transfer — the environment in which a governed medicines, devices, cost and time model becomes research infrastructure rather than a warehouse exercise.

HEAL Italia co-members: Palermo · Bologna · Sapienza · Tor Vergata · Milano-Bicocca · Modena and Reggio Emilia · Polytechnic University of Marche · Pisa · Cagliari · Catania · Foggia · VeronaOfficial HEAL Italia extended-partnership list; Bologna leads the smart-health/data spokeConsortium / network

Consortium co-membership establishes an academic network around IFO and creates formal opportunities for shared research and data infrastructures. It does not, by itself, evidence a bilateral agreement between IFO and each listed university.

Roma Tre University2025 framework agreement with the Department of Science

Training and orientation internships; biology, chemistry, mathematics, physics, engineering, computer science and scientific communication capabilities relevant to data quality and modelling.

University of Siena2025 curricular training and orientation internship agreement

Pharmacology, pharmacovigilance and pharmacoepidemiology; participation in IFO pharmacology and skin-cancer therapeutic research initiatives.

University of L'Aquila2024 PNRR-supported protocol with IFO–San Gallicano

Omics approaches to psoriasis and atopic dermatitis; the programme supplies contextual evidence on resources, products and services associated with advanced research pathways.

Unitelma Sapienza2024 internship agreement and structured staff education

Digital learning, public administration, legal and management disciplines supporting governance, workforce development and institutional adoption.

Università Campus Bio-Medico di RomaConvention for the tutoring of medical trainees

Integrated medical, engineering and biomedical profile relevant to therapeutic pathways, medical devices and clinical process design.

Università Cattolica del Sacro CuoreWith Fondazione Policlinico Universitario Agostino Gemelli

Oncology, HIV-related observational databases, oncofertility and an Italy–China lung-cancer research initiative announced in July 2026; supports feasibility analysis and real-world evaluation.

LUMSA UniversityEuropean Researchers' Night and science-engagement partnerships

Communication, psychology, social sciences, ethics and organisational studies supporting the human dimension of pharmaceutical governance.

University of Cassino and Southern LazioIFO-linked European research and public-engagement programmes

Engineering, economics, data science, communication and territorial health systems.

University of ParmaEuropean Researchers' Night partnerships; specialist trainees hosted at IFO

Specialist medical education and biomedical research circulating between IFO and extra-regional academic networks.

University of PisaIFO research programmes and the 2024 science-engagement network

Biology, pharmacology, clinical research and evidence synthesis connected with the analytical use of pharmaceutical logistics data.

Ferrara · Naples Federico II · Padua · InsubriaSpecialist-training relationships documented by IFO clinical units

IFO acts as an extra-network training site for highly specialised clinical disciplines.

Charité–Universitätsmedizin Berlin · Thomas Jefferson UniversityInternational academic connections in the IFO public record

Specialist training and collaborative cancer research, consistent with IFO's recognition of relationships with foreign universities.

Accademia di Belle Arti di RomaScience-communication network

Public information design and citizen engagement around institutional research.

Standards and specifications

ATC classificationAnatomical Therapeutic Chemical classification system

Drug dimensions are organised along ATC classification for pharmaceutical-consumption analysis.

Dimensional modellingAnalytical modelling literature

Facts and conformed dimensions model actual, budget and forecast consumption scenarios.

Idempotent MERGE semanticsTransactional data-management theory

Deterministic business-key hashes and idempotent MERGE guarantee repeatable loads.

IRCCS-grade data governanceItalian Scientific Institutes for Research, Hospitalisation and Healthcare

Governed views make data research-ready without bypassing institutional responsibilities.

Scientific and clinical disciplines

Pharmaceutical logistics and health economicsHealthcare operations

Consumption, forecast, drugs, devices and organisations are reconciled per scenario.

Pharmacoeconomics and health-technology assessmentHealth economics research

Cost, quantity, therapeutic class, organisational setting and time support value assessment and resource-allocation studies.

Pharmacoepidemiology and pharmacovigilanceClinical pharmacology

Consumption patterns can be combined, under appropriate governance, with clinical and safety information to investigate use, variation and potential risk signals.

Health-services research and multicentre readinessClinical research methodology

Explicit definitions, provenance and quality controls make participation in multicentre studies feasible.

Oncology and dermatology care pathwaysIRE and ISG clinical missions

Availability, use, cost and traceability of drugs and devices are connected to treatment pathways, trials and safety monitoring.

Formal and mathematical foundations

Cost identityReconciliation arithmetic

Total Cost = Quantity × Unit Cost, reconciled per scenario across 59 field-level rules and 17 transformation categories.

Versioned semanticsTemporal and provenance theory

Mappings, classifications and organisational structures are versioned so historical analyses remain intelligible.

Written treatment

• Data Quality Is a Computational Property
• The Task as the Atomic Unit of Data Computation

Standards and specifications

OMOP Common Data ModelObservational Health Data Sciences and Informatics (OHDSI) community

Principal objects and fields are catalogued with field-level contracts for implementation.

Standard and source conceptsOMOP standardised vocabularies

Vocabulary resolution maps heterogeneous source concepts to standard concepts.

SNOMED-oriented terminologyClinical terminology

Semantic-domain validation is anchored in controlled clinical terminology.

Scientific and clinical disciplines

Observational health data scienceClinical research informatics

The catalogue supports comparable, research-grade representation of clinical events.

Medical informatics interoperabilityHL7 v2, HL7 FHIR, CDA and C-CDA

Source-to-OMOP coverage analysis is designed against established clinical interchange standards.

Formal and mathematical foundations

Structural and semantic validationFormal validation theory

Field contracts drive metadata-driven ETL generation and validation ahead of publication.

Written treatment

• Every Source Speaks a Dialect

Standards and specifications

CSDM and CMDBCommon Service Data Model / configuration management

Service topology and configuration are modelled semantically instead of replicated as ticket tables.

ITIL-aligned service managementIT service-management practice

Work, service, change, risk and service level are modelled as governed Data Objects.

Scientific and clinical disciplines

Organisational knowledge representationKnowledge management

Knowledge and evidence resolution are linked to operational work as a computable structure.

Formal and mathematical foundations

Cross-domain traceabilityLineage theory

Change decisions can be traced through topology to measured service level.

Written treatment

• What Is a Data Object?

Cited literature

1. Searle, J. R. The Construction of Social Reality (1995).
2. Evans, E. Domain-Driven Design (2003).
3. Kahn, R. E., and Wilensky, R. “A Framework for Distributed Digital Object Services” (1995).
4. Guarino, N., writings on ontology-driven information systems.

Research and institutional partners

IRCCS San RaffaeleScientific Institute for Research, Hospitalisation and Healthcare

Research partner within the PR Calabria FESR FSE 2021–2027 programme.

Università Magna GraeciaUniversity research partner

Academic partner within the PR Calabria FESR FSE 2021–2027 programme.

Romolo HospitalClinical partner

Clinical partner contributing to the bio-clinical research platform.

More CareIndustrial partner

Industrial partner within the funded research consortium.

Standards and specifications

GDPR and pseudonymisationEuropean data-protection regulation

Direct patient identity is separated from laboratory sample identity through pseudonymised identifiers.

OpenAPI, REST and Kafka contractsInterface specification standards

Biobanks and external systems integrate through documented API contracts and controlled file exchange.

Genomic formats (FASTQ, VCF)Bioinformatics community standards

NGS molecular ground truth is carried in established sequencing formats.

Scientific and clinical disciplines

Precision medicine and molecular diagnosticsTranslational clinical research

Patient → Clinical Request → Sample → Biobank → NGS and Raman Acquisition → Prediction → Validation.

Raman spectroscopyAnalytical spectroscopy

Acquisition parameters, raw and extracted spectral data are traced through preprocessing and feature extraction.

Machine-learning validation against ground truthApplied ML methodology

Predictions, confidence scores and concordance are validated against NGS molecular ground truth.

Biobank and specimen traceabilityBiobanking research practice

Sample, reagent and preparation provenance is preserved end to end.

Formal and mathematical foundations

Provenance chainScientific provenance theory

Every prediction can be traced back to the specimen and clinical request that produced it.

Written treatment

• Determinism Is Not Repetition: The Conditions of Reproducible Data

Standards and specifications

Enterprise data warehousingAnalytical architecture literature

Governed reporting and analytical models were established across banking operations.

Scientific and clinical disciplines

Quantitative analysis and data scienceStatistical modelling

Analytical and predictive work supported business and risk reporting.

Formal and mathematical foundations

Reconciliation controlsNumerical validation practice

Cross-layer reconciliation underpinned trust in published figures.