AIDA Data Hub
Data Science Platform
A home for your research.
AIDA Data Hub
Services for Clinical Innovation in Data Driven Precision Health
National e-infrastructure supporting the Analytic Imaging Diagnostic Arena (AIDA)
Hosted by LiU and the Center for Medical Image Science and Visualization (CMIV)�Part of SciLifeLab Bioinformatics platform (NBIS)
250319 Welcome to the Launch of the AIDA Data Hub DSP
AIDA & AIDA Data Hub
AIDA Community - medtech4health.se/aida
National collaboration arena for AI research and innovation in medical imaging diagnostics.
AIDA Data Hub - datahub.aida.scilifelab.se
E-infrastructure set up to support AIDA.
AIDA Data Hub
E-infrastructure for research and clinical innovation in data driven precision health.
Data services
Data Science Platform for Sensitive data
Support
AIDA Data Hub
Started in 2018, as part of AIDA.
Support in sharing medical imaging data.
Policy support in sensitive data for research.
Support to ethical review applications.
Agreement support.
Template agreements for data sharing and transfer.
2018
Data Sharing Policy
Comprehensive resource describing best practices in handling and sharing medical imaging data for research in Sweden and similar countries.
Concrete guidelines and examples, with references to original sources in law.
Developed multi-professionally with AIDA network and stakeholders.
Key insights have been published in Nature Scientific Data.
2019
Using Clinical Imaging Data for Research
Common practice in Sweden and similar countries, 1-paragraph summary:
�The common practice is that caregivers disclose data to research institutions for specific activities described in approved ethical review applications, to be carried out under appropriate technical and organizational protective measures and supervised by a named competent researcher. The research institution is then data controller and copyright holder for the disclosed data, and is responsible for ensuring that data is processed and shared only as described in the approved ethical review application, with data processing agreements, pseudonymization, anonymization and licensing as tools, and with an obligation to store relevant data for 10 years after last use for purposes of research validation.
Data In
Datasets | Scans | Annotations | Size | |
Total | 50 | 154917 | 48787 | 56.95TB |
15 | 6590 | 48387 | 2.15TB | |
11 | 13240 | 34186 | 10.86TB | |
39 | 141677 | 14601 | 44.47TB | |
| | | | |
2 | 106448 | 1006448 | 124.32GB | |
Data Out
Tryggve�Nordic collaboration on �Sensitive personal data for research
Joel Hedlund, Executive manager
Joel Hedlund
NeIC and Nordic ELIXIR nodes.
Collaboration between Nordic platforms for research using sensitive personal biomedical data.
Tryggve 1: Processing platform development.
Tryggve 2: Sensitive data archiving - Federated EGA.
Joel Hedlund
Nordic Register Genomics in Psychiatry �- A Tryggve2 research study��
Lu Yi, PhD. lu.yi@ki.se
Patrick Sullivan, Prof.
Psychiatric Genomics Institute,
Karolinska Institutet
Schizophrenia Basics
Website & Contact: neic.no/tryggve/
Slides: goo.gl/ru9b4y
Schizophrenia Genetics
A major clue, from generations of past work.
Probabilistic not deterministic:
No convincing single gene causes.
Website & Contact: neic.no/tryggve/
Slides: goo.gl/ru9b4y
Nordic registers
Website & Contact: neic.no/tryggve/
Slides: goo.gl/ru9b4y
Study Numbers
Descriptor | Denmark | Norway | Sweden | TOTAL |
Vital statistics: Q4 2017 | | | | |
– total population | 5,781,190 | 5,295,619 | 10,120,242 | 21,197,051 |
– births | 61,397 | 56,633 | 115,416 | 233,446 |
– foreign born (%) | 0.085 | 0.141 | 0.185 | 0.137 |
Register analyses | | | | |
– lifetime Schizophrenia (SCZ) | 36,676 | 9,002 | 29,072 | 74,750 |
– lifetime Major depression (MD) | 75,771 | 87,540 | 595,743 | 683,283 |
– lifetime Postpartum depression (PPD) | 50,176 | 8,572 | 93,960 | 152,708 |
– lifetime Eating disorders in females (ED) | 21,816 | 4,857 | 34,238 | 60,911 |
Microarray data: Q2 2018 | | | | |
– Ns with GWAS | 89,273 | 2,850 | 183,966 | 276,089 |
– SCZ cases | 5,247 | 800 | 4,924 | 10,971 |
– MD cases | 25,431 | 0 | 5,059 | 30,490 |
– PPD cases | 1,600 | 0 | 1,381 | 2,981 |
– ED cases | 5,114 | 0 | 4,118 | 9,232 |
Microarray data: Q4 2021 | | | | |
– Ns with GWAS | 425,000 | 386,000 | 300,000 | 1,111,000 |
– SCZ cases | 9,622 | 2,240 | 12,000 | 23,862 |
– MD cases | 45,701 | 11,750 | 10,000 | 67,451 |
– PPD cases | 3,600 | 1,000 | 2,881 | 7,481 |
Website & Contact: neic.no/tryggve/
Slides: goo.gl/ru9b4y
Tryggve2
A federated system that enables data sharing and analysis in a secure, streamlined & intelligent way
#2-7
Distributed compute solution via singularity container
Website & Contact: neic.no/tryggve/
Slides: goo.gl/ru9b4y
AIDA Data Hub�Sensitive Data Services
First GPU system in Sweden for research using sensitive personal biomedical data.
Set up at CMIV, in collaboration with Nvidia, serving researchers in the AIDA community.
Hosted in Region Östergötland secure data centers hosting for hospital electronic health record production systems.
Hosting VINNOVA funded SCAPIS data lab, where AI researchers can securely process SCAPIS data for research.
AIDA DGX-2 Service
Service for best-in-class researchers in �Swedish medical imaging diagnostic AI.�Secure enough for medical personal data.
2020
AIDA Data Hub�Data Science Platform
Secure data science platform co-located with national/European flagship compute systems.
Supporting advanced data usage patterns:�long term primary storage, collaborate, annotate, share, federate, train AI...
Customers make security decisions as appropriate; outgoing connections to home institution servers, collaborators...
User fees for sustainable operations and development. Discounts to incentivize data sharing and maximize high impact research.
Based on Bigpicture/GDI technologies.
2025
AIDA Data Hub contribution to Arrhenius sensitive data
NAISS Arrhenius: next flagship system for academic computing in Sweden, part of EuroHPC.
AIDA Data Hub has contributed to the sensitive data capability procurement group, with experience from systems for sensitive data and AI innovation in medical imaging diagnostics.
10% of budget to sensitive data support.
AI Factories for Life Science
2+ BEUR investment.
SE system Mimer to be operated by NAISS, in collaboration with RISE.
To be built in the same computer rooms as AIDA Data Hub DSP, and offer similar functionalities.
Recruitment of 52 FTE ongoing, to be in place in 2025h1.
Bigpicture Petabyte platform for European digital pathology AI
AIDA Data Hub leading repository infrastructure development, which is carried out in� collaboration with sensitive data teams at the NBIS Systems Development unit and CSC.fi.
Establishing Bigpicture Federated node on AIDA Data Hub Data Science Platform.
�
EUCAIM European Federation for Cancer Images
AIDA Data Hub Data Science Platform to support EUCAIM use cases. �� Collaboration with sensitive data teams at the NBIS Systems Development unit.
ASHA - Använda Standardiserade Hälsodata som Accelerator
RÖ led VINNOVA Systems demonstrator for Data lake systems for primary and secondary use.�AIDA Data Hub Data Science Platform provides spaces for secondary use.
Thank you!
AIDA Data Hub�Services for Clinical Innovation in Data Driven Precision Health
National data infrastructure supporting the Analytic Imaging Diagnostic Arena (AIDA)
Hosted by LiU and the Center for Medical Image Science and Visualization (CMIV)�Part of SciLifeLab Bioinformatics platform (NBIS)