Knowledge management

Knowledge engineering expertise for innovative data visualization, exploration, confirmation, learning, prediction, recommendation and decision support



HPXWeb, is our software for annotating data with complex metadata. We assist clients in designing and reusing relevant ontologies for their health data initiatives with the goal of increasing their Health Data Equity. We help clients identify and apply appropriate data analysis to generate conclusions, including predictions and outcome recommendations for action. HPXWeb Knowledge Management is used by businesses seeking to increase their health data value by adding metadata and using it to perform data analysis.

  • Data
  • conclusions, predictions &

Data Quality

At Hilbert Paradox, we emphasize the data quality before focusing on quantity. Valuable data allows you to fully exploit the data potential. In our Data Equity solution, Hilbert Paradox assists you in discovering the full potential of your data to maximize its equity.


Case Example

  1. A medical device connects through a wireless connection with your computer or phone.
  2. The medical data will be safely uploaded to the HPX Cloud through a Web API which includes storage, retrieval, and processing efficiencies.
  3. Using the HPX Knowledge Engineering Modules Explore, Explain, Learn and Inform, Hilbert Paradox’ neural network can perform a thorough analysis of your medical data to discover patterns and correlations that lead to better decision making and increased health data equity.


Our first step to increase your data equity in knowledge engineering. We visualize your data in charts, graphs and tables. After a first exploratory analysis is performed, notification management is enabled, and the creation and delivery of clear and comprehensive reports.


In this second step, custom metadata is added to make ontologies. These are connected to established glossaries, terminologies, such as from WHO, CDISC, FHIR, HL7, etc.


After adding metadata and ontologies, correlations can be identified by observing the data trends. In this step, hypotheses are tested and reports can be produced from the identified trends and correlations.


The ultimate step enables recommendations. Based on expert knowledge in scientific research, industry standards and best practices, rules can be added to produce recommendations from the data, and users can be notified of such recommendations.

  1. Through a user interface, the results are displayed to the data end user, usually the clinician who can take action and start appropriate treatment if deemed necessary.


While many companies and research institutions focus on only one aspect or discipline of Health Data, Hilbert Paradox recognizes that by correlating different types of data such as ECG, EEG, environmental factors, glucose levels, Whole Exome/Genome Sequencing, etc. more connections can be created and data equity can grow exponentially.

Taking a more holistic and integrated approach is important in the current health situation. The population is aging and a greater percentage must deal with chronic disease. To analyze and find treatment, it is important to have a variety of broad data so patient-specific treatment can be determined. One can both discover patterns in a patient’s health evolution during the treatment and recognize more general population health trends when aggregating anonymized data. Often organizations are focused solely on one type of data and deliver a lone piece of the puzzle, but Hilbert Paradox works together with its partners to complete the full picture.

Hilbert Paradox is currently running a clinical pilot study to verify the feasibility of a broad variety of individual health data collection. The study has been approved by the Ethical Committee of ZNA in Antwerp, Belgium, and is available on