National mandate · Israel Innovation Authority
Turning medical data
into integrative intelligence.
HCIMI, the Hadassah Center for Integrative Medical Intelligence, is a national consortium unifying clinical, genomic, and community data.
Enabling personalized medicine, agentic clinical AI, and accelerating discovery at Hadassah Medical Center.
Integrative Intelligence Personalized Medicine Discovery
Explore the platform →
20+Years longitudinal data
1.4MCommunity patients
33Integrated systems
2National IIA mandates
The Platform Architecture
Three layers. One intelligent platform.
From raw clinical data to agentic AI decisions. Every layer built to enable personalized medicine and accelerate discovery.
Data Foundation LAYER 1 — FOUNDATION Data Integration 33 Systems · NLP · AWS Lakehouse Imaging · Genomics · Organoids Synthetic · 20+ yrs Wide · Historical · Longitudinal · Multimodal The substrate for personalized medicine Modeling Engine LAYER 2 — INTELLIGENCE Modeling Engine Mechanistic Bio · Physics ODE · Diffusion Biological priors Grounded in biology Embedding Vector spaces Multimodal Clinical patterns Learned patterns Hybrid Mech + Embed Fused models Dual reasoning Best of both Mechanistic · Embedding · Hybrid in service of personalized medicine Cognitive AI Layer LAYER 3 — COGNITIVE AI & Discovery Agentic AI LLMs Knowledge Graphs Semantic Search Discovery Engine Clinical Decision Personalized Med. Explainable AI Real-time Inference INTELLIGENCE IN SERVICE OF The clinician at the point of care. The scientist seeking discovery. The pharma partner accelerating trials. The patient who deserves their own equation. Data flow through layers HCIMI · Hadassah Center for Integrative Medical Intelligence
A new kind of entity: where clinical depth and computational power are equals.
HCIMI was founded on a premise that is simple but consequential: the most important medical breakthroughs of the next decade will emerge from the intelligent synthesis of clinical knowledge and computational power, two disciplines brought together as full equals for the first time.
"The most consequential medical insights of the next decade will be computational. We are building the infrastructure to produce them."
Mandated by the Israel Innovation Authority and anchored at Hadassah Medical Center, we bring together the nation's foremost oncologists, neurologists, pathologists, and data scientists as absolute peers: clinicians and data scientists advancing together as co-founders of a new scientific discipline, from the first day.
Operated in partnership with Hadasit, Hadassah's technology transfer and innovation company, and funded through the Israel Innovation Authority's national programme.
Our mandate is national. Our infrastructure is unparalleled in Israel. Our ambition: to establish Hadassah as a global node in Integrative Medical Intelligence, a peer of the world's leading biomedical research consortia.
We build the infrastructure and the science that makes it possible to treat this patient, find this biomarker, and make this discovery: at the intersection of data, biology, and artificial intelligence.
This is what we mean by Integrative Medical Intelligence. A new scientific discipline.
01
Integrative Intelligence
Unifying clinical, genomic, imaging, and community data into a single coherent, AI-ready platform: the connective tissue that makes everything else possible.
02
Personalized Medicine
Using that infrastructure to treat each patient as a unique biological equation. Mechanistic, embedding, and hybrid models that reason about this patient, at this moment in their disease.
03
Discovery
Generating new scientific knowledge from the platform: biomarkers, disease subtypes, and treatment response patterns that emerge only at the intersection of clinical depth and computational scale.
National standing
National grant · IIA Track 10
Integrative Medical Intelligence: National Platform for Clinical AI
Hadassah · Hadasit · Meuhedet · AWS · NVIDIA
IIA Infrastructure Grant · Year 3
Secure Virtual Research Environment with NLP Enrichment and Synthetic Data Trust Layer
Hadassah · MDClone · AWS · Cross-national validation with Sheba Medical Center
People & Computers · 2023–2025
IT Innovation Award, three consecutive years
Iron Swords Emergency C&C (2023) · Real-Time Hospital Management Systems (2024) · Oncology Data Mart & cBioPortal + Real-Time C-Section Prediction System (2025)
From unified data
to living patient models.
The term borrows from Apollo 13, where NASA engineers used a continuously updated virtual model of the spacecraft to bring a stranded crew home.
Formal definition · National Academies of Sciences, Engineering and Medicine · 2024
A Digital Twin in medicine is a personalised, continuously updated computational representation of a specific patient, capable of simulating counterfactual treatment decisions in real time. A new class of medical intelligence, one that requires exactly the kind of longitudinal, multimodal, wide data infrastructure that HCIMI is building.
Three conditions have now matured simultaneously to make this possible: data at longitudinal depth, compute at unprecedented scale, and models: mechanistic, embedding, and hybrid, designed to reason and simulate. HCIMI is building the infrastructure and the science that sit at this convergence.
All four foundational methodologies are under active development today, across oncology, IBD, and neurology.
01 · Vectorisation
The patient journey as mathematics
Mapping decades of multimodal patient history into high-dimensional embedding space where clinical similarity becomes computable distance. The basis for personalized reasoning at scale.
02 · Hybrid Modelling
Mechanistic + embedding models
Physics-informed biological models combined with learned embeddings. Mechanistic models generalise from established biological priors. Embedding models extend beyond those priors into patterns the data reveals. Hybrid models bring both together.
03 · Simulation
Counterfactual clinical decisions
What if treatment A instead of B? What if the dose were adjusted at week 3? The model simulates. The decision-maker acts. The loop closes in operational use: in clinical practice and pharmaceutical development alike.
04 · Scalability
Built for chronic disease at population scale
Designed for longitudinal, chronic conditions where the value of a living model compounds over time: IBD flares, neurological progression, oncology trajectories across years of follow-up.
Invited Lecture · ILAMI 2026
Digital Twins in Personalized Medicine
Dr. Ruth Lev Bar Or · Dorit Tzur
Dr. Dorit Shemesh · Prof. Shai Rosenberg
Watch the lecture English narration
ILAMI 2026 · Digital Twins in Medicine
Wide. Historical.
Longitudinal. Multimodal.
The substrate for computational patient models and personalized clinical AI is data of rare depth and breadth: longitudinal across decades, integrative across institutions, computable across every modality. We have spent years building exactly this. We believe it has no parallel in Israel.
Tertiary clinical data
20+ Years Hadassah Records
Complete longitudinal records across oncology, neurology, gastroenterology, and surgery: structured and unstructured, harmonised across 33 source systems into a unified AWS Lakehouse with NLP enrichment.
Community continuum
1.4M Meuhedet Patients
The hospital-community gap, closed. Primary care, specialist referrals, chronic disease management: the full patient journey from community to tertiary care and back, across decades.
Privacy & federation
Synthetic Data Trust Layer
MDClone synthetic data generation validated against real-world ground truth via the TSTR methodology: Train Synthetic, Test Real. Cross-institutional research with patient data remaining in place. National validation standard under development with Sheba Medical Center.
Biology
Genomics + Organoid Biobank
Foundation One genomic sequencing linked to clinical outcomes: somatic variants, germline profiles, and mutational landscapes. Patient-derived organoid models bridging computational prediction and biological validation at the molecular level.
360° Patient 360° Research Environment Hospital Tertiary · 20+ yr Community 1.4M · longitudinal AWS RES · VRE Virtual Research Rooms SageMaker · Bedrock Mechanistic · Embedding · Hybrid MDClone · TSTR Train synthetic → Test real · Sheba
Three clinical domains.
One platform.
Applied across three disease domains: longitudinal depth turned into personalized insight, and new knowledge.
Oncology Gastroenterology Neurology
Predicting chemotherapy resistance in colorectal cancer
Predicting IBD flares before they occur
Living models of brain tumour progression
Fusing whole-slide image pathomics with CT radiomics into a Deep Microenvironment Score that predicts resistance to adjuvant chemotherapy in Stage II to III colorectal cancer, before treatment begins. The aim: non-genetic biomarkers not visible on standard review.
A longitudinal multimodal model, combining structured clinical data, free-text NLP, and patient-reported outcomes, predicts inflammatory bowel disease flares weeks in advance. It enables pre-emptive treatment adjustment and reduces hospitalisation across the hospital-community care continuum.
Mechanistic-neural hybrid models of glioblastoma growth, parameterised from serial MRI, enabling personalised simulation of radiation response and surgical planning. Counterfactual treatment comparison before the first incision. Every patient deserves their own equation.
Clinical Lead: Prof. Shai Rosenberg · Prof. Aron Popovtzer
Pathology: Dr. Tzahi Neuman · Dr. Nir Pillar
Embedding model: Deep Microenvironment Score (DMS) · Multimodal fusion
Clinical Lead: Gastroenterology and Liver Diseases, Hadassah
Data: hospital and community continuum · Hybrid model: mechanistic + temporal embeddings
Clinical Lead: Prof. Shai Rosenberg
Framework: Reaction-Diffusion equations + Neural ODE · Serial MRI parameterisation
A national consortium.
An international ambition.
HCIMI operates at the centre of an ecosystem: each partner chosen because they complete a capability that no single institution can provide alone.
Elements of the ecosystem
01
TERTIARY
Hd
Hadassah
20+ yr · 33 systems
02
COMMUNITY
Mu
Meuhedet
1.4M · longitudinal
03
INNOVATION
Hs
Hadasit
Tech transfer
INTEGRATIVE
Hc
HCIMI
National platform
04
CLOUD
Aw
AWS
Lakehouse · Bedrock
05
COMPUTE
Nv
NVIDIA
GPU · NIM
06
PRIVACY
Mc
MDClone
Synthetic · TSTR
07
VALIDATION
Sh
Sheba
National
Clinical consortium
Meuhedet
1.4M community patients. The primary care continuum that closes the hospital-community gap and makes longitudinal patient models possible at population scale.
Cloud · AI infrastructure
Amazon Web Services
AWS Research and Engineering Studio (RES), S3 Lakehouse, Bedrock. The compute and storage infrastructure that makes population-scale modelling achievable.
AI acceleration
NVIDIA
NIM microservices, TensorRT-LLM, RTX 6000 Ada NLP server. The GPU layer enabling real-time clinical inference from multimodal data.
Privacy · Synthesis
MDClone
Synthetic data generation and federated query across 33 integrated source systems. Cross-institutional research without moving patient data.
National validation
Sheba Medical Center
Independent validation of synthetic data generalisability across diverse patient populations, establishing a national standard.
Innovation engine
Hadasit
Hadassah's technology transfer company, and the legal vehicle behind the consortium: the bridge between academic discovery and commercial translation.
Clinical and computational.
Co-equal. Co-founders.
HCIMI is built on a structural principle rare in academic medicine: clinical and data/AI leadership presented with identical weight, co-architects of a new discipline together.
Clinical leadership
Prof. Shai Rosenberg
Senior Neuro-Oncologist · Head of AI in Oncology
Head, Cancer Computational Biology Lab
Prof. Aron Popovtzer
Director, Sharett Institute of Oncology
Head and Neck Oncology · Radiotherapy
Built and leads Hadassah's Cancer Computational Biology Lab and AI in Oncology Unit, developing MESiCA and TP53_PROF, machine-learning models now used to interpret mutational signatures and predict the clinical impact of cancer mutations. Trained at the Technion under Karl Skorecki, Wolf Prize for excellence, and directs the Faculty of Medicine's dual MD/Computer Science programme.
Directs the Sharett Institute of Oncology, Israel's first comprehensive cancer centre, named by Newsweek among the world's leading cancer institutes. Co-author of KEYNOTE-048 (The Lancet, 2019) and other pembrolizumab trials in NEJM and The Lancet Oncology that changed the standard of care for head and neck cancer. Trained at the University of Michigan, Beth Israel Hospital, and Memorial Sloan Kettering.
Prof. Myriam Grunewald
Director of Research, Hebrew University
Head, Hadassah Organoid Center
Dr. Tzahi Neuman
Chairman, Diagnostic Pathology
Secretary, Israel Association of Pathologists
Founded and directs Israel's first living organoid biobank, recently extended to organoid models of inflammatory bowel disease. Her lab's papers in Cell (2006) and Science (2021) established mechanisms now foundational to the study of blood vessel formation and vascular aging.
Chairs Hadassah's Department of Diagnostic Pathology and leads its FoundationOne molecular testing programme, one of a handful worldwide performing this level of tumour-mutation profiling. Recipient, Israel Ministry of Health prize for personalized medicine; advises industry on diagnostics and biotech.
Dr. Nir Pillar
MD/PhD · Senior Pathologist
Head, Computational Pathology Lab
Leads Hadassah's Computational Pathology Lab, working with UCLA's Aydogan Ozcan lab on the trust problems now central to clinical AI: in Nature Biomedical Engineering, one of the first frameworks for detecting AI hallucination in digital pathology, and in Nature Communications, virtual tissue staining that replaces chemical processing. A decade-long focus on bringing computational methods into daily clinical practice.
Data & AI leadership Visit the Data & AI Unit →
Dorit Tzur
Head of Data & AI · Chief Technology Officer
Dr. Ruth Lev Bar Or
Vice Head of Data & AI
Scientific Modeling Lead
Directed the IDF Medical Corps' BI and Medical Data Analysis Unit for over 20 years before joining Hadassah in 2023. Co-author of over 100 peer-reviewed publications drawn from national adolescent health cohorts, including several in NEJM and JAMA. Member, Hadassah's Helsinki and Ethics Committees. Co-recipient, with Ruth Lev Bar Or and Dorit Shemesh, of the 2025 People & Computers IT Award for the oncologic DataMart.
An active research record spanning three decades: with Haim Sompolinsky (PNAS, 1995), one of the founding papers in continuous attractor network theory, and with Uri Alon and Moshe Oren (PNAS, 2000), a landmark early model in systems biology, predicting p53-Mdm2 oscillations after DNA damage later confirmed by single-cell imaging. More recently, co-author of 2025 studies on kidney function and nationwide melanoma screening. PhD in Applied Mathematics, Weizmann Institute.
Andrey Tass
Senior DBA & Data Engineer · Big Data Architect
Dr. Dorit Shemesh
Senior Cancer and AI Researcher · PhD Quantum Chemistry
Architects the platform's Read-Scale data infrastructure and its Spark/Python engineering spine, following a decade doing similar work across Israel's Ministry of Health, Clalit, and Hadassah. Works across Azure, AWS, Cloudera, and the Spark, Python, Nifi, and Airflow toolchain, and ensures compliance with national technology standards.
AI scientist leading data infrastructure and applied AI for precision oncology at Hadassah: architecture of a 275,000-patient clinical database, AI/NLP pipelines for clinical data extraction, and predictive models for treatment response and disease progression. Data scientist for the international Consortium for Liquid Biopsy, analysing multi-omics blood-sample data across a pan-cancer cohort. PhD in quantum chemistry.
Sara Harris
Data Warehouse Engineer
Two decades as an applicative database administrator in Israel's public sector, specialising in complex query development and performance optimisation. Joined Hadassah's Data & Analytics division and, within a year, built the data warehouse now spanning roughly 1,000 tables across the hospital's systems; supports its data pipelines at the technical, logical, and application layers. Two-time recipient, Data & AI Excellence award.
Research & Innovation
Prof. Eyal Mishani
Director, Research Fund of the Hadassah
Medical Organization · Board Member, Hadasit
Masha Zavin
Industry & Data Collaborations Lead
Hadasit
Built Israel's first PET/Cyclotron facility and, in 2022, registered LuNET-SRY, the first locally-manufactured radiopharmaceutical in Israel. Directs the Hadassah Medical Organization's Research Fund: over 300 research staff and a multi-million-dollar project portfolio.
Leads Hadasit's industry partnerships across pre-clinical trials, prospective and retrospective studies, EMR integration, and regulatory support, including international accelerator programmes with IBM, drawing on her own background in law alongside an MBA in biomed. For HCIMI, will coordinate data-sharing agreements between the hospital and commercial partners.
We are open
to the right partners.
HCIMI partners on shared scientific and clinical stakes: a capability, a question, or a commitment that advances the science of Integrative Medical Intelligence, and the practice of personalized, discovery-driven medicine. If that is you, we want to hear from you.
Dorit Tzur
Chief Technology Officer, HCIMI · Hadassah Medical Center doritzur@hadassah.org.il
General inquiries: contact@hcimi.hadassah.org
Academic research
Joint research programmes with universities, institutes, and clinical centres who share our scientific agenda in Integrative Medical Intelligence and personalized medicine.
Pharma & biotech
External control arms, real-world evidence cohorts, and custom AI model development for clinical trial acceleration and drug discovery.
Already in active discussions with leading global pharmaceutical and biotech companies.
Technology
Companies with AI, compute, data, or biological capabilities that extend the platform's reach or deepen its scientific capacity.
Philanthropy
Strategic donors who want to fund the infrastructure of a new discipline: a sustained research and discovery capability.
Digital Twins in Medicine  ·  HCIMI  ·  ILAMI Invited Lecture
1 / 14
Lecture slide
0:00 4:56