Information For Patients
The Medical Informatics Platform
Medicine is producing an unprecedented level of valuable medical data at a staggering pace 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 analyze and advance understanding of neurological and psychiatric diseases. Analyses will lead to real possibilities for early diagnosis and personalized medicine, addressing one of the major healthcare challenges facing the EU. Data isn't moved from the hospital, uploaded to the cloud, or copied and pasted elsewhere, and only anonymous data is shared.
The MIP provides methods to analyze federated data from hospitals, research centers, and biobanks. Clinical scientists can develop, share, and release the results of their research. The MIP will bring people across professional and scientific fields together, encouraging them to actively contribute to the design and development of the services that the MIP provides. Representing the fusion of medicine and computer science, we aim to break down the traditional barriers of patient care, brain science, and clinical research to minimize the delays in diagnosis of brain diseases and implement the most effective treatments. Data security is a top priority for us. The MIP design will preserve hospital ownership and control of data by providing federated access to analysis tools across hospitals, without moving patient data from original servers and ensuring patient privacy.
WHO USES THE MIP?
We provide online evidence-based medicine tools via a web portal to users such as neuroscientists, computational scientists, epidemiologists, and the pharmaceutical industry. Its analyses integrate open-access research data repositories with brain disease features generated in participating hospitals using MIP tools.
Hospitals and research centers that have installed the MIP:
• 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
As in many IT contexts, data security is treated as a top priority, the need for which is made even more pressing due to the long-standing commitment of the medical profession to patient confidentiality. We aim to preserve hospital ownership and control of data by developing a federated query engine within the hospitals, leaving patient data in its original location and format. This is a fundamental difference compared to traditional schemes in which data is moved to accommodate the needs of the query engine. The research team is also developing techniques to ensure that it will not be possible to infer personal information about patients from query results while performing advance machine learning analytics.
The MIP is committed to respecting the Data Policy Manual (DPM) produced by the Data Governance Working Group of the HBP and approved by The Directorate (DIR) and the Science Infrastructure Board (SIB).
- Integrating clinical data from hospital databases
- Redescription mining augmented with random forest of multi-target predictive clustering trees
- Extending Redescription Mining to Multiple Views
- Ensembles for multi-target regression with random output selections
- Enabling Data Integration Using MIPMap
- Feature Ranking for Multi-target Regression with Tree Ensemble Methods
- Semantic Annotation of Data on Neurodegenerative Diseases in Patients using Ontologies
- Regional volumetric change in Parkinson's disease with cognitive decline
- Multi-label classification via multi-target regression on data streams
- Tree-based methods for online multi-target regression
- Data-driven identification of prognostic tumor subpopulations using spatially mapped t-SNE of mass spectrometry imaging data
- Comparison of Tree-Based Methods for Multi-target Regression on Data Streams
- Multiple Linear Regression: Bayesian Inference for Distributed and Big Data in the Medical Informatics Platform of the Human Brain Project
- Option Predictive Clustering Trees for Multi-target Regression
- Homogeneous clusters of Alzheimer's disease patient population
- Semi-supervised classification trees
- Option Predictive Clustering Trees for Hierarchical Multi-label Classification
- HINMINE: heterogeneous information network mining with information retrieval heuristics
- Neurobiological origin of spurious brain morphological changes: A quantitative MRI study
- A framework for redescription set construction
- Predictive Clustering of Multi-dimensional Time Series Applied to Forest Growing Stock Data for Different Tree Sizes
SUPPORT FROM PATIENT ASSOCIATIONS