KDI Project for optimizing patient therapy

Scientists started a project for collecting some clinical data.

Type of content: Assets
Type of asset:
Use case
Big data potential
Policy domains: Health
Phase in the policy cycle:
Policy Design and Analysis
Open license availability
Tags: Big data
SWOT Analysis for
KDI Project for optimizing patient therapy
Helpful Harmful
Strengths• Patient data are evaluated holistically: structured data (e.g. laboratory values), unstructured data (e.g. free-text findings), images and OMICS data (e.g. SNPs and gene expression data) are merged to give each patient a holistic picture of all data traces.
• Identifying and evaluating at first inconspicuous dependencies across departmental boundaries.
• Research Database: An important foundation for data intelligence / big data solutions in the healthcare sector is a research database for the storage and management of patient-specific data (both from the clinical process as well as from molecular biological and genetic analyzes)
• Data Analytics: consists of extensions to the approaches of machine learning in semantic networks that have been developed in THESEUS 'Core Technology Cluster. They represent the state-of-the-art and are based on the mathematics of matrix and tensor factorization. The approaches have proven to be particularly effective for the high-dimensional sparse data in clinical data.
• Platform for medical apps: develop concepts for app implementation in order to support innovative usage and business models.
Weaknesses• Low ease of use
• Large volume of data needed
• Web page in German only (English version not working properly
• Need to properly represent and process the security / insecurity of information sources
• Training personnel needed
• Data security
• Patients’ Privacy: High security requirements in terms of data security and patient privacy
Opportunities• Data intelligence: solutions are developed and validated directly from a typically large data set: data reflect the complexity of reality with all its nuances and developed solutions found by the direct means of validation clinical acceptance.
• In order to provide a synopsis of the data sources for everyday medical practice but also for subsequent (Remind-) projects or methods of artificial intelligence, is a systematic analysis of the data and diversity of a concept for the ontologically guided work-up and utilization of data aim of this project.
• Improving patient care
• Detect deviations from standard and the reasons for.
• An important new aspect is the modelling of temporal information and a modelling of the sequential processes in the clinic. With very large numbers of patients and patient-specific data, growing training times may require scalable distributed computing software, so we will explore implementations of our approach in the Hadoop framework.
Threats• Diversity: in all fields of medicine very many different databases, in the context of digitization, were built. The diversity currently makes an integrated visualization or even processing with underlying common ontologies or ordering hierarchies impossible.
• Low ease of use
• Data quality: need to properly represent and process the security / insecurity of information sources
• Training personnel.
• Data security, Patients’ Privacy: High security requirements in terms of data security and patient privacy.

Open data - Download the Knowledge base

You are free to download the data of this Knowledge base.

To do this you must be an authenticated user: log in or sign in now.

All the data are licensed as Creative Common CC-BY 4.0.