Medical Data Analytics

Medical Informatics Platform

A powerful federated data processing and analysis system that preserves patient privacy

Artwork by Gerd Altmann, Pixabay, https://pixabay.com/users/geralt-9301/

The MIP is designed to help clinicians, clinical scientists, and clinical data scientists aiming to adopt advanced analytics for diagnosis and research in clinics.

Users can explore harmonised medical data extracted from pre-processed neuroimaging, neurophysiological and medical records and research cohort datasets without transferring original clinical data.

The MIP links brain science, clinical research and patient care and promotes collaborative research. In a unique and accessible space it provides the tools to improve knowledge, diagnosis, early prediction and innovative treatment of brain diseases.

  • Data Federation - A virtual database that seamlessly presents aggregated data from distributed sources
  • Access and analyse imaging and clinical data, securely stored in hospitals, research archives and public databases
  • Optimised leverage and re-use of decentralised patient data and research cohort datasets, without transferring original data
  • Modelling, integration and data analysis tools for the exploration of harmonised medical data
  • Integrated statistical analysis tools
  • Predictive machine learning algorithm for data exploration

Tools

Access the Public MIP

The MIP PUBLIC platform allows EBRAINS users to explore and analyse variables from fully synthetic datasets (generated via computer simulation). No installation, no downloading required.

What you can do:

  • Experiment on synthetic datasets in epilepsy, dementia, F-Tract, mental health and traumatic brain injury
  • Select datasets, variables and co-variables to define your experimental parameters
  • Create your experiment to perform descriptive statistical analysis for the variables of interest
  • Choose the available algorithms
  • Run and download your experiment (jpeg, png, pdf, svg vector image)
  • Access MIP Training Materials
  • Access the Referential Ontology Hub for Applications within Neurosciences (ROHAN)

Access to the Public MIP requires an EBRAINS account.

Log in (EBRAINS account required)

Ethics and Legal Requirements

Request extended access to the MIP

Get access to the MIP as:

  • a clinician, for confirming objective diagnosis and treatment of a brain disease
  • a clinical neuroscientist or data scientist, for the application and testing of new models and methods
  • a pharmaceutical or biotechnology company, for disease target discovery or to federate your data

The MIP is organised into data federations, where data from multiple locations are harmonised and described by a common data model. There are currently federations for Dementia, Traumatic Brain Injury and Mental Health.

Access to the different federations will be granted on a case-by-case basis, and is only possible after you have signed in on the platform and your user account has been registered. You have several options in terms of admittance:

Participate - request to be admitted becoming partner (and include your data) in an established federation on a specific pathology (currently dementia, epilepsy, traumatic brain injury and mental health)

Lead - request to be admitted becoming partner (and include your data) in an established federation on a specific pathology, define a new data model and lead a new federated analysis with all or some of the partners in the federation

Be the first - start to use the MIP for a new federated analysis on data available in your consortium, on a specific pathology, not yet available in the MIP.

Requested access to one or more federations by sending an email to support@ebrains.eu

Background

How the MIP works

General operation of the Medical Informatics Platform

The MIP is an innovative data processing, analysis and data collection system. It provides an interface for various investigators (clinicians, neuroscientists, epidemiologists, researchers, health managers) to access and analyse aggregated, federated medical data stored in hospital, research centers and public databases, without moving the data from their place of origin, and without infringing patient privacy. Any institution or centre may request MIP installation.

The installation of the MIP explained in three steps:

  1. Installation of a local MIP instance in the participating hospital/lab or set-up of a secured personal virtual machine for the participating hospital/center, enabling harmonisation of local data via pre-processing, data integration and pseudonymisation/anonymisation techniques.

  2. The participating institution signs a MIP Data Sharing Agreement, ensuring data privacy, allowing end-users to only query aggregated findings. Databases cannot be copied, downloaded or uploaded.

  3. The central MIP engine enables federated web-based queries, which execute the same algorithm in each federated node in a coordinated way to provide aggregated findings.

Background

Data Federation

The Medical Informatics Platform allows clinicians and clinical scientists to explore and research harmonised medical data.

Medicine is producing an unprecedented level of valuable medical data around the world, with millions of brain images and terabytes of associated medical data being produced every day. The Medical Informatics Platform (MIP) aims to create a virtual database that aggregates data from various sources, and gives them a common data model. This technology is called data federation.

Data federation provides a single source of data for experts to analyse and advance understanding of neurological and psychiatric diseases. Analyses will lead to real possibilities for early diagnosis and personalised medicine, addressing one of the major healthcare challenges facing the EU.

Most notably - original clinical data is neither moved from the hospital, uploaded to a cloud, nor copied or pasted elsewhere - only aggregated findings, produced by combining information from multiple sources, are analysed in those federated analyses.

MIP FEDERATIONS

The Medical Informatics Platform (MIP) supports the development of disease-oriented federations for hospitals and (medical) research centres that wish to engage in collaborations, share data without transferring them, and perform federated analysis of large-scale distributed clinical datasets.

For each federation, clinical researchers create specific data models based on well-accepted common data elements (CDE), approved by all participating centres. The MIP teams provide support to the centres to facilitate coordination and communication, and assist in the creation of these data models.

Current Federations on the MIP:
MIP Federation in Dementia
MIP Federation in Traumatic Brain Injury
MIP Federation in Mental Health

Background

Analytical Tools

The MIP currently integrates 14 algorithms for data analysis, applicable in subsets to the different pathologies.

Medical conditions - Federations set up in the MIP:

  • Dementia
  • Traumatic brain injury
  • Epilepsy
  • Mental health

Data types accessible in the MIP:

  • Clinical information
  • Regional brain volumes
  • Intracerebral EEG

Federated Analysis Algorithms Descriptions
More information on the available federated analysis algorithms, retrievable through the individual component list, can be directly found on GitHub. This includes documentation on the existing algorithm federation approach as well as information related to creating a new algorithm.

GitHub repository

Community

Join the MIP network

MIP installations across Europe, Map data ©2019 Google

Since the beginning of the Human Brain Project (HBP) in 2013, the MIP has been developed and installed in an increasing number of participating hospitals. The Lausanne University Hospital (CHUV) is the institution coordinating this activity.

The MIP enables medical progress in the field of brain diseases and related medical conditions through unique access to a large volume of patient datasets. It provides methods to analyse federated data from hospitals, research centres, and biobanks, encouraging the development of new scientific ideas. Collaborators can share and release results of their research with others via the platform.

What is the MIP Network?

The MIP network links a growing number of hospitals and research centers in 16 countries across the world. The institutions either installed the MIP or are lined up for the deployment process.

  • CHUV / Switzerland
  • Brescia Hospital / Italy
  • Plovdiv Hospital / Bulgaria
  • CHRU Lille / France
  • Niguarda Hospital / Italy
  • Freiburg Hospital / Germany
  • Institute Mario Negri / Italy
  • IRCSS Neurological Institute Carlo Besta / Italy
  • IRCSS Fondazione Istituto Neurologico Nazionale Casimiro Mondino / Italy
  • St. Anne’s hospital / Czech Republic
  • Motol university hospital / Czech Republic
  • Danish epilepsy Filadelfia/ Denmark
  • Hospital del Mar / Spain
  • Sahlgrenska University Hospital /Sweden
  • Grenoble Hospital / France
  • IRCSS Don Carlo Gnocchi / Italy
  • UKAachen / Germany
  • IRCSS Ospedale San Camillo / Italy
  • Max Plank Psychiatric Institute / Germany
  • Karolinska Institute - INCF (International Neuroinformatics Coordinating Facility) / Sweden
  • IRCCS Institute of Neurological Sciences / Italy
  • Paracelsus Medical University / Austria
  • Centro Hospitalar do Porto / Portugal
  • Moscow Research and Clinical Center for Neurospychiatry / Russia
  • CHU Liège / Belgium
  • Novo Nordisk / Denmark
  • Hospital de Santa Maria - Centro de Referencia para a Area de Epilepsias Refractarias / Portugal
  • Neurospin / France
  • Rigshospitalet / Denmark
  • Tel-Aviv Medical Center (TLVMC) / Israel

Key benefits to expect from joining the MIP network:

  • Participate or lead Federated Analysis on large volumes of decentralised clinical data available in the network of MIP-Federated Nodes
  • Develop new scientific collaborations
  • Train and use novel state-of-the-art analytical tools, including machine learning algorithms and the MIP PUBLIC
  • Investigate and discover novel findings using the MIP with your own data
  • Increase the chance of future successful national or European grant applications being part of the MIP network

Request Access to the MIP Network:

For more information on the MIP Community and to request access, contact sandra.schweighauser@chuv.ch

Access all EBRAINS services